AudioMuse-AI

Self-hosted music analysis and playlist-generation service for discovering sonically related tracks in supported music servers.

Deployment layout

Compose: /srv/docker/audiomuse/docker-compose.yml
Static configuration/source: no additional static configuration or build source
Non-secret environment: service roles, timezone, PostgreSQL topology, Redis URL, and temporary-audio paths are declared inline in Compose; there is no project-local .env
Secrets: /srv/docker/secrets/audiomuse/.env
NVMe application state: /srv/appdata/audiomuse/postgres, /srv/appdata/audiomuse/redis, /srv/appdata/audiomuse/temp-audio-flask, /srv/appdata/audiomuse/temp-audio-worker, /srv/appdata/audiomuse/plugins-flask, and /srv/appdata/audiomuse/plugins-worker
Bulk HDD data: none
Other mounts: none

Secrets are never committed to this repository. The central secret environment is supplied to the PostgreSQL, Flask application, and worker services.

Networks/dependencies: the Flask application and worker depend on PostgreSQL and Redis through the private audiomuse_internal network; both also use the external npm_proxy network at 192.168.97.25 and 192.168.97.26
Back up: /srv/appdata/audiomuse/postgres, /srv/appdata/audiomuse/plugins-flask, /srv/appdata/audiomuse/plugins-worker, and the separately protected AudioMuse-AI secret environment
Re-creatable: the containers, Redis queue state, and temporary-audio directories