zhangdxchn

icon indicating a website link https://www.jianshu.com/u/85c5c609b260 icon indicating an email [email protected]

icon location Xiamen(Amoy) China icon location var zhangdxchn = new 攻城狮();

Results 3 comments of zhangdxchn

NotImplementedError: Could not run 'aten::empty_strided' with arguments from the 'CUDA' backend.

same issue, [see this](https://github.com/FunAudioLLM/CosyVoice/pull/365/files), work for me.

TransformerLM.__init__() missing 1 required positional argument: 'sampling'

Re-download Model. Try it. work for me. ``` # SDK模型下载 from modelscope import snapshot_download snapshot_download('iic/CosyVoice-300M', local_dir='pretrained_models/CosyVoice-300M') snapshot_download('iic/CosyVoice-300M-SFT', local_dir='pretrained_models/CosyVoice-300M-SFT') snapshot_download('iic/CosyVoice-300M-Instruct', local_dir='pretrained_models/CosyVoice-300M-Instruct') snapshot_download('iic/CosyVoice-ttsfrd', local_dir='pretrained_models/CosyVoice-ttsfrd') ```

This modification is compatible with MacOS M chip and runs in CPU mode.

> > > @alkaidzone The versions of torchaudio and torch should be the same. I use version 2.4.1. For all the dependencies, you can refer to the requirements.txt file in...