Tacotron-2
Tacotron-2 copied to clipboard
Question: How can I train the model on my own data
I just have a big folder full of wav-files and want to train the model with these.
What do I have to do to make the magic happen?
I am thankful for any help.
You should make your corpus have the same style with the LJspeech , ie, there is a wavs folder in your corpus and there is a metadata label file in your corpus too
Ok, thanks