modelscope / FunASR

A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.
https://www.funasr.com
Other
6.99k stars 744 forks source link

vad 模式切割的音频在转录后会自己删除吗 #2197

Open lesrose opened 1 week ago

lesrose commented 1 week ago

vad_model="fsmn-vad", vad_kwargs={"max_single_segment_time": 60000},

加了这两个参数会切割音频,在转录玩音频后,切割的音频文件还在吗