Rayhane-mamah / Tacotron-2

DeepMind's Tacotron-2 Tensorflow implementation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Compare to Nvidia's Pytorch Tacotron-2, the synthesizing speed is much slower

Thien223 opened this issue · comments

Who can please tell me, why there is this different?

As I thought, Tensorflow implementation should faster than pytorch.

Which part make this repo synthesizes slower than Nvidia?

Any comment is highly appreciated. Thank you guys