what is the simple step to add a new language support?

[dream-1ab]

Hello, firstll of all I love this project because of its excellent performance, so I decided to add my own language support to this project.

This project may a good bridge between my language and other GPT-SoviTTS supported languages. This connects people and their cultures.

Uyghur language is my native language and 10 million of people using it today.

I have basic neural network and CNN model creation and training experience.

By the way there are few questions in my mind:

1. what is the simplified steps to add new language support to this project?
2. What type of datasets are needed to be add my language support? And how much? (Both simple voice cloning and cross lingual voice conversion)
3. What is the requirement of datasets? (For example segmentation, Tokenizer preparation, Uyghur-English parallel text corpus or something like that)
4. What neural network architecture is this project uses? Is GPT-SoviTTS itself becomes a new neural network architecture?
5. What neural network architectures are I should learn for understanding how this project works?

And there may other questions that are not in my mind so let me know.
Thank you for your excellent work guys.
I wish one day I become a contributor of this project.