Force MHA QKV onto fp32 #5391
Merged
Force MHA QKV onto fp32 #5391
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: smajumdar <[email protected]>
github-actions bot
pushed a commit
that referenced
this pull request
Nov 10, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]>
titu1994
added a commit
that referenced
this pull request
Nov 10, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]>
1-800-BAD-CODE
pushed a commit
to 1-800-BAD-CODE/NeMo
that referenced
this pull request
Nov 13, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Signed-off-by: 1-800-bad-code <[email protected]>
Davood-M
added a commit
that referenced
this pull request
Nov 15, 2022
* remove stage wrapper from parallel Signed-off-by: Oleksii Kuchaiev <[email protected]> * fixed the onnx bug in conformer for non-streaming models. (#5242) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> * [Tools][ASR] Tool for generating data using simulated RIRs (#5158) [Tools][ASR] Tool for generating data using simulated RIRs Signed-off-by: Ante Jukić <[email protected]> * Add fully torch.jit.script-able speaker clustering module (#5191) * Add files for commit Signed-off-by: Taejin Park <[email protected]> * Added parallelism on p-value search Signed-off-by: Taejin Park <[email protected]> * Changed speaker clustering to accept torch.tensor Signed-off-by: Taejin Park <[email protected]> * Cleaned up the code and tested to have identical output Signed-off-by: Taejin Park <[email protected]> * update on Notebook demo Signed-off-by: Taejin Park <[email protected]> * Added eigvalsh for faster eig val calculation: Signed-off-by: Taejin Park <[email protected]> * Remove NMESC_JitScriptedModule.ipynb Signed-off-by: Taejin Park <[email protected]> * Cleaned code and style fix Signed-off-by: Taejin Park <[email protected]> * Modified MSDD framework to fit torch-scripted clustering Signed-off-by: Taejin Park <[email protected]> * LGTM fix Signed-off-by: Taejin Park <[email protected]> * removed all string based timestamps Signed-off-by: Taejin Park <[email protected]> * Removed unnecessary lines Signed-off-by: Taejin Park <[email protected]> * removed redundant lines Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Update perturb.py (#5231) * Update perturb.py Add checking for channels mismatch for audio and noise data, throw an exception if they have different number of channels. Also fixed `perturb_with_foreground_noise` as done in `perturb_with_input_noise` Signed-off-by: He Huang (Steve) <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * update check and teest Signed-off-by: stevehuang52 <[email protected]> * fix test Signed-off-by: stevehuang52 <[email protected]> Signed-off-by: He Huang (Steve) <[email protected]> Signed-off-by: stevehuang52 <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * remove CV requirements. (#5233) Signed-off-by: Xuesong Yang <[email protected]> * Fix link to inference notebook (#5247) (#5251) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Co-authored-by: Jocelyn <[email protected]> * checks for accepted adapter type at module level (#5194) * add accepted adapter functionality into transformer, mlp and attention Signed-off-by: arendu <[email protected]> * fix to t5 adapter and ia3 evals due to predict_step dictionary key changes Signed-off-by: arendu <[email protected]> * use mixin logic for adapters in ParallelAttention and ParallelMLP classes Signed-off-by: arendu <[email protected]> * typo fix Signed-off-by: arendu <[email protected]> * updates Signed-off-by: arendu <[email protected]> * moved adapter tools Signed-off-by: arendu <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix error with t5 adapter Signed-off-by: arendu <[email protected]> * updates' Signed-off-by: arendu <[email protected]> * replace ColumnParallelLinear with nn.Linear in export_utils Signed-off-by: arendu <[email protected]> * remove ColumnLinear Signed-off-by: arendu <[email protected]> * typo fix Signed-off-by: arendu <[email protected]> * update to check config targets Signed-off-by: arendu <[email protected]> * updates Signed-off-by: arendu <[email protected]> * refactor so that mixin is adapter name agnostic Signed-off-by: arendu <[email protected]> * fix merge conflict Signed-off-by: arendu <[email protected]> * minor Signed-off-by: arendu <[email protected]> * minor Signed-off-by: arendu <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * using class comparison instead of string match Signed-off-by: arendu <[email protected]> * fix test fail Signed-off-by: arendu <[email protected]> * fixed checks for add_adapter Signed-off-by: arendu <[email protected]> * fixed checks for add_adapter Signed-off-by: arendu <[email protected]> Signed-off-by: arendu <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Oleksii Kuchaiev <[email protected]> * fix groovy syntax Signed-off-by: Oleksii Kuchaiev <[email protected]> * fix hypotheses return (#5253) Signed-off-by: nithinraok <[email protected]> Signed-off-by: nithinraok <[email protected]> * Update ASR scores table (#5254) (#5255) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> * Support for inserting additional subsampling in conformer encoder (#5224) * Change the default position of the reduction position to null and rename subsampling reduction to striding Signed-off-by: Shantanu Acharya <[email protected]> * Put the caching logic outside the conformer encoder Signed-off-by: Shantanu Acharya <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add description of the reduction parameters in the configs Signed-off-by: Shantanu Acharya <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update test_asr_exportables with correct reduction position value Signed-off-by: Shantanu Acharya <[email protected]> Signed-off-by: Shantanu Acharya <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * asr and nmt tests in parallel Signed-off-by: Oleksii Kuchaiev <[email protected]> * add more users who can trigger blossom-ci Signed-off-by: Oleksii Kuchaiev <[email protected]> * path fix Signed-off-by: Oleksii Kuchaiev <[email protected]> * fix paths, remove redundant test Signed-off-by: Oleksii Kuchaiev <[email protected]> * add symlink Signed-off-by: Oleksii Kuchaiev <[email protected]> * Modernize RNNT ONNX export and add TS export (#5248) * Upgrade rnnt export for CUDA/CPU/TRT Signed-off-by: smajumdar <[email protected]> * Update runtime script for onnx exported model to modern API Signed-off-by: smajumdar <[email protected]> * Finalize code Signed-off-by: smajumdar <[email protected]> * Remove comments Signed-off-by: smajumdar <[email protected]> * Remove redundant stuff from tests Signed-off-by: smajumdar <[email protected]> * Update test Signed-off-by: smajumdar <[email protected]> * Remove onnx rnnt export test due to lack of onnxruntime install Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * update tutorials to use meeting config as default and VAD (#5237) * update tutorials to use meeting config as default and VAD Signed-off-by: nithinraok <[email protected]> * update model path Signed-off-by: nithinraok <[email protected]> Signed-off-by: nithinraok <[email protected]> * Fix links to speaker identification notebook (#5260) (#5261) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> Co-authored-by: Sean Naren <[email protected]> * add shm-size Signed-off-by: Oleksii Kuchaiev <[email protected]> * [TTS] Fastpitch energy condition and refactoring (#5218) * Incorporating Energy conditioning in FastPitch Signed-off-by: subhankar-ghosh <[email protected]> * Minor fixes in Energy conditioning in FastPitch Signed-off-by: subhankar-ghosh <[email protected]> * Add Energy conditioning in FastPitch to infer method Signed-off-by: subhankar-ghosh <[email protected]> * adding fn to function names Signed-off-by: subhankar-ghosh <[email protected]> * Incorporating Energy conditioning in FastPitch Signed-off-by: subhankar-ghosh <[email protected]> * Minor fixes in Energy conditioning in FastPitch Signed-off-by: subhankar-ghosh <[email protected]> * Add Energy conditioning in FastPitch to infer method Signed-off-by: subhankar-ghosh <[email protected]> * adding fn to function names Signed-off-by: subhankar-ghosh <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * remove ifelse from batching, minor refactoring changes in energy code Signed-off-by: subhankar-ghosh <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Refactor based on PR comments. Signed-off-by: subhankar-ghosh <[email protected]> * Added support for not learning alignment in energy Signed-off-by: subhankar-ghosh <[email protected]> * Fix typo in assert statemetn Signed-off-by: subhankar-ghosh <[email protected]> * Renaming