Industrial-grade speech recognition: 170x realtime, 50+ languages, speaker diarization
FunASR is an industrial-grade end-to-end speech recognition toolkit. 170x faster than Whisper with 50+ language support, speaker diarization, emotion detection, streaming processing, and OpenAI-compatible API for seamless integration.
Author: modelscope
Stars: 17,976
Install: pip install funasr