Feature: Add FunASR/SenseVoice as local ASR engine

[LauraGPT]

Note

License and capability clarification (2026-07-14): FunASR is a toolkit, not a single checkpoint. The FunASR and SenseVoice repository source code is MIT; model weights follow each model card. SenseVoiceSmall supports Chinese, Cantonese, English, Japanese, and Korean, and its weights use the linked FunASR Model Open Source License Agreement. Fun-ASR-Nano-2512 is Apache-2.0. Language coverage, punctuation, and performance depend on the selected model and runtime configuration.

Hi! Great voice input tool.

Would you consider FunASR or SenseVoice as a local ASR engine?

Why?

• SenseVoice-Small — ~70ms for 10s audio, ultra-fast for voice input
• Built-in punctuation — Clean text output ready to paste
• Streaming — Paraformer-streaming for real-time input
• CJK excellence — State-of-the-art on Chinese/Japanese/Korean
• 50+ languages, fully local, Apache 2.0

References

• FunASR: https://github.com/modelscope/FunASR (16K+ stars)
• SenseVoice: https://github.com/FunAudioLLM/SenseVoice (8K+ stars)