average_pitch to average_features Signed-off-by: subhankar-ghosh <[email protected]> * Renaming len variable name as it is a keyword Signed-off-by: subhankar-ghosh <[email protected]> * Renaming len variable name as it is a keyword Signed-off-by: subhankar-ghosh <[email protected]> Signed-off-by: subhankar-ghosh <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Xuesong Yang <[email protected]> * [TTS] HiFi-TTS Download Script (#5241) * Hifi tts download script Signed-off-by: Oleksii Volkovskyi <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oleksii Volkovskyi <[email protected]> * comment and remove imports Signed-off-by: Oleksii Volkovskyi <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oleksii Volkovskyi <[email protected]> Signed-off-by: Oleksii Volkovskyi <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Specifying audio signal dropout separately for the Conformer Encoder (#5263) * Fixed bug in transcribe_speech.py where decoding strategy was not being updated. Signed-off-by: Shantanu Acharya <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add option to specify audio dropout separately for conformer encoders Signed-off-by: Shantanu Acharya <[email protected]> * Add audio dropout option to test_asr_exportables Signed-off-by: Shantanu Acharya <[email protected]> * Rename dropout_audio to dropout_pre_encode Signed-off-by: Shantanu Acharya <[email protected]> * Update the comments in squeezeformer configs referring to conformer modules Signed-off-by: Shantanu Acharya <[email protected]> Signed-off-by: Shantanu Acharya <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * created (#5268) * created * bug Signed-off-by: Dima Rekesh <[email protected]> Signed-off-by: Dima Rekesh <[email protected]> Co-authored-by: Dima Rekesh <[email protected]> * [TTS] Add Mandarin/English Bilingual Recipe for Training Fastpitch Models (#5208) * Add Chinese TTS tokenizer and G2P. * Add data process script. * Add tutorial. Signed-off-by: Yuekai Zhang <[email protected]> * Minor typo fixes in TTS tutorial (#5266) (#5272) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Co-authored-by: Jocelyn <[email protected]> * Fix failing speaker counting for short audio samples (#5267) * Add files for commit Signed-off-by: Taejin Park <[email protected]> * Added parallelism on p-value search Signed-off-by: Taejin Park <[email protected]> * Changed speaker clustering to accept torch.tensor Signed-off-by: Taejin Park <[email protected]> * Cleaned up the code and tested to have identical output Signed-off-by: Taejin Park <[email protected]> * update on Notebook demo Signed-off-by: Taejin Park <[email protected]> * Added eigvalsh for faster eig val calculation: Signed-off-by: Taejin Park <[email protected]> * Remove NMESC_JitScriptedModule.ipynb Signed-off-by: Taejin Park <[email protected]> * Cleaned code and style fix Signed-off-by: Taejin Park <[email protected]> * Modified MSDD framework to fit torch-scripted clustering Signed-off-by: Taejin Park <[email protected]> * LGTM fix Signed-off-by: Taejin Park <[email protected]> * removed all string based timestamps Signed-off-by: Taejin Park <[email protected]> * Removed unnecessary lines Signed-off-by: Taejin Park <[email protected]> * removed redundant lines Signed-off-by: Taejin Park <[email protected]> * Add enhanced speaker count back Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed minor docstrings Signed-off-by: Taejin Park <[email protected]> * removed import Counter Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (#5271) (#5273) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> * Add Gradio App to ASR Docs (#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix bug into Dialogue tutorial (#5277) (#5280) Co-authored-by: Zhilin Wang <[email protected]> * [TTS] fixed type of filepath and rename openslr. (#5276) Signed-off-by: Xuesong Yang <[email protected]> * O2bert + apex pipeline functions (#5221) * Global batch size support for validation * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Global batch size support for bert * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bert batch support * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bert batch size support * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * O2 support for bert * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_pretraining.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update megatron_bert_model.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_config.yaml Signed-off-by: Shanmugam Ramasamy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_model.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * Bug fix * Bug fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Bug fix * Bug fix * Bug fix * Update megatron_bert_config.yaml Signed-off-by: Shanmugam Ramasamy <[email protected]> * Addressed Sandeeps comments Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update Jenkinsfile Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update Jenkinsfile Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update megatron_bert_model.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update Jenkinsfile Signed-off-by: Shanmugam Ramasamy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Shanmugam Ramasamy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Typo fix (#5288) (#5291) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> * Upperbound PTL (#5302) * Upperbound PTL Signed-off-by: smajumdar <[email protected]> * Upperbound PTL Signed-off-by: smajumdar <[email protected]> * Upperbound PTL Signed-off-by: smajumdar <[email protected]> * Upperbound PTL Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Add support for Sampled Softmax for RNNT Joint (#5216) * Initial prototype of SampldRNNTJoint Signed-off-by: smajumdar <[email protected]> * Implement randperm based noise selection algo Signed-off-by: smajumdar <[email protected]> * First working prototype of sampled rnnt ! Signed-off-by: smajumdar <[email protected]> * Add note for why we need this remap before i forget Signed-off-by: smajumdar <[email protected]> * Finalize version that works with sampling Signed-off-by: smajumdar <[email protected]> * Update docs for rnnt decoder and joint Signed-off-by: smajumdar <[email protected]> * Remove the adjustment_val for softmax Signed-off-by: smajumdar <[email protected]> * Update config and docs for Sampled Softmax Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Update Interface(s) phonetic entry (#5212) * change interface(s) phone Signed-off-by: Jason <[email protected]> * push version Signed-off-by: Jason <[email protected]> * update dict path Signed-off-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * [TTS] remove obsolete torch_tts unit test marker and replace with run_only_on('CPU') (#5307) Signed-off-by: Xuesong Yang <[email protected]> * Fixes for Conformer-xl export (#5309) * Fixing runtime check for ONNX > 2G Signed-off-by: Boris Fomitchev <[email protected]> * Fixing ONNX export Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * add label inference support to EncDecSpeakerLabel class (#5278) * add label inference support to EncDecSpeakerLabel class Signed-off-by: nithinraok <[email protected]> * add necessary tests Signed-off-by: nithinraok <[email protected]> * reflect on comments Signed-off-by: nithinraok <[email protected]> * grammatical correction Signed-off-by: nithinraok <[email protected]> * minor doc string changes Signed-off-by: nithinraok <[email protected]> Signed-off-by: nithinraok <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * [TTS] bugfix IPAG2P and refactor to remove duplicate process. (#5304) * [TTS] bugfix IPAG2P and refactor to remove duplicate process. * added type hints and rename func. * unify str and list(str) as list(str). * revise logging message when phoneme_dict_obj is empty Signed-off-by: Xuesong Yang <[email protected]> * Update path to get_data.py in TTS tutorial (#5311) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Add italian model checkpoints (#5315) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * Text Memmap Parsing Improvements (#5265) * 1. Fixed text-memmap issue when boundary (new-line) is missing from end of file). Signed-off-by: Micha Livne <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * 1. Fixed style. Signed-off-by: Micha Livne <[email protected]> * 1. Added support in paratial sample loading and alternative decoding. Signed-off-by: Micha Livne <[email protected]> * 1. Fixed syntax issues. Signed-off-by: Micha Livne <[email protected]> * 1. Minor change. Signed-off-by: Micha Livne <[email protected]> * 1. Extended flexibility of mapping indices. Signed-off-by: Micha Livne <[email protected]> * 1. Added validation ofdtype of indexing function. Signed-off-by: Micha Livne <[email protected]> Signed-off-by: Micha Livne <[email protected]> Co-authored-by: Micha Livne <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * [TTS] Replace IPA lambda arguments with locale string (#5298) * [TTS] Replace IPA lambda arguments with locale string * [TTS] Add locale validation * Fixed typos * Return punctuation as sorted list Signed-off-by: Ryan <[email protected]> * Remove onnx graphsurgery from Dockerfile (#5320) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Update refspec (#5321) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Force wav file format for audio_filepath (#5323) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Updates to T0 Dataset and Model (#5201) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fixes Signed-off-by: MaximumEntropy <[email protected]> * Update config Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Restore function needed for NMT Signed-off-by: MaximumEntropy <[email protected]> * Fix config Signed-off-by: MaximumEntropy <[email protected]> * Change output file format from JSON to JSONL Signed-off-by: MaximumEntropy <[email protected]> * Add T0 data preproc scripts Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Merge and multiprocessing Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix for is_correct Signed-off-by: MaximumEntropy <[email protected]> * Refactor T0 dataset Signed-off-by: MaximumEntropy <[email protected]> * Add script to merge train folder into individual training files to minimize number of blends Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Minor changes Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Remove bin compat Signed-off-by: MaximumEntropy <[email protected]> * Fix header lines Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * [DOC] add sphinx-copybutton requirement to copy button on code snippets. (#5326) Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * [TTS] expand to support flexible dictionary entry formats in IPAG2P. (#5318) * expand to support flexible dictionary entry formats in IPAG2P. * removed unused imports in test.collections.tts * removed unused imports in nemo.collections.tts.modules * removed unused imports in nemo_text_processing.text_normalization.zh * updated unit tests with new cases * renamed test function names because we only test IPAG2P rather than all classes in the modules.py. * revise current test dict with a single space between word and pronunications. Signed-off-by: Xuesong Yang <[email protected]> * small bugfix for r1.13.0 (#5310) (#5325) * typo fix * udpate transcribe Signed-off-by: fayejf <[email protected]> Co-authored-by: fayejf <[email protected]> * Option to pad the last validation input sequence if its smaller than the encoder sequence length for MegatronGPT (#5243) * Option to pad the last input sequence of validation dataset if its smaller than the encoder sequence length for MegatronGPT * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added default value for drop last argument Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Anmol Gupta <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Add support for Hydra multirun to NeMo (#5159) * Update execution doc and remove old snippet Signed-off-by: smajumdar <[email protected]> * Fix types Signed-off-by: smajumdar <[email protected]> * Fix defaults Signed-off-by: smajumdar <[email protected]> * Fix types for ParallelAdapterConfig Signed-off-by: smajumdar <[email protected]> * Add hash for config cache Signed-off-by: smajumdar <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add support to delete redundant ckpt files for HP search Signed-off-by: smajumdar <[email protected]> * Correct config for IA3 Signed-off-by: smajumdar <[email protected]> * Fix check to <= 0 Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * typo fix (#5328) * Speed up HF data processing script for ASR (#5330) * Correct hydra issue with relative filepaths Signed-off-by: smajumdar <[email protected]> * Improve speed of dataset processing Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * add precommit hood to automatic sort entries in requirements. (#5333) Signed-off-by: Xuesong Yang <[email protected]> * [TTS] update organization of model checkpoints and their pointers. (#5327) * [TTS] update orgnization of model checkpoints and their pointers. Signed-off-by: Xuesong Yang <[email protected]> * move model name column to the 2nd col and correct model names as predefined_model_name. Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Add speaker clustering arguments to forward function (#5306) * Move arguments to forward function Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Resolved type issue Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Taejin Park <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (#5340) (#5341) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * [TTS] bugfix for the script of generating mels. (#5344) Signed-off-by: Xuesong Yang <[email protected]> * Fixing de-autocast (#5319) * Fixing de-autocast Signed-off-by: Boris Fomitchev <[email protected]> * Cleanup Signed-off-by: Boris Fomitchev <[email protected]> * Refining export with max_dim/batch Signed-off-by: Boris Fomitchev <[email protected]> * Moving cast utils to its own module Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * Pcla tutorial fixes (#5313) (#5347) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> * bug (#5348) Co-authored-by: Dima Rekesh <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> * [Bugfix] Added rm -f / wget- nc command to avoid bash error in multispeaker sim notebook (#5292) * Added rm -f command to avoid error message Signed-off-by: Taejin Park <[email protected]> * removed unnecessary changes Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * [DOC] added ipython dependency to support IPython.sphinxext extension (#5345) * [DOC] added ipython dependency to support IPython.sphinxext extension Signed-off-by: Xuesong Yang <[email protected]> * revert ipython extension in the doc and replace ipython block with shell-session. Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Fix dialogue tutorial bug (#5297) (#5303) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Eric Harper <[email protected]> * Fix issue with HF Model upload tutorial (#5359) (#5360) * Add Gradio App to ASR Docs (#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a387e3b3d9cdf511f7b9bbb5e94925e48cc2) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> * Bug fix (removing old compute consumed samples) (#5355) Signed-off-by: Shanmugam Ramasamy <[email protected]> Signed-off-by: Shanmugam Ramasamy <[email protected]> * removed uninstall nemo_cv and nemo_simple_gan and relax numba version… (#5332) * Update reinstall.sh and requirements. * removed nemo_cv and nemo_simple_gan in reinstall.sh. * relaxed numba version limits. * added tensorboard requirement to avoid any incpmpatible issue. Signed-off-by: Xuesong Yang <[email protected]> * revert changes for numba Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Pipeline paralleism in Bert (#5293) * Global batch size support for validation * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Global batch size support for bert * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bert batch support * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bert batch size support * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * O2 support for bert * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_pretraining.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * Update megatron_bert_model.