AudioMuse-AI is an open-source, self-hosted tool that analyzes music libraries using sonic analysis to rediscover songs and generate playlists, with multi-server support and various discovery features. It supports Docker Compose, native platforms, and Kubernetes deployment, and integrates with several self-hosted music servers.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
AudioMuse-AI is an opensource and self-hosted tool that uses sonic analysis to rediscover forgotten songs in your music library and generate groove-aware playlists that also capture the meaning behind each track, without relying on metadata or external APIs.
You can run it locally with Docker Compose or Podman, deploy it at scale in a Kubernetes cluster (AMD64 and ARM64 supported), or use native applications available for macOS, Windows, and Linux. It integrates with major self-hosted music servers including Navidrome, Jellyfin, LMS, Lyrion, Emby, and Plex, with more integrations planned.
Prefer not to self-host? Elestio offers AudioMuse-AI as a managed cloud service.
AudioMuse-AI features include:
- Multiple Music Servers (from v3.0.0): connect several media servers to a single AudioMuse-AI deployment; built-in duplicate detection ensures each track is analyzed only once.
- Clustering: automatically groups sonically similar songs to create playlists based on sound.
- Instant Playlists: generate a playlist by specifying criteria like "high-tempo, low-energy music".
- Music Map: visual 2D genre-based map of the collection.
- Playlist from Similar Songs: create a discovery playlist from a track’s sonic signature.
- Song Paths: bridge two songs with intermediate recommendations.
- Sonic Fingerprint: playlists based on listening habits.
- Song Alchemy: mix vibe, mark tracks as ADD or SUBTRACT to create a curated playlist and export to a server.
- Text Search: search by mood, instruments, and genre.
- Lyrics Search: search by theme or meaning; Lyrics language support covers 72 languages.
The project documents its architecture, algorithm description, deployment strategies, and configuration parameters in the docs folder.
Getting started
AudioMuse-AI can be deployed via Docker Compose for containerized deployment. Prerequisites include Docker and Docker Compose, and a running media server (Navidrome, Jellyfin, Lyrion, Emby, or Plex). Steps in the Quick Start Deployment (Containerized) include:
- Create your environment file:
cp deployment/.env.example deployment/.envand customizedeployment/.envas needed. - Start the services:
docker compose -f deployment/docker-compose.yaml up -d. - Access the application at
http://localhost:8000and the API docs athttp://localhost:8000/apidocs/. - Run your first analysis via the web UI in the Analysis and Clustering page.
- Stop the services:
docker compose -f deployment/docker-compose.yaml down.
AudioMuse-AI is designed to work with PostgreSQL v15 in the deployment example.
Native deployment is available for macOS, Linux, and Windows, with per-platform release bundles that include the embedded stack.
Recent releases
Latest releases include:
- v3.1.1 AudioMuse AI v3.1.1: Maintenance Release (2026-08-01)
- v3.1.0 AudioMuse AI v3.1.0: Plugin System Improvement (2026-07-30)
- v3.0.6 AudioMuse AI v3.0.6: maintenance update (2026-07-29)
- v3.0.5 AudioMuse AI v3.0.5: Bugfixing (2026-07-25)
- v3.0.4 AudioMuse AI v3.0.4: Migration Imrpovement (2026-07-24)
Traction
Stars: 2379
Behind the repo
The project lists multiple related repositories such as AudioMuse-AI Helm Chart, Navidrome Jellyfin plugin, Lyrion plugin, and a MusicServer variant, indicating an ecosystem around the core tool. The repository is licensed under AGPL-3.0 and created on 2025-05-24.
Caveats
License: AGPL-3.0 Created: 2025-05-24 Last push: 2026-08-04 Open issues: 29 Languages: Python