This is the most painful AI half-vocoder I've ever seen lol, great tutorial XD

There has been some changes in the colab script code in the past few days, I would recommend going to official doc and grabbing training links from there:

i'm using so-vits-svc-fork-4.0 , and connection interrupted , now what are the exact steps to do when i reload the colab to resume training , the tensorboard stopped at 5000 Steps, but in my google drive it only saved ( D_960.pth and G_960.pth) so please could you tell me what to do cell by cell starting from check GPU to resume training, Thanks

I tried this, but got lost somewhere along the way - my own fault! I want to undo everything I did because I'm concerned about all the things I installed and permissions I gave. How do I do this? I want to make it like I never attempted it.

Big apologize for people trying to run the UI 4.0, it looks like the code is different enough that it can't run on the same environment as 3.0 (made a bad assumption that it did without actually doing the testing). Here are the code lines I've used for new conda environment to make it work:

nice

When getting error that looks like this: 'ValueError: n_samples=598 should be >= n_clusters=4000. 0 0.583984375 MB , shape: (598, 256) float32'

i need your helpppp

For those who want to train the cluster model, go throught the entire process of preparing the wavs file for training, BUT just before the final training cell add this code (and let it run for ten or so minutes):

i cant find solution to "Could not find TensorRT" error :/

when i run the third ones, i saw this error:

Familiar voice, I'd say a Gothic voice

Temporary solution for Colab not connecting/finding the G_ and D_ files is to create a new cell above 'start training' cell and fill it with this code:

4:55 I put that no such file or directory, what I do?

that background music sounds like something out of CIV

I need anymore than google chrome? Can I do this on Mac os Catalina intel