Ian Boyles

I agree I have just finishing training but it won't inference correctly. I need the colab to be updated for inferencing with a custom trained model and more documentation.

you probably ran out of compute time during proccessing, I think I ran this on google colab also may not just work in jupyter environment.

The first processing steps are compute heavy and take a few hours.

I second this what hardware is required for real time? On colab for running inference after having to run asr.py on any audio before hand makes it 4 to 5...

Dang this doesn't work on colab as you can't install anything that low due to python version fixed at 3.10.10

lol use my makeittalk colab it works

Yeah we need a Colab for this stuff that doesn't crash from ram out of memory lol

Yeah I have yet to see any model truly offer real time for facial animation yet that I could use

Hey I have a repo posted with need instructions and changes to make it work in present day.

There's was a few changes for package versions I believe, there's also a TWO_D change as well. they are in caps when replaced.