Credits:

AInonymous (App)
Clipper (Clipping voices)
CookiePPP (Notebooks)
Synthbot (Data preprocessing)
Flutteranon (Original Synthesis notebook)
Clipper (Hosting the /mlp/ dataset and many nights of clipping)
SortAnon (adding HiFi-GAN support and audio super-resolution)
Anons from /mlp/

NVIDIA/tacotron2
Natural TTS Synthesis By Conditioning Wavenet
On Mel Spectrogram Predictions
HiFi-GAN: Generative Adversarial Networks
for Efficient and High Fidelity Speech Synthesis