search

Kyubyong / tacotron

A TensorFlow Implementation of Tacotron: A Fully End-to-End Text-To-Speech Synthesis Model
Apache License 2.0
1.83k stars 436 forks source link

remove prenet in decode2 #52

Closed candlewill closed 7 years ago

candlewill commented 7 years ago

According to the paper, in the post processing net (namely decode2 in code), the pre-net is not used.

candlewill commented 7 years ago

dimension error