Skip to content

Uses the powerful WhisperS2T and Ctranslate2 libraries to batch transcribe multiple files

Notifications You must be signed in to change notification settings

BBC-Esq/WhisperS2T-transcriber

Repository files navigation

🚀WhisperS2T-transcriber🚀

  • Uses the powerful WhisperS2T and Ctranslate2 libraries to batch transcribe multiple files

Requirements

  1. 🐍Python 3.11
  2. 📁Git
  3. 📁Git Large File Storage
  4. 🪟 Windows (linux not yet supported)

    You can modify the installation instructions manually for Linux, I just don't have Linux in order to test them reliably. You can do this by analyzing the setup_gpu.py or setup_cpu_only.py file to see what libraries are required.

Installation

Download the latest release and extract the files your computer. Navigate to the respository folder, create a command prompt, and run the following commands:

python -m venv .
.\Scripts\activate

Run this again to activate the environment each time you restart the program.

Select GPU (includes CPU) or CPU-only installation:

python setup_gpu.py
python setup_cpu_only.py

Usage

python whispers2t_batch_gui.py

The program will process any and all of the following file types:

  • .mp3, .wav, .flac, .wma, .aac, .m4a, .avi, .mkv, .mp4, .asf, .amr.

Important

All transcriptions are output in the same folder of the file that was transcribed. If you'd like to change this behavior put an issue on Github requesting it.