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_config.yaml Signed-off-by: Shanmugam Ramasamy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_model.py Signed-off-by: Shanmugam Ramasamy <[email protected]> * Bug fix * Bug fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Bug fix * Bug fix * Bug fix * Update megatron_bert_config.yaml Signed-off-by: Shanmugam Ramasamy <[email protected]> * PPBert * PPBert * PPBert * PPBert * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update megatron_bert_config.yaml Signed-off-by: Shanmugam Ramasamy <[email protected]> * bug fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bug fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bug fix * bug fix * bug fix * bug fix Signed-off-by: Shanmugam Ramasamy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Oleksii Kuchaiev <[email protected]> * tutorial fixes (#5354) (#5361) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> * Enable mlflow logger (#4893) * Enable mlflow logger Signed-off-by: whrichd <[email protected]> * fix style Signed-off-by: whrichd <[email protected]> * Add doc lines. Signed-off-by: whrichd <[email protected]> * change default value Signed-off-by: whrichd <[email protected]> * fix doc Signed-off-by: whrichd <[email protected]> * addressed comments, added dataclass Signed-off-by: whrichd <[email protected]> * fix style Signed-off-by: whrichd <[email protected]> * fix doc Signed-off-by: whrichd <[email protected]> Signed-off-by: whrichd <[email protected]> * Add SDP documentation (#5274) (#5376) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (#5378) (#5381) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (#5379) (#5384) Co-authored-by: Adi Renduchintala <[email protected]> * [TTS] Add Spanish model documentation (#5390) Signed-off-by: Ryan <[email protected]> * [TTS] Add Spanish FastPitch training configs (#5383) * [TTS] Add Spanish FastPitch training configs * [TTS] Add single speaker Spanish configs Signed-off-by: Ryan <[email protected]> * Fix Python type hints according to Python Docs (#5370) * Remove duplicated type annotations Signed-off-by: Vladimir Bataev <[email protected]> * Fix tuple annotations in function return types Signed-off-by: Vladimir Bataev <[email protected]> * Add necessary imports Signed-off-by: Vladimir Bataev <[email protected]> * Add necessary imports Signed-off-by: Vladimir Bataev <[email protected]> * Fix types in obvious places Signed-off-by: Vladimir Bataev <[email protected]> * Fix types in obvious places Signed-off-by: Vladimir Bataev <[email protected]> * Fix unused import (avoid quotes in type annotations) Signed-off-by: Vladimir Bataev <[email protected]> * Revert "Fix unused import (avoid quotes in type annotations)" This reverts commit ea433efcd9916abf8944879e791484a0a1437f83. Signed-off-by: Vladimir Bataev <[email protected]> * Remove problematic import Signed-off-by: Vladimir Bataev <[email protected]> * Fix list_available_models method type Signed-off-by: Vladimir Bataev <[email protected]> * Revert some changes Signed-off-by: Vladimir Bataev <[email protected]> * Revert quotes in list_available_models Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> * Force MHA QKV onto fp32 (#5391) (#5395) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> * Add cpWER for evaluation of ASR with diarization (#5279) * Add cpWER calculation feature Signed-off-by: Taejin Park <[email protected]> * added notebook Signed-off-by: Taejin Park <[email protected]> * updated notebook and diarization_utils Signed-off-by: Taejin Park <[email protected]> * Minor update on tutorial notebook Signed-off-by: Taejin Park <[email protected]> * Style fix Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update on missing docstrings Signed-off-by: Taejin Park <[email protected]> * Fixed an unfinished docstring Signed-off-by: Taejin Park <[email protected]> * Removed unused variables Signed-off-by: Taejin Park <[email protected]> * Fixed dict input to list input Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Style fix Signed-off-by: Taejin Park <[email protected]> * fixed LGTM issues Signed-off-by: Taejin Park <[email protected]> * Fixed error in cpWER cal Signed-off-by: Taejin Park <[email protected]> * fixed docstrings Signed-off-by: Taejin Park <[email protected]> * fixed docstrings Signed-off-by: Taejin Park <[email protected]> * Fix some of the typing issues, lower case names Signed-off-by: SeanNaren <[email protected]> * Replaced bruteforce with LSA alg for cpWER Signed-off-by: Taejin Park <[email protected]> * Reflected PR comments Signed-off-by: Taejin Park <[email protected]> * Cleaned notebook Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated notebook Signed-off-by: Taejin Park <[email protected]> * Fixed LGTM warnings Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added test_diar_metrics.py Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed typos Signed-off-by: Taejin Park <[email protected]> * Fixed wrong type annotations Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added bruteforce mode and its unit-test Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * LGTM issues fixed Signed-off-by: Taejin Park <[email protected]> * reolve LGTM issues Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * unified speaker key in trans_dict Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Removed unused variable and imports Signed-off-by: Taejin Park <[email protected]> * Update nemo/collections/asr/parts/utils/diarization_utils.py Co-authored-by: Sean Naren <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Update nemo/collections/asr/parts/utils/diarization_utils.py Co-authored-by: Sean Naren <[email protected]> Signed-off-by: Taejin Park <[email protected]> * moved all the diarization eval to der.py Signed-off-by: Taejin Park <[email protected]> * Update tests/collections/asr/test_diar_metrics.py Co-authored-by: Sean Naren <[email protected]> Signed-off-by: Taejin Park <[email protected]> * der.py update on tests Signed-off-by: Taejin Park <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * unused imports and style fix Signed-off-by: Taejin Park <[email protected]> * style fix Signed-off-by: Taejin Park <[email protected]> * unused import Signed-off-by: Taejin Park <[email protected]> * reflected review comments Signed-off-by: Taejin Park <[email protected]> * Fixed an import bug in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: SeanNaren <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: SeanNaren <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * Added cast Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Ante Jukić <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: He Huang (Steve) <[email protected]> Signed-off-by: stevehuang52 <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: arendu <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: Shantanu Acharya <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: subhankar-ghosh <[email protected]> Signed-off-by: Oleksii Volkovskyi <[email protected]> Signed-off-by: Dima Rekesh <[email protected]> Signed-off-by: Yuekai Zhang <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Shanmugam Ramasamy <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Micha Livne <[email protected]> Signed-off-by: Ryan <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: whrichd <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: David <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: anteju <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: He Huang (Steve) <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Shantanu Acharya <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Subhankar Ghosh <[email protected]> Co-authored-by: Oleksii Volkovskyi <[email protected]> Co-authored-by: Dima Rekesh <[email protected]> Co-authored-by: Dima Rekesh <[email protected]> Co-authored-by: Yuekai Zhang <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Shanmugam Ramasamy <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Micha Livne <[email protected]> Co-authored-by: Micha Livne <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Ryan Langman <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: anmolgupt <[email protected]> Co-authored-by: Anmol Gupta <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: Eric Harper <[email protected]> Co-authored-by: Riqiang Wang <[email protected]> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: David <[email protected]>
hainan-xv
pushed a commit
to hainan-xv/NeMo
that referenced
this pull request
Nov 29, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Signed-off-by: Hainan Xu <[email protected]>
hainan-xv
pushed a commit
to hainan-xv/NeMo
that referenced
this pull request
Nov 29, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Signed-off-by: Hainan Xu <[email protected]>
JimmyZhang12
pushed a commit
to JimmyZhang12/NeMo
that referenced
this pull request
Dec 14, 2022
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]>
andrusenkoau
pushed a commit
to andrusenkoau/NeMo
that referenced
this pull request
Jan 5, 2023
Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Signed-off-by: andrusenkoau <[email protected]>
treacker
added a commit
that referenced
this pull request
Jan 25, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (#5277) * Typo fix (#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (#5359) * Add Gradio App to ASR Docs (#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (#5379) * Fixed bug in notebook (#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (#5421)" (#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (#5242) (#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (#5339) (#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
