Hello @A-Jacobson.

Great work with your implementation and more importantly with you clear representation of the model in your README (100% better that the one presented in the paper x) ).

So I am actually also working on Tacotron 2 implementation (in tensorflow) and there are few things I wanted to check with you, maybe we could help each other out. (implementation here)

• Does your attention mechanism work? Mine (based on the original tacotron) doesn't seem to capture the alignment correctly.
• Does your loss decrease insanely fast?

Again, impressive work.

Do you mind posting some samples?

Thanks!

Hi,
I am encountering this error in the decoding-helpers file. When I googled, it suggested me to install pytorch from source and so I did.
It is still showing me this error.
Can you help me?
stop_tokens[t] = stop_token
RuntimeError: expand(torch.cuda.FloatTensor{[12, 1]}, size=[12]): the number of sizes provided (1) must be greater or equal to the number of dimensions in the tensor (2)

Thanks!