Is there any Chinese-English-Mixed model weights?

[busyyang]

Notice: In order to resolve issues more efficiently, please raise issue following the template.
（注意：为了更加高效率解决您遇到的问题，请按照模板提问，补充细节）

❓ Questions and Help

In our application, we would like recognize mixed Chinese and Engish. I used paraformer-large in docker demo, it works very good in Chinese, but poor performance in English. The main reason could be the datasets including 60K hours Chinese audio and small scales in English audio.

Is there plan to release a Chinese-English-mixed models, or anyone refined a weight?

[dengcunqin]

Yes, I trained a Chinese-English mixed model, and it performed very well. Specifically, I fine-tuned it by adding an English dataset.

[busyyang]

@dengcunqin do you open the weigits for testing?