search

NVIDIA / nv-wavenet

Reference implementation of real-time autoregressive wavenet inference
BSD 3-Clause "New" or "Revised" License
735 stars 126 forks source link

can't reproduce good wav #44

Closed azraelkuan closed 6 years ago

azraelkuan commented 6 years ago

i train a model with the train.py in the dir pytorch, but can not generate good wav using inference.py

wavs.zip

azraelkuan commented 6 years ago

image this is my config, and i use the ljspeech as my dataset

azraelkuan commented 6 years ago

after training to 200k, get good result. close this.