Terminal

Your project is doing a great job. Will it support generating Chinese in the future? If so, when do you expect to release a version that supports Chinese?

fix-issues-963:preprocess_audio.py--The .npy file failed to save

1

https://github.com/babysor/MockingBird/issues/963 

请问如何训练方言

1

佬，比如我要在底模基础上训练一个上海方言，我需要哪些前置准备呢。 比如 1. 需要多少小时的方言数据才能达到比较好的效果 2. 音素符合需要找一套方言的吗 3. 需要自备新的G2P吗 万分感谢！！！

TypeError: StableDiffusionImg2ImgPipeline.__call__() missing 1 required positional argument: 'init_image'

2

pip install diffusers==0.10.0

创空间的CosyVoice3很好用，但是开源的webui效果很差

2

创空间：https://www.modelscope.cn/studios/FunAudioLLM/Fun-CosyVoice3-0.5B 为啥创空间的效果很好，本地部署的会乱讲话，有空白呢