BaoCut integrates Fudan's moss transcription model with local speaker diarization

BaoCut integrates Fudan's moss-transcribe-diarize model (1.73 GB, local) for speaker diarization. Tests by dotey show it is slower than Qwen but more accurate on mixed Chinese-English audio.

2026-10-29 ~ 2026-10-29 · 2 related posts

• BaoCut ships local moss-transcribe-diarize model with speaker ID, alongside qwen3-asr and whisper — huangyun_122 · 2026-10-29
• dotey tries BaoCut transcription: slower than Qwen but better at mixed Chinese-English — dotey · 2026-10-29