Home
This repository contains a training recipe for training FastSpeech2 with Hybrid Segmentation (HS), a state-of-the-art text-to-speech (TTS) model. The training utilizes the Espnet toolkit and is tailored for the Indian languages, covering 13 major languages of India.
An overview of the training has been given in the Readme.md file. All the details regarding pre-processing, preparation of files, caution and tips are given in this wiki.
Please look for more details on the side space.
