Explanation for loss_disc, loss_gen, loss_fm,loss_mel, loss_kl?

jaywalnut310 / vits

VITS: Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech

https://jaywalnut310.github.io/vits-demo/index.html

MIT License

6.91k stars 1.27k forks source link

Explanation for loss_disc, loss_gen, loss_fm,loss_mel, loss_kl? #173

Open chigkim opened 1 year ago

chigkim commented 1 year ago

Could someone explain what loss_disc, loss_gen, loss_fm,loss_mel, loss_kl? Especially, Which number should I watch in order to spot model convergence/early stopping? Thanks!

bryansmart commented 1 year ago

I would also appreciate some detail about these values. Right now, I can only evaluate training progress by listening at each step.

Poney01234 commented 1 year ago

Up plz

tranctan commented 1 year ago

These are somewhat answered in this thread.