How to control the length of generated wav???

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

How to control the length of generated wav??? #71

Closed zuoxiang95 closed 7 years ago

zuoxiang95 commented 7 years ago

hey guys, i want to ask how to control the length of generated wav. i find the generated wav is about 4s. If my text is too long, it can not be finished in 4s. what should i do?

keicoon commented 7 years ago

I found this script in eval.py 44 line.

timesteps = 100  # Adjust this number as you want

Modify this script to want you increasing timesteps value.

zuoxiang95 commented 7 years ago

@keicoon You are right. Thanks!