daanelson / whisperx

Accelerated transcription of audio using WhisperX

  • Public
  • 49.9K runs
  • GitHub
  • Paper
  • License
  1. Author
    @daanelson
    Version
    22.04

    9aa6ecad

    Latest

    Release notes

    Whisper large-v2 model. Supports transcription in all whisper languages. Alignment is only in English for efficiency - the model has to load a separate alignment model for each language. Reach out if you need alignment in other languages.