# lucidrains/audiolm-pytorch

Implementation of AudioLM, a SOTA Language Modeling Approach to Audio Generation out of Google Research, in Pytorch

Repository: https://github.com/lucidrains/audiolm-pytorch
Canonical: https://ross.abutalabs.com/products/audiolm-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: artificial-intelligence, attention-mechanisms, audio-synthesis, deep-learning, transformers
Last push: 2025-01-12T15:40:45+00:00

## Health v2 (maintenance only)
Score: 34/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 1, release rhythm 40, longevity 100
- inputs: {"age_days": 1454, "days_push": 598, "days_rel": 598, "gap_med": 0.0, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2627, forks 278 (observed 2026-08-28T04:07:05.443887+00:00)

## What it is
A PyTorch implementation of AudioLM, Google Research's language modeling approach to audio generation, including a MIT-licensed SoundStream implementation compatible with EnCodec. It extends the original work with T5 classifier-free guidance conditioning, enabling text-to-audio and TTS (VALL-E style) training.

## Use cases
- generate audio with a language model in pytorch
- train a text-to-speech model like VALL-E
- implement AudioLM from the Google Research paper
- train a SoundStream or EnCodec neural audio codec
- do text-to-audio generation with T5 conditioning
- replicate AudioLM research experiments

## When to choose
- you want a PyTorch codebase to train or experiment with AudioLM-style audio generation
- you need an open-source SoundStream implementation or EnCodec-compatible codec training
- you want to train a VALL-E-like TTS model from scratch

## When to avoid
- you need a production-ready, pretrained text-to-speech service out of the box
- you want a simple inference-only audio generation tool without training
- you need a maintained product with guaranteed support rather than a research implementation

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, audio-processing, tts, llm-training
- domain: deep-learning, artificial-intelligence, machine-learning
- platform: python
- tags: audiolm, audio-generation, pytorch, transformers, soundstream, text-to-audio, research-implementation, audio

## Member repositories
- lucidrains/audiolm-pytorch (main) score 34

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:05.443887+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 2026-08-30T02:19:45.120099+00:00, confidence not recorded.
  - readme: https://github.com/lucidrains/audiolm-pytorch (fetched 2026-08-28T04:07:05.443887+00:00, sha 2dbf03916096)
- Data as of 2026-08-30T08:39:29.467469+00:00.