# LAION-AI/CLAP

Contrastive Language-Audio Pretraining

Repository: https://github.com/LAION-AI/CLAP
Canonical: https://ross.abutalabs.com/products/laion-ai-clap
Homepage: https://arxiv.org/abs/2211.06687
Language: Python
License: CC0-1.0
License Family: permissive
Last push: 2025-05-15T19:55:29+00:00

## Health v2 (maintenance only)
Score: 42/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 21, release rhythm 35, longevity 100
- inputs: {"age_days": 1641, "days_push": 475, "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 2256, forks 214 (observed 2026-08-28T04:06:31.422389+00:00)

## What it is
CLAP (Contrastive Language-Audio Pretraining) is a Python library from LAION that learns joint audio and text embeddings using contrastive learning, trained on the LAION-Audio-630K dataset. It enables extracting latent representations of audio and text for downstream tasks such as text-to-audio retrieval and zero-shot audio classification, and is also integrated into HuggingFace Transformers.

## Use cases
- find audio clips matching a text description
- compute embeddings for audio files and text queries
- zero-shot classify sounds or music by label prompts
- build a text-to-audio search engine
- extract audio features for a downstream model
- retrieve music by natural language queries

## When to choose
- you need joint audio-text embeddings for retrieval or classification
- you want zero-shot audio classification without training a custom classifier
- you want a CLIP-style model for the audio modality with pretrained checkpoints

## When to avoid
- you need image or video understanding rather than audio
- you need speech-to-text transcription rather than semantic audio representation
- you cannot run GPU inference or handle large model checkpoints

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, audio-processing, nlp, search-engine
- domain: machine-learning, deep-learning
- platform: python, cross-platform
- tags: contrastive-learning, multimodal, audio-text-retrieval, zero-shot-classification, laion, pytorch, pretrained-models, embedding, audio, natural-language-processing, gpu

## Member repositories
- LAION-AI/CLAP (main) score 42

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:31.422389+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 2026-08-30T02:43:41.380809+00:00, confidence not recorded.
  - readme: https://github.com/LAION-AI/CLAP (fetched 2026-08-28T04:06:31.422389+00:00, sha 032496f2515f)
  - homepage: https://arxiv.org/abs/2211.06687 (fetched 2026-08-29T10:23:38.527725+00:00, sha 5eff35ee2928)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T10:23:38.530693+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T10:23:38.534266+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T10:23:38.536078+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T10:23:38.532518+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.