Kipok
added a commit
to Kipok/NeMo
that referenced
this pull request
Jan 31, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (NVIDIA#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (NVIDIA#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (NVIDIA#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (NVIDIA#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (NVIDIA#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (NVIDIA#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (NVIDIA#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (NVIDIA#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (NVIDIA#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (NVIDIA#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (NVIDIA#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (NVIDIA#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (NVIDIA#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (NVIDIA#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (NVIDIA#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (NVIDIA#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (NVIDIA#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (NVIDIA#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (NVIDIA#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (NVIDIA#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (NVIDIA#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (NVIDIA#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (NVIDIA#5277) * Typo fix (NVIDIA#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (NVIDIA#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (NVIDIA#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (NVIDIA#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (NVIDIA#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (NVIDIA#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (NVIDIA#5359) * Add Gradio App to ASR Docs (NVIDIA#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (NVIDIA#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (NVIDIA#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (NVIDIA#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (NVIDIA#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (NVIDIA#5379) * Fixed bug in notebook (NVIDIA#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (NVIDIA#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (NVIDIA#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (NVIDIA#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (NVIDIA#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (NVIDIA#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (NVIDIA#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (NVIDIA#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421)" (NVIDIA#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (NVIDIA#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (NVIDIA#5242) (NVIDIA#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (NVIDIA#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (NVIDIA#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (NVIDIA#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (NVIDIA#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (NVIDIA#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (NVIDIA#5339) (NVIDIA#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (NVIDIA#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (NVIDIA#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (NVIDIA#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (NVIDIA#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (NVIDIA#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (NVIDIA#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (NVIDIA#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
ericharper
added a commit
that referenced
this pull request
Jan 31, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (#5277) * Typo fix (#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (#5359) * Add Gradio App to ASR Docs (#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (#5379) * Fixed bug in notebook (#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (#5421)" (#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (#5242) (#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (#5339) (#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
ericharper
added a commit
that referenced
this pull request
Jan 31, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (#5277) * Typo fix (#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (#5359) * Add Gradio App to ASR Docs (#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (#5379) * Fixed bug in notebook (#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (#5421)" (#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (#5242) (#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (#5339) (#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
Kipok
added a commit
to Kipok/NeMo
that referenced
this pull request
Jan 31, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (NVIDIA#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (NVIDIA#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (NVIDIA#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (NVIDIA#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (NVIDIA#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (NVIDIA#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (NVIDIA#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (NVIDIA#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (NVIDIA#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (NVIDIA#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (NVIDIA#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (NVIDIA#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (NVIDIA#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (NVIDIA#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (NVIDIA#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (NVIDIA#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (NVIDIA#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (NVIDIA#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (NVIDIA#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (NVIDIA#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (NVIDIA#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (NVIDIA#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (NVIDIA#5277) * Typo fix (NVIDIA#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (NVIDIA#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (NVIDIA#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (NVIDIA#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (NVIDIA#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (NVIDIA#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (NVIDIA#5359) * Add Gradio App to ASR Docs (NVIDIA#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (NVIDIA#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (NVIDIA#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (NVIDIA#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (NVIDIA#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (NVIDIA#5379) * Fixed bug in notebook (NVIDIA#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (NVIDIA#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (NVIDIA#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (NVIDIA#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (NVIDIA#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (NVIDIA#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (NVIDIA#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (NVIDIA#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421)" (NVIDIA#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (NVIDIA#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (NVIDIA#5242) (NVIDIA#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (NVIDIA#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (NVIDIA#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (NVIDIA#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (NVIDIA#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (NVIDIA#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (NVIDIA#5339) (NVIDIA#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (NVIDIA#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (NVIDIA#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (NVIDIA#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (NVIDIA#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (NVIDIA#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (NVIDIA#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (NVIDIA#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
