Empty generated waves about tacotron HOT 7 OPEN

Can you share your training loss graph?

from tacotron.

from tacotron.

From the TensorBoard. I guess 200 epochs, or 1000 global steps are far from enough. Note that the paper said they trained for 2 million global steps. And I think that's one of the biggest reasons why people including me have had difficulties getting a good result. (The other reason is, of course, some possible bugs in the implementation. But it's also true that mostly we don't have enough gpus or time or patience!) But the training curve keeps slowly decreasing, I think that's a good sign.

from tacotron.

@hmubarak I have the same problem here with yours.

All of the generated voice is like this:

Loss

from tacotron.

@candlewill As said before, we need to be more patient with the training. 800 global steps is very unlikely to yield good results.

from tacotron.

I faced the same problem. Did you solve this problem by keeping training?(i am now over 26k steps)

from tacotron.

@candlewill @Kyubyong @hmubarak How did you solve this issue? I trained for almost 50k steps. But still when I synthesize using synthesize.py, it is generating empty wavs. Can you please solve this issue?

from tacotron.