search

JunityZhan / Understanding-VITS

In this repository, you will learn how code works in VITS(Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech) in Jupyter Notebooks, including normalizing data, training process, inference process, and model's details.
GNU General Public License v3.0
159 stars 24 forks source link

<happy to help> #1

Open fkwlqm opened 1 year ago

fkwlqm commented 1 year ago

Hey, i just came across this repo. I have some understanding about the model as well and I am happy to help you if you are ok. I will wait for your notebook to get uploaded first and then fill in any details that I think are missing. Good day!

JunityZhan commented 1 year ago

Thank you, I am so happy that someone is coming here after I just create this repo, I am still working on it and I am glad if someone can point out any fault that I make in my understand.

fkwlqm commented 1 year ago

I was surprised too. Cause I wanted to do something similar and as soon as I searched, I found your repo as soon as you made first commit. small world

preslaff commented 6 months ago

Hi all! Can you give me any directions to train the VITS on a different language - E.g. Bulgarian. I used espeak-ng to phonemize my dataset, so I can skip some preprocessing steps I suppose?