This is a repository to train TTS models. All the training/finetuning are focusing on Bengali language here but can be used for any languages with minimal changes. The most confusing part is the data formatting; once you get it right for the dataset in hand, the rest just becomes easier. I'll be posting more about it gradually. I will also improve the other parts of the repo as well.
Bangla TTS contains a list of models trained/tuned using scripts from this repository. Model that we have added to the collections so far are: