PyTorch implementation of Robust Universal Neural Vocoding

A speaker independent vocoder trained on the ZeroSpeech 2019: TTS without T English dataset


This page contains audio samples from https://github.com/bshall/UniversalVocoding:
a PyTorch implementation "Robust Universal Neural Vocoding", J. Lorenzo-Trueba Jaime et al, (2018).

Notable differences from the paper:

In-Domain Speakers


Speaker - V001

original generated

Speaker - V002

original generated

Out-of-Domain Speakers (English)


Speaker - S002

original generated

Speaker - S005

original generated

Speaker - S030

original generated

Out-of-Domain Speakers (Surprise Austronesian Language)


original generated