# fatchord/WaveRNN

WaveRNN Vocoder + TTS

Repository: https://github.com/fatchord/WaveRNN
Canonical: https://ross.abutalabs.com/products/wavernn
Homepage: https://fatchord.github.io/model_outputs/
Language: Python
License: MIT
License Family: permissive
Topics: wavernn, pytorch, neural-vocoder, speech-synthesis, tts, tacotron, text-to-speech
Last push: 2022-07-02T14:21:35+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3092, "days_push": 1523, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2190, forks 686 (observed 2026-08-28T04:06:24.070966+00:00)

## What it is
A PyTorch implementation of DeepMind's WaveRNN neural vocoder plus a Tacotron text-to-speech system, trained on LJSpeech. It supports training custom models and generating speech from text with pretrained weights.

## Use cases
- synthesize speech from text with a neural vocoder
- train a WaveRNN vocoder on my own dataset
- run Tacotron TTS with pretrained LJSpeech models
- generate high-quality audio from mel spectrograms
- experiment with efficient neural audio synthesis

## When to choose
- you want a lightweight, efficient neural vocoder in PyTorch
- you need a complete Tacotron + WaveRNN TTS pipeline with pretrained models
- you want to train a vocoder on LJSpeech or custom audio

## When to avoid
- you need actively maintained code or modern TTS architectures
- you want production-grade, multi-speaker or multilingual TTS
- you lack a CUDA GPU for reasonable generation speed

## Facets
- artifact type: library
- maturity: maintenance
- function: tts, machine-learning, deep-learning, audio-processing
- domain: speech-processing, machine-learning, deep-learning
- platform: python, cross-platform
- tags: wavernn, vocoder, tacotron, speech-synthesis, pytorch, ljspeech, neural-vocoder, audio, gpu, linux

## Member repositories
- fatchord/WaveRNN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:24.070966+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:47:41.269639+00:00, confidence not recorded.
  - readme: https://github.com/fatchord/WaveRNN (fetched 2026-08-28T04:06:24.070966+00:00, sha 1c09103d3b50)
  - homepage: https://fatchord.github.io/model_outputs/ (fetched 2026-08-29T10:27:56.773362+00:00, sha 189ced7bb34d)
- Data as of 2026-08-30T08:39:29.467469+00:00.