Skip to content

majiajue/whisper-web

 
 

Repository files navigation

Whisper Web

Web application for accurate speech-to-text conversion, powered by OpenAI Whisper and Pyannote.

whisper.hydev.org

Features

  • Speech-to-text with timestamps
  • Speaker identification

Performance

Currently, the model can process 2 hours of audio in 12 minutes on an RTX 3060 graphics card.

Deployment Instructions

TODO

About

Insanely Fast Whisper + Diarization + Web

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 43.7%
  • Svelte 42.9%
  • CSS 7.7%
  • TypeScript 2.2%
  • HTML 2.1%
  • JavaScript 1.4%