- Add gradio_asr_demo_api_video.py: Gradio web UI supporting audio/video upload,
streaming output, hotwords, and Cloudflare tunnel
- Add demo/asr_demo/: demo audio and video files for the Gradio interface
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- Add MPS device choice and auto-detect MPS availability
- Change default attention implementation to 'auto' with smart fallback
- Auto-detect flash_attention_2 availability on CUDA, fallback to sdpa
- Use sdpa for MPS and CPU devices (flash_attention_2 not supported)
- Use float32 dtype for MPS/CPU devices for better compatibility
Fixes#206
The model_path was being converted to a Path object and then back to string
for from_pretrained() calls. This is unnecessary since HuggingFace accepts
strings directly, and causes issues on Windows where Path() converts forward
slashes to backslashes (e.g., "microsoft/VibeVoice-Realtime-0.5B" becomes
"microsoft\VibeVoice-Realtime-0.5B").
This fix:
- Keeps model_path as a string (no behavior change on Linux/macOS)
- Fixes Windows compatibility for HuggingFace repo IDs
- Removes redundant str() conversions
JianweiYu