soobinseo/Transformer-TTS

how control wav length ?

Opened this issue · 1 comments

In synthesis.py,there's a parameter"--max_len",does this code artificially control the length of the synthesized audio through this parameter?Rather than being determined by the model?

Hi, sorry for late comment.
I tried to train the model with stop token loss, but it was failed. So far there is no way to control the wav length by the model.

Thanks.