0.2.17

QuentinFuxa/WhisperLiveKit0.2.17Dec 9, 2025by QuentinFuxa

AI Summary

Introduces native MLX AlignAtt for faster decoding and implements one shared ONNX Silero VAD model instance across sessions for better multi-speaker support.

Key Highlights

  • Native MLX AlignAtt for faster decoding (previously mixed MLX/PyTorch)
  • One shared ONNX Silero VAD model instance across sessions
  • Fix translation instructions in Docker container
  • Fix custom MLX path handling

New Features

  • Native MLX AlignAtt decoding
  • Shared ONNX Silero VAD model across sessions

Full Release Notes

## Improvements
- Native MLX AlignAtt for faster decoding (previously mixed MLX/PyTorch).
- One shared ONNX Silero VAD model instance across sessions for better multi-speaker support.

## Bug fixes
- Fix translation instructions when using Docker container (#307)
- Correct Chrome extension path (#274)
- Fix TypeError when translating (#301)
- Fix custom MLX path handling (#299)
- Fix LocalAgreement issue (#308)