titu1994
added a commit
to titu1994/NeMo
that referenced
this pull request
Mar 24, 2023
* Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount * new structure for tts datasets in script folder Signed-off-by: Oktai Tatanov <[email protected]> * remove cmudict downloading Signed-off-by: Oktai Tatanov <[email protected]> * rename mixertts dataset, add vocoder dataset Signed-off-by: Oktai Tatanov <[email protected]> * add libritts processing Signed-off-by: Oktai Tatanov <[email protected]> * update tts dataset and libritts get data Signed-off-by: Oktai Tatanov <[email protected]> * fix bugs in vocoder ds Signed-off-by: Oktai Tatanov <[email protected]> * add ds * changed vits yaml * rm yaml * fix yaml and model * Added scaler * refactored yaml * managed to run in fp16 * refactoring Signed-off-by: Oktai Tatanov <[email protected]> * fix small bugs and add new todos Signed-off-by: Oktai Tatanov <[email protected]> * fix optimizers Signed-off-by: Oktai Tatanov <[email protected]> * Port Variational Inference with Adversarial Learning (VITS) to NeMo TTS (NVIDIA#6) * Add vits files Add vits_losses.py, vits_modules.py and vits.py. * Move non-vits models to modules * Add vits.yaml * Add _loader to vits.py * Add basic template for vits * Update vits.yaml with vits parameters * Remove extra space * Add top level training script * Add some variables to vits yaml * Add forward and training methods * Fix imports * Added validation step * Log training losses * Update loss calls to use class attributes * Add VITS to models list * Fix all imports * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Fix imports for VITS * Remove old module calls * Fix typo in monotonic align import * Modified validation step 1. reverted to tensorboard 2. validation_step logs audio, mel-spec for batch 0 3. validation_step_alt logs audio, mel-spec for batch 0 and loss_mel * Add parameters from original VITS config * Fix config file * Fix imports and generate spec from audio * Fix incorrect dimensions * Progress update * Fix loss * Fix cuda thing * Fix monotonic align import * Fix typos in vits.py * Disable loss typecheck * Fix spectrogram lengths * Remove Precision 16 requirement * Address lgtm alerts * clean up unused code * Address lgtm alerts * Refactor audio_to_mel_torch method * Use NeMo FilterBank to get melspec Todo: set self.fb * Fix filterbank max frequency to match with original VITS * Fix filterbank features correct length * Address lgtm issues * Remove print statements * Remove stft_pad_amount Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Jason <[email protected]> Signed-off-by: Jason <[email protected]> * make new commit Signed-off-by: Jason <[email protected]> * add copyright headers Signed-off-by: Jason <[email protected]> * style Signed-off-by: Jason <[email protected]> * rename README Signed-off-by: Oktai Tatanov <[email protected]> * fix style without vits_modules Signed-off-by: Oktai Tatanov <[email protected]> * add numba code, fix style and add todos Signed-off-by: Oktai Tatanov <[email protected]> * small fix * fix some todos * added numba mas * added DDP sampler * specified versions * fixed for new librosa version * added feature loss * added IPA phonemizer * refactored IPA g2p * added vits losses * some ref * fix * added checkpointing * cp * cfg * merged some 1.8.0 fixes * plt fix * fix logging * fix checkpoint loading * refactored inference * fp32 run * update branch Signed-off-by: ericharper <[email protected]> * update package info Signed-off-by: ericharper <[email protected]> * new exp * update branch Signed-off-by: ericharper <[email protected]> * Restored tests previously disabled for 22.03 base (NVIDIA#4109) Signed-off-by: Boris Fomitchev <[email protected]> * add augmentation to label models (NVIDIA#4113) * add augmentation to label models Signed-off-by: nithinraok <[email protected]> * duration fix Signed-off-by: nithinraok <[email protected]> * Call register_bert_model after assigning self.bert_model variable (NVIDIA#4116) Signed-off-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> * Tutorial on ITN with Thutmose tagger and small fixes (NVIDIA#4117) * 1. Add tutorial. 2. Move a function to fix import in tutorial. 3. Merge multiple spaces into one space in the final output Signed-off-by: Alexandra Antonova <[email protected]> * fixes for code review Signed-off-by: Alexandra Antonova <[email protected]> * Add tutorial to tutorials.rst Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> * update the default (NVIDIA#4135) Signed-off-by: ekmb <[email protected]> * Draft: Fix restoring from checkpoint for case when `model.common_dataset_parameters.label_vocab_dir` is provided (NVIDIA#4136) * Fix restoring from checkpoint with label vocab dir Signed-off-by: PeganovAnton <[email protected]> * Add tests for various ways to pass label ids to model Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Fix typo Signed-off-by: PeganovAnton <[email protected]> * Do not create tmp directory Signed-off-by: PeganovAnton <[email protected]> * Fix parameter name Signed-off-by: PeganovAnton <[email protected]> * finish cherry-pick op Signed-off-by: PeganovAnton <[email protected]> * Fix labels errors Signed-off-by: PeganovAnton <[email protected]> * Remove duplicate stage Signed-off-by: PeganovAnton <[email protected]> * Change target branch Signed-off-by: PeganovAnton <[email protected]> * fix typo (NVIDIA#4140) Signed-off-by: Yang Zhang <[email protected]> * Fix/punctuation avoid overwritting tmp files (NVIDIA#4144) * Add draft of fixing tmp files overwritting Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Remove accidental changes Signed-off-by: PeganovAnton <[email protected]> * Use built-in tempfile library Signed-off-by: PeganovAnton <[email protected]> * Fix code style Signed-off-by: PeganovAnton <[email protected]> * bug_fix_diarization_manifest_creation (NVIDIA#4125) Signed-off-by: Yang Zhang <[email protected]> Co-authored-by: Nithin Rao <[email protected]> * fix doc (NVIDIA#4146) Signed-off-by: Yang Zhang <[email protected]> * Tacotron2 retrain (NVIDIA#4103) * fix yaml Signed-off-by: treacker <[email protected]> * Fix for new TTSDataset class Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * added wandb logging Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * fix numpy version Signed-off-by: treacker <[email protected]> * inference fix Signed-off-by: treacker <[email protected]> * removed old code Signed-off-by: treacker <[email protected]> * updated parser logic Signed-off-by: treacker <[email protected]> * reverted version update Signed-off-by: treacker <[email protected]> * refactored parser logic Signed-off-by: treacker <[email protected]> * Updated Jenkinsfile Signed-off-by: treacker <[email protected]> * Refactored tutorial for Tacotron2 Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Made backward compatibility Signed-off-by: treacker <[email protected]> * Update Jenkinsfile Signed-off-by: treacker <[email protected]> * Update tacotron.yaml Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * cleaned up TN/ ITN doc (NVIDIA#4119) * cleaned up TN/ ITN doc Signed-off-by: Yang Zhang <[email protected]> * fix typo Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> * fix image Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: treacker <[email protected]> * Check implicit grad acc in GLUE dataset building (NVIDIA#4123) * Check implicit grad acc in GLUE dataset building Signed-off-by: MaximumEntropy <[email protected]> * Fix jenkins test for GLUE/XNLI Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Fixed jenkins Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> * Refactoring Signed-off-by: treacker <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> * Multiprocess improvements (NVIDIA#4127) * initial commit Signed-off-by: nithinraok <[email protected]> * start fix Signed-off-by: nithinraok <[email protected]> * improve multiprocessing speed while creating speaker dataset Signed-off-by: nithinraok <[email protected]> * updated scp to filelist Signed-off-by: nithinraok <[email protected]> * WaveGlow input type fixes (NVIDIA#4151) Signed-off-by: Jocelyn Huang <[email protected]> * notebooks' link, typo and import fix (NVIDIA#4158) * redo missing pr 4007 Signed-off-by: fayejf <[email protected]> * remove extremely unreliable links Signed-off-by: fayejf <[email protected]> * Thutmose tagger bug fixes (NVIDIA#4162) * add pretrained ngc model, small fixes Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * fix model location Signed-off-by: Alexandra Antonova <[email protected]> * 1. fix typos. 