Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update decoding script for gigaspeech and remove duplicate files. #361

Merged
merged 1 commit into from
May 13, 2022
Merged

Update decoding script for gigaspeech and remove duplicate files. #361

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 20 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,13 +12,14 @@ for installation.
Please refer to <https://icefall.readthedocs.io/en/latest/recipes/index.html>
for more information.

We provide four recipes at present:
We provide 6 recipes at present:

- [yesno][yesno]
- [LibriSpeech][librispeech]
- [Aishell][aishell]
- [TIMIT][timit]
- [TED-LIUM3][tedlium3]
- [GigaSpeech][gigaspeech]

### yesno

Expand Down Expand Up @@ -197,6 +198,23 @@ The best WER using modified beam search with beam size 4 is:

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1je_1zGrOkGVVd4WLzgkXRHxl-I27yWtz?usp=sharing)

### GigaSpeech

#### Conformer CTC

| | Dev | Test |
|-----|-------|-------|
| WER | 10.47 | 10.58 |

#### Pruned stateless RNN-T

| | Dev | Test |
|----------------------|-------|-------|
| greedy search | 10.59 | 10.87 |
| fast beam search | 10.56 | 10.80 |
| modified beam search | 10.52 | 10.62 |


## Deployment with C++

Once you have trained a model in icefall, you may want to deploy it with C++,
Expand Down Expand Up @@ -225,4 +243,5 @@ Please see: [![Open In Colab](https://colab.research.google.com/assets/colab-bad
[aishell]: egs/aishell/ASR
[timit]: egs/timit/ASR
[tedlium3]: egs/tedlium3/ASR
[gigaspeech]: egs/gigaspeech/ASR
[k2]: https://github.com/k2-fsa/k2
8 changes: 4 additions & 4 deletions egs/gigaspeech/ASR/RESULTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,10 +5,10 @@

#### Conformer encoder + embedding decoder

Conformer encoder + non-recurrent decoder. The encoder is a
reworked version of the conformer encoder, with many changes. The
decoder contains only an embedding layer, a Conv1d (with kernel
size 2) and a linear layer (to transform tensor dim). k2 pruned
Conformer encoder + non-recurrent decoder. The encoder is a
reworked version of the conformer encoder, with many changes. The
decoder contains only an embedding layer, a Conv1d (with kernel
size 2) and a linear layer (to transform tensor dim). k2 pruned
RNN-T loss is used.

Results are:
Expand Down
Loading