Open shahuzi opened 4 years ago
@shahuzi yes because this way I am able to train a model easily. But in this repo I am sampling from Gaussian distribution.
@shahuzi yes because this way I am able to train a model easily. But in this repo I am sampling from Gaussian distribution.
Can you make more explanations about this repo ? For example, the kl_loss between VAE-Tacotron and GMVAE-Tacotron is same ?
Did u find some other GMVAE repo , which can follow?
Hi, I've read your code, and it seems that your implementation is a VAE-Tacotron, but not a GMVAE-Tacotron? Because the loss here seems is no difference from this repo.?