2. write magic functions without space Signed-off-by: Alexandra Antonova <[email protected]> * add example of inference with pretrained model Signed-off-by: Alexandra Antonova <[email protected]> * changed model location to nemo Signed-off-by: Alexandra Antonova <[email protected]> * style fix Signed-off-by: Alexandra Antonova <[email protected]> * fix space Signed-off-by: Alexandra Antonova <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> * update speaker docs (NVIDIA#4164) * update speaker docs Signed-off-by: nithinraok <[email protected]> * chunks -> segments Signed-off-by: nithinraok <[email protected]> * Khz -> kHz Signed-off-by: nithinraok <[email protected]> * changed to vits g2p * refactoring * added cosineLR * Updated whitelist path * added vanilla torch grad scaler * Fixed lightning version * added warmup and wd * switched to cosineLR * refactored data classes for vits * some fixes * fixed import * changeg train loop * fixed scheduler bug * refactoring for exps * Refactored loss logic * Ref for exps * added coqui stuff * exps * bugfix * added side file * bugfix * reverted * fixed sampler behaviour * updated for ptl 1.7.2 * refactored dataloader func * some cleaning * reverted to vanilla loss * modified for pickling * added dataset class * fixed torch version * added autocast for fp training * removed coqui files * 'Fixed tokenizer' * Fix tokenizer * update branch Signed-off-by: ericharper <[email protected]> * Fix link to inference notebook (NVIDIA#5247) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Update ASR scores table (NVIDIA#5254) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Fix links to speaker identification notebook (NVIDIA#5260) Signed-off-by: SeanNaren <[email protected]> Signed-off-by: SeanNaren <[email protected]> * Minor typo fixes in TTS tutorial (NVIDIA#5266) Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> * Pcla tutorial fixes (NVIDIA#5271) * Fixed typos Signed-off-by: Matvei Novikov <[email protected]> * Fixed cell type and tatoeba reference Signed-off-by: Matvei Novikov <[email protected]> * Fixed typo Signed-off-by: Matvei Novikov <[email protected]> * Fixed branch variable Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix bug into Dialogue tutorial (NVIDIA#5277) * Typo fix (NVIDIA#5288) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Fix dialogue tutorial bug (NVIDIA#5297) * set add_pooling_layer=False for huggingface bert model * remove add_pooling_layer=False and set find_unused_parameters=True * set num_prompt_tokens to 0 for huggingface * small bugfix for r1.13.0 (NVIDIA#5310) * typo fix Signed-off-by: fayejf <[email protected]> * udpate transcribe Signed-off-by: fayejf <[email protected]> Signed-off-by: fayejf <[email protected]> * Add italian model checkpoints (NVIDIA#5316) Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Igor Gitman <[email protected]> * [STT] Add Ru ASR Conformer-CTC and Conformer-Transducer (NVIDIA#5340) * [STT] Add stt_ru_conformer_ctc_large Signed-off-by: Sasha Meister <[email protected]> * [STT] Add stt_ru_conformer_transducer_large Add stt_ru_conformer_transducer_large Signed-off-by: Sasha Meister <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Pcla tutorial fixes (NVIDIA#5313) * fixes Signed-off-by: Matvei Novikov <[email protected]> * fixes Signed-off-by: Matvei Novikov <[email protected]> * moved `create_text_and_labels` to token_classification_utils.py Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * a lot of refactoring * strict ptl version * strict ptl version * reverted plt version * Added base text2audio class * Fix issue with HF Model upload tutorial (NVIDIA#5359) * Add Gradio App to ASR Docs (NVIDIA#5270) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> (cherry picked from commit e4b6a38) * Fix issue with normalized config for dataset name Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * tutorial fixes (NVIDIA#5354) Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> * Add SDP documentation (NVIDIA#5274) * Add details to SDP README.md Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to WriteManifest processor Signed-off-by: Elena Rastorgueva <[email protected]> * Add docstring to CreateInitialManifestMLS Signed-off-by: Elena Rastorgueva <[email protected]> * Add ModifyManifestTextProcessor docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add ASRInference docstring Signed-off-by: Elena Rastorgueva <[email protected]> * Add base_processor docstrings Signed-off-by: Elena Rastorgueva <[email protected]> * Add minimal SDP docs page Signed-off-by: Elena Rastorgueva <[email protected]> * Update tools/speech_dataset_processor/README.md Co-authored-by: Igor Gitman <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * Write simple README for SDP and move complex explanations to docs Signed-off-by: Elena Rastorgueva <[email protected]> * Remove incorrect type hints Signed-off-by: Elena Rastorgueva <[email protected]> * Make config example less confusing Signed-off-by: Elena Rastorgueva <[email protected]> * Fix typo Signed-off-by: Elena Rastorgueva <[email protected]> * Clarify that YAML file is config file in README Signed-off-by: Elena Rastorgueva <[email protected]> * Remove unused imports Signed-off-by: Elena Rastorgueva <[email protected]> * Remove SDP docs for now Signed-off-by: Elena Rastorgueva <[email protected]> * Remove links to docs in SDP README Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Co-authored-by: Igor Gitman <[email protected]> * [Bugfix] Added rm -f / wget- nc command in multispeaker sim notebook to r1.13.0 (NVIDIA#5375) * Fix minor error in notebook Signed-off-by: Taejin Park <[email protected]> * changed branch name in tutorial notebook Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Taejin Park <[email protected]> * Rename Speech Dataset Processor to Speech Data Processor (NVIDIA#5378) Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> * fix for num worker 0 causing issues in losses after 1 epoch (NVIDIA#5379) * Fixed bug in notebook (NVIDIA#5382) Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Virginia Adams <[email protected]> * Force MHA QKV onto fp32 (NVIDIA#5391) Signed-off-by: smajumdar <[email protected]> Signed-off-by: smajumdar <[email protected]> * Added scheduling variety * ref * Fix for prompt table restore error (NVIDIA#5393) * Fix for prompt table restore error Signed-off-by: Virginia Adams <[email protected]> * Added more saftey checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Added more condition checks Signed-off-by: Virginia Adams <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Virginia Adams <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Fix args (NVIDIA#5410) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * bugfix * import tests * Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421) Signed-off-by: Yu Yao <[email protected]> Signed-off-by: Yu Yao <[email protected]> * Megatron Export Update (NVIDIA#5343) * export update for Megatron + change ORT optimization Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated export_utils to use autocast instead of manually casting >:/ Signed-off-by: David Mosallanezhad <[email protected]> * removed dtype from LayerNorm Signed-off-by: David Mosallanezhad <[email protected]> * added comment Signed-off-by: David Mosallanezhad <[email protected]> * reverting changes on FloatCast Signed-off-by: David Mosallanezhad <[email protected]> * Cherry-picked changes from megatron-norm Signed-off-by: Boris Fomitchev <[email protected]> * updated asr_model import to cast_utils Signed-off-by: David Mosallanezhad <[email protected]> * updated del onnx_model place Signed-off-by: David Mosallanezhad <[email protected]> * changed ort optimization to basic -> temp fix Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Boris Fomitchev <[email protected]> * disable pc test (NVIDIA#5426) Signed-off-by: ekmb <[email protected]> Signed-off-by: ekmb <[email protected]> * Fix GPT generation when using sentencepiece tokenizer (NVIDIA#5413) * Fix Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Disable sync_batch_comm in validation_step for GPT (NVIDIA#5397) * disable sync_batch_comm in validation_step Signed-off-by: ericharper <[email protected]> * Read sync_batch_comm from config or default to False Signed-off-by: Markel Sanz Ausin <[email protected]> * Update megatron_gpt_config to default sync_batch_comm to False to avoid CUDA error Signed-off-by: Markel Sanz Ausin <[email protected]> * Empty Signed-off-by: MaximumEntropy <[email protected]> * Comment out test Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Revert "Add temporary fix for CUDA issue in Dockerfile (NVIDIA#5421)" (NVIDIA#5431) This reverts commit 0718b17. * Revert workaround for T5 that sets number of workers to 0 & sync_batch_comm=False (NVIDIA#5420) * Revert workers workaround Signed-off-by: MaximumEntropy <[email protected]> * Fix in config Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Fixed discrepancies * updated Jenkisfile * updated Jenkisfile * Cleaning * fixed the onnx bug in conformer for non-streaming models. (NVIDIA#5242) (NVIDIA#5446) Signed-off-by: Vahid <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> * Set sync_batch_comm in other places (NVIDIA#5448) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Radtts 1.13 (NVIDIA#5451) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * [TTS] add CI test for RADTTS training recipe. Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Radtts 1.13 plus (NVIDIA#5457) * [TTS] Fixing RADTTS training - removing view buffer and fixing accuracy issue (NVIDIA#5358) * Fixing RADTTS training - removing view buffer and fixing accuracy issue * Fixes for Torchscript/Triton * Added autocast to radtts UT * using cuda() for training example Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> * Add num layers check (NVIDIA#5470) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Change to kwargs (NVIDIA#5475) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * Support for finetuning and finetuning inference with .ckpt files & batch size refactoring (NVIDIA#5339) (NVIDIA#5478) * Initial refactor Signed-off-by: MaximumEntropy <[email protected]> * Resolve config before passing to load_from_checkpoint Signed-off-by: MaximumEntropy <[email protected]> * Fixes for model parallel and nemo restore Signed-off-by: MaximumEntropy <[email protected]> * Fixes for eval Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Revert config changes Signed-off-by: MaximumEntropy <[email protected]> * Refactor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix typo Signed-off-by: MaximumEntropy <[email protected]> * Remove comments Signed-off-by: MaximumEntropy <[email protected]> * Minor Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix validation reconfiguration Signed-off-by: MaximumEntropy <[email protected]> * Remove old comment Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fixes for test_ds Signed-off-by: MaximumEntropy <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Signed-off-by: MaximumEntropy <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * export_utils bugfix (NVIDIA#5480) * updated export_utils Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Export fixes for Riva (NVIDIA#5496) * Export fixes for Riva Signed-off-by: Boris Fomitchev <[email protected]> * Cleaning up training_utils Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> * minor bug fix (NVIDIA#5521) Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> * added set_start_method + function param bugfix (NVIDIA#5539) * added set_start_method + function param bugfix Signed-off-by: David Mosallanezhad <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * upper bound torchmetrics Signed-off-by: ericharper <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: ericharper <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: ericharper <[email protected]> * remove notebook (NVIDIA#5548) Signed-off-by: ericharper <[email protected]> Signed-off-by: ericharper <[email protected]> * Remove broadcast (NVIDIA#5558) Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * cleaning * Fix all gather while writing to a file during T5 finetuning (NVIDIA#5561) * Gather from data parallel only instead of all ranks Signed-off-by: MaximumEntropy <[email protected]> * Fix Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> * update readme Signed-off-by: ericharper <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added copyright * fixed imports * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * cleaning * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed filesize check * last cleaning Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * updated cmudict path * fixed merge bug Signed-off-by: Evgeniy Shabalin <[email protected]> * warnings fix * fix warnings Signed-off-by: Evgeniy Shabalin <[email protected]> * storing * updated version Signed-off-by: Evgeniy Shabalin <[email protected]> * update Jenkinsfile versions Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed issues Signed-off-by: Evgeniy Shabalin <[email protected]> * fixed more issues * more fixes Signed-off-by: Evgeniy Shabalin <[email protected]> * added experimental tag * Clarification updates Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * remove old cython code Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * docstring fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * Enhancements Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * imports fix Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix typo Signed-off-by: Evgeniy Shabalin <[email protected]> * excessive comtutations fix Signed-off-by: Evgeniy Shabalin <[email protected]> * typecheck fix Signed-off-by: Evgeniy Shabalin <[email protected]> * Small refactoring * Small refactoring Signed-off-by: Evgeniy Shabalin <[email protected]> * reversed exp_manager params Signed-off-by: Evgeniy Shabalin <[email protected]> * Fixed call for new function signature Signed-off-by: Evgeniy Shabalin <[email protected]> * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci Signed-off-by: Oktai Tatanov <[email protected]> Signed-off-by: Jason <[email protected]> Signed-off-by: ericharper <[email protected]> Signed-off-by: Boris Fomitchev <[email protected]> Signed-off-by: nithinraok <[email protected]> Signed-off-by: Yang Zhang <[email protected]> Signed-off-by: MaximumEntropy <[email protected]> Signed-off-by: ekmb <[email protected]> Signed-off-by: PeganovAnton <[email protected]> Signed-off-by: Jocelyn Huang <[email protected]> Signed-off-by: fayejf <[email protected]> Signed-off-by: smajumdar <[email protected]> Signed-off-by: SeanNaren <[email protected]> Signed-off-by: Matvei Novikov <[email protected]> Signed-off-by: Igor Gitman <[email protected]> Signed-off-by: Sasha Meister <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Elena Rastorgueva <[email protected]> Signed-off-by: Taejin Park <[email protected]> Signed-off-by: Virginia Adams <[email protected]> Signed-off-by: Yu Yao <[email protected]> Signed-off-by: David Mosallanezhad <[email protected]> Signed-off-by: Markel Sanz Ausin <[email protected]> Signed-off-by: Oleksii Kuchaiev <[email protected]> Signed-off-by: Vahid <[email protected]> Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Evgeniy Shabalin <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: [email protected] <[email protected]> Co-authored-by: Oktai Tatanov <[email protected]> Co-authored-by: jasonjjl1999 <[email protected]> Co-authored-by: martynwei <[email protected]> Co-authored-by: Ryan Hong <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: ericharper <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Nithin Rao <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: Ramanathan Arunachalam <[email protected]> Co-authored-by: bene-ges <[email protected]> Co-authored-by: Alexandra Antonova <[email protected]> Co-authored-by: Yang Zhang <[email protected]> Co-authored-by: Sandeep Subramanian <[email protected]> Co-authored-by: Evelina <[email protected]> Co-authored-by: PeganovAnton <[email protected]> Co-authored-by: Jocelyn <[email protected]> Co-authored-by: fayejf <[email protected]> Co-authored-by: Somshubra Majumdar <[email protected]> Co-authored-by: Sean Naren <[email protected]> Co-authored-by: Matvei Novikov <[email protected]> Co-authored-by: Zhilin Wang <[email protected]> Co-authored-by: Igor Gitman <[email protected]> Co-authored-by: Sasha Meister <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Elena Rastorgueva <[email protected]> Co-authored-by: Taejin Park <[email protected]> Co-authored-by: Adi Renduchintala <[email protected]> Co-authored-by: Virginia Adams <[email protected]> Co-authored-by: yaoyu-33 <[email protected]> Co-authored-by: David <[email protected]> Co-authored-by: David Mosallanezhad <[email protected]> Co-authored-by: Boris Fomitchev <[email protected]> Co-authored-by: Yi Dong <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Markel Sanz Ausin <[email protected]> Co-authored-by: Oleksii Kuchaiev <[email protected]> Co-authored-by: Vladimir Bataev <[email protected]> Co-authored-by: Vahid Noroozi <[email protected]> Co-authored-by: Xuesong Yang <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Signed-off-by: smajumdar [email protected]
What does this PR do ?
Attempt to fix fp16 with MHA in ASR
Collection: [ASR]
Before your PR is "Ready for review"
Pre checks:
PR Type: