Simple Whisper

Transcription library/cli/server based on OpenAI Whisper model written using whisper-rs.

What is included?

• Simple Whisper lib: Implements the Whisper model via:

  • whisper.cpp Backend. Weights are automatically downloaded from Hugging Face repo.
  • Supported codec: flac, vorbis, wav, mp3

• Simple Whisper cli: CLI application useful to transcribe audio file. For more information see the README.md.

• Simple Whisper server: Websocket server that transcribe uploaded files.

Goals

• Show how malleable RUST is, scaling from server to GPU code.
• Support a high variety of platforms.
• Fast enough on every platform.

No Goals

• It is NOT intended to be the fastest/accurate Whisper implementation.

Credits

The project was inspired by:

• Candle implementation: rwhisper.