Add FunASR — open-source speech recognition toolkit #107

Hi! Great collection of local AI tools.

Would you consider adding FunASR to the speech/audio section?

FunASR

• What: Industrial-grade open-source speech recognition toolkit (ASR, VAD, punctuation, speaker diarization)
• Why local: Runs fully offline, no API calls needed, Apache 2.0
• Performance: SenseVoice-Small processes 10s audio in ~70ms
• Languages: 50+
• GitHub: https://github.com/modelscope/FunASR (16K+ stars)
• Install: pip install funasr

Also relevant

• SenseVoice — Fast multi-task speech model (ASR + emotion + audio events): https://github.com/FunAudioLLM/SenseVoice (8K+ stars)
• Fun-ASR-Nano — Encoder-LLM model, HuggingFace native: https://github.com/FunAudioLLM/Fun-ASR

All run locally on consumer hardware.