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
4.46k stars 493 forks source link

优化merge segments 参数,解决新闻联播男女主持人“晚上好”合并一个speakid问题 #1861

Closed wuhongsheng closed 9 hours ago