itechmusic
itechmusic
Hi there, thanks for your waiting. [MuseTalk](https://github.com/TMElyralab/MuseTalk) is now released. Please try it out and let us know.
@rmasiso hi，you can try install mediapipe
It seems the openmim is all right. What's your pc system? Have you re-pull image, or re-run docker image？
@rmasiso To figure out, you can try look into whole_body.py, and move try, leave from import only, and then see detail error.
> 足够了，推理耗时有统计吗？如果A10卡推理能撑得住吗？ @ApolloRay MuseV Hg GUI就是a10卡搭的。a100 12帧需要9s。
谢谢提醒，我们后面加到readme上
你好，请问有更详细的报错信息吗？
@kacecarol hi, can you upload your input and output？ we can update guidance later.
> Traceback (most recent call last): > File "/workspace/MuseV/scripts/inference/text2video.py", line 16, in > from diffusers.models.autoencoder_kl import AutoencoderKL > ModuleNotFoundError: No module named 'diffusers.models.autoencoder_kl' > > > 用docker拉取的 @dabaonidetou 需要用我们的修改过的diffusers仓哈，版本和huggingface diffusers不一样，最好按照我们的readme里介绍的进行环境配置
@dabaonidetou 具体原因我还不清楚，但的确是diffusers包版本切换的问题。 另外，代码和模型都可以在镜像外的系统里下载哈，然后挂载的方式挂载到docker 容器上。在容器里直接cd 到MuseV目录外设置pythonpath即可。