TalkNet2-pytorch icon indicating copy to clipboard operation
TalkNet2-pytorch copied to clipboard

Synthesized speech to fast

Open Saartjes opened this issue 4 years ago • 1 comments

Hi there, I must say I’m impressed by the results Talknet is able to achieve with relatively short training. I have been playing around with it but keep running into a problem. The synthesized voice speaks very fast. Way faster than the input training data.

Is this something you have encountered as well? Do you know of any tricks that might help counter this issue?

Saartjes avatar Jul 24 '21 13:07 Saartjes

How were you able to get started with the repo?

Helaly96 avatar May 21 '22 22:05 Helaly96