Feature Request: Add SenseVoice as STT option (faster than Whisper)

[LauraGPT]

Feature Request

Verbi is a great modular voice assistant for experimenting with STT models. Suggesting SenseVoice as an additional STT option.

Why SenseVoice?

• 5x faster than Whisper — lower conversation latency
• Non-autoregressive — constant latency
• 234M params — lightweight
• 50+ languages — multilingual
• Emotion detection — could enable emotion-aware assistant responses
• OpenAI-compatible API — funasr-server serves /v1/audio/transcriptions

Integration

from funasr import AutoModel
model = AutoModel(model="iic/SenseVoiceSmall", vad_model="fsmn-vad")
result = model.generate(input=audio)

• FunASR: https://github.com/modelscope/FunASR (16.6K stars)
• SenseVoice: https://github.com/FunAudioLLM/SenseVoice (8.3K stars)