fatchord / WaveRNN

WaveRNN Vocoder + TTS

Home Page:https://fatchord.github.io/model_outputs/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

AssertionError: python preprocess.py

shipcommit opened this issue · comments

I'm getting this error log after running python preprocess.py:

13100 wav files found in "../Training data/LJSpeech-1.1/wavs/"

Traceback (most recent call last):
  File "preprocess.py", line 72, in <module>
    text_dict = ljspeech(path)
  File "D:[directory]\WaveRNN-master\utils\text\recipes.py", line 9, in ljspeech
    assert len(csv_file) == 1
AssertionError
(pytorch) PS D:[directory]\WaveRNN-master> 

I downloaded the LJ Speech Dataset, and kept the default folder structure.

Am I supposed to specify the sentence training data CSV somewhere? If I try and move it to the same folder as my wav files it results in a runtime error.

commented

Try removing "/wavs/" from your path, pointing to the directory with .csv file should be sufficient.