search

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
6.94k stars 737 forks source link

TeleSpeech-ASR1.0对方言支持的比较好，建议将这个模型添加到funasr中使用 #1768

Open Hunter6324 opened 5 months ago

Hunter6324 commented 5 months ago

❓ Questions and Help

TeleSpeech-ASR1.0对方言支持的比较好，建议将这个模型添加到funasr中使用。模型地址：https://www.modelscope.cn/models/TeleAI/TeleSpeech-ASR1.0/summary

What is your question?

目前funasr对普通话的支持程度已经很完善了，但是对方言的识别还不够好。与之相反，TeleSpeech-ASR1.0对方言支持的比较好，两模型存在互补关系。建议将这个模型也添加到funasr中，最好同样支持docker部署。

LauraGPT commented 5 months ago

Wait for two weeks.

Zzzhq947 commented 4 months ago

请问现在能支持TeleSpeech-ASR1.0模型了吗