skinny.

Latest / Kana & Mari's SoundRepos (English)

BogiHsu/Tacotron2-PyTorch | PyTorch text to speech synthesis | GitHub

A PyTorch implementation of Tacotron 2 that converts text into mel spectrograms for speech synthesis. It includes LJ Speech data preprocessing, single- and multi-GPU training, checkpoint-based inference, and compatibility with external WaveGlow and HiFi-GAN vocoders. Related: BogiHsu/Tacotron2-PyTorch / PyTorch text to speech synthesis / GitHub

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.