Skip to content

Text file language translation app that translates .txt, .csv, and .xlsx files on the local device using lightweight, open-source LLMs from Hugging Face (no internet connection needed). Uses batch processing and token chunking which provides the ability to translate large file sizes while preserving semantic meaning across paragraphs / sections.

License

Notifications You must be signed in to change notification settings

thefilesareinthecomputer/offline_file_translation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

98f4145 · Sep 28, 2024

History

6 Commits
Apr 18, 2024
Apr 18, 2024
Apr 18, 2024
Jan 29, 2024
Sep 28, 2024
Apr 18, 2024
Apr 18, 2024

Repository files navigation

Text File Translation App

Capabilities

  • Translates text files from one language to another using Helsinki-NLP's OPUS-MT models from Hugging Face.
  • Supports multiple file formats: .txt, .csv, .xlsx.
  • Batch processing for handling large volumes of text.
  • Token chunking for efficient translation of lengthy text while avoiding memory issues.

Usage

  • Place your text file in an accessible location.
  • Select the desired source and target languages from the lists in the UI.
  • Select the file you want to translate in the UI file explorer window.
  • The translated file will be saved alongside the original, with the file name appended with the target language.

Benefits

  • Flexibility: Works with multiple languages.
  • Privacy: Performs translations locally and offline, ensuring data privacy.
  • Convenience: Easy to set up and run, making translations hassle-free.

Specifications

  • Language Support: Dependent on available Helsinki-NLP/OPUS-MT models.
  • File Format Support: Text (.txt), Comma-Separated Values (.csv), and Excel (.xlsx).
  • Dependencies: Python, pandas, torch, transformers, tqdm, openpyxl, pyinstaller.

Installation & Setup

git clone {REPO_URL} {REPO_FOLDER}

cd {REPO_FOLDER}

python3.11 -m venv {VENV_NAME}

source {VENV_NAME}/bin/activate
 
pip install --upgrade pip pip-check-reqs wheel python-dotenv

pip install -r requirements.txt

pip install {ADDITIONAL_PACKAGES}

pip freeze > requirements.txt

echo "{VENV_NAME}/
_archive/
_notes/
_notes.txt
generated_data/
venv/
__pycache__/
*.pyc
*/migrations/*
db.sqlite3
.env
staticfiles/" > .gitignore

cat .gitignore

git init

git add .

git commit -m "Initial commit"

optionally, set a remote repository and push the new code to it

About

Text file language translation app that translates .txt, .csv, and .xlsx files on the local device using lightweight, open-source LLMs from Hugging Face (no internet connection needed). Uses batch processing and token chunking which provides the ability to translate large file sizes while preserving semantic meaning across paragraphs / sections.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages