TMElyralab
Which version of hugging_face to use here?
Hey @anuj509 not exactly related to this, but I've been having issues trying to run MuseV on colab, with a similar issue relating to cache https://github.com/TMElyralab/MuseV/issues/173#issue-2830723725. After looking at your colab notebook I'm going to try removing the cache line and see if it works. Just wanted to know if that is the issue there too and you had any idea?
Here's a working Colab notebook I cobbled together : https://colab.research.google.com/drive/14SaeDn80Dnau8Uh8dRVAtdDkQhb40sEF?usp=sharing
I'm sure it could be streamlined by someone with more knowledge than I, but it DOES work, if you follow the instructions (I've included some important comments with the "#" lines, be sure to follow those.) Let me know if anything is unclear or if it's not working for you.
Thank you @ActionStripes it seems to be working, though for me i had to change to python 3.11.
Glad to be of help! BTW, if you don't like the look of the mouth/face in the output video, Tencent recently released their GFPGAN version 1.4 model, which does wonders to restore quality, and it's quite fast: https://github.com/TencentARC/GFPGAN
@ActionStripes Thanks once again, i was meaning to implement something similar, i was considering to use codeformers repo, but since i want this to be as low latency as possible, GFPGAN would be something to check out
Hey I have created colab notebook you can try that
ImportError: cannot import name 'cached_download' from 'huggingface_hub' (/usr/local/lib/python3.10/dist-packages/huggingface_hub/init.py) Getting this error again after both cp commands!! python --version - 3.10.12
Hey I have created colab notebook you can try that
ImportError: cannot import name 'cached_download' from 'huggingface_hub' (/usr/local/lib/python3.10/dist-packages/huggingface_hub/init.py) Getting this error again after both cp commands!! python --version - 3.10.12
You might try:
!pip uninstall huggingface_hub !pip install huggingface-hub==0.25.2
This is the only version of huggingface_hub I've gotten to work with this project.
@ActionStripes can you please share a colab link that works .. I have been trying for some days .. sad talker or wav 2 lip is not giving me eye blink and head movement as required
@Vaishnavi-seetharama I will take a look at the Colab notebook I posted previously and see if it can be updated, although I don’t know if that will help with issues regarding eye blink or head movement - MuseTalk only performs lip sync on already existing videos, as far as I know. Is this: https://github.com/TMElyralab/MuseV more what you’re looking for?