0.2.2

QuentinFuxa/WhisperLiveKit0.2.2Jul 4, 2025by QuentinFuxa

AI Summary

Replaces the ffmpeg-python library with raw ffmpeg calls to resolve systematic crashes and improve server stability.

Key Highlights

Replaced ffmpeg-python with raw ffmpeg calls to fix crashes after 9 minutes on some machines.
Updated to the latest SimulWhisper for improved compatibility with English-only models.
Fixed buffer overflow prevention when no tokens are created.

New Features

Raw ffmpeg implementation
SimulWhisper update compatibility
Fixed buffer overflow prevention

Full Release Notes

### New:
- **Replace ffmpeg-python with raw ffmpeg calls**:
     - Fixes systematic crashes after 9 minutes on some machines
     - Improves reboot and restart handling
     - Allows ffmpeg to restart without crashing the server on conversion errors

- **Update to latest SimulWhisper:**
     - Adds compatibility with English-only models
     - Infers word-level timestamps for better diarization alignment
     - Other improvements: https://github.com/ufal/SimulStreaming/commits/main/
- Prevent buffer from growing indefinitely when no tokens are created
- Fix Hugging Face token file handling in Docker
- Remove default 8000 port in WebSocket when no port is provided