r9y9/wavenet_vocoder
A PyTorch implementation of WaveNet vocoder for generating high-quality raw speech samples from linguistic or acoustic features.

This repository provides a neural vocoder implementation based on the WaveNet architecture. It generates raw audio waveforms conditioned on linguistic or acoustic features, supporting mixture of logistics, mixture of Gaussians, and single Gaussian distributions for 16-bit audio modeling. The vocoder is used as a component in text-to-speech systems to convert acoustic representations into speech, and integrates with the ESPnet speech processing toolkit.
Frequently asked
- What is r9y9/wavenet_vocoder?
- A PyTorch implementation of WaveNet vocoder for generating high-quality raw speech samples from linguistic or acoustic features.
- Is wavenet_vocoder open source?
- Yes — r9y9/wavenet_vocoder is an open-source project tracked on heatdrop.
- What language is wavenet_vocoder written in?
- r9y9/wavenet_vocoder is primarily written in Python.
- How popular is wavenet_vocoder?
- r9y9/wavenet_vocoder has 2.4k stars on GitHub.
- Where can I find wavenet_vocoder?
- r9y9/wavenet_vocoder is on GitHub at https://github.com/r9y9/wavenet_vocoder.