# facebookresearch/encodec

State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.

Repository: https://github.com/facebookresearch/encodec
Canonical: https://ross.abutalabs.com/products/encodec
Language: Python
License: MIT
License Family: permissive
Last push: 2024-01-04T01:58:44+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1413, "days_push": 973, "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 4041, forks 360 (observed 2026-08-28T04:08:33.512890+00:00)

## What it is
EnCodec is a deep learning based neural audio codec from Meta AI that compresses mono 24 kHz and stereo 48 kHz audio to bitrates from 1.5 to 24 kbps with high fidelity. It provides pre-trained encoder/decoder models with residual vector quantization, plus optional entropy coding via a small language model for up to 40% further compression.

## Use cases
- compress audio with a neural codec
- get audio tokens for training audio language models
- encode music at low bitrates with high quality
- use EnCodec through Hugging Face Transformers
- build text-to-audio or music generation pipelines needing discrete audio tokens

## When to choose
- you need state-of-the-art neural audio compression at low bitrates
- you need discrete audio token representations for generative audio models
- you want pre-trained 24 kHz mono or 48 kHz stereo models

## When to avoid
- you need a production codec for real-time communication (use Opus instead)
- you require a permissive license for commercial use of the PyPI weights (CC BY-NC)
- you need active development or frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: audio-processing, machine-learning, deep-learning, compression
- domain: machine-learning, deep-learning
- platform: python, cross-platform
- tags: neural-codec, audio-compression, vector-quantization, pretrained-models, research, audio, gpu

## Member repositories
- facebookresearch/encodec (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:33.512890+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 2026-08-29T18:23:45.814962+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/encodec (fetched 2026-08-28T04:08:33.512890+00:00, sha 06fc36e33a54)
  - registry_pypi: https://pypi.org/pypi/encodec/json (fetched 2026-08-29T09:16:11.777770+00:00, sha 91e7cb21553d)
- Data as of 2026-08-30T08:39:29.467469+00:00.