Vietnamese Text to Speech library
Full Changelog: https://github.com/NTT123/vietTTS/compare/v0.3.1...v0.4.1
Improve audio quality with fine-tuned HiFi-GAN.
Add word separator token!
Set silence audio segments to zero to reduce noise.
Use the nvidia/tacotron2/prenet with two dropout layers.
Use HiFiGAN vocoder for faster and better vocoder.
Update waveRNN pretrained model.
Model training is now working correctly.
We have a working quick start script.