Mylo

Yeah, once i fix that, v2 48k should also work

> 48k failed for me on training. I got the d/s converted. The 2nd wav folder doesn't generate files. The one where the # of items would be double. All...

> > > 48k failed for me on training. I got the d/s converted. The 2nd wav folder doesn't generate files. The one where the # of items would be...

Try running it again, it might work. Looks like it ran out of memory there.

It might have something to do with recent changes to pytorch, as the new default CUDA version for pytorch is now 12.1 instead of 11.8. I'll lock it back to...

This could really help improve outputs from models such as AudioLDM, I'll add it to the backlog.

That's weird, that's a circular import in audiocraft, not my code. Either there's something wrong with your python install, or with audiocraft, and seeing that it hasn't released an update...

It's a bit complicated. windows/linux + nvidia -> regular pytorch cuda windows + amd/intel/nvidia -> pytorch-directml linux + amd -> pytorch rocml And I'm not even sure if they are...

Check if you can reach https://pytorch.org/

I have pushed a commit to change the torch version to 11.8, as suggested by @hggz My apologies for my inactivity. I've been very busy with work. The goal was...