AudioMuse AI v1.0.1: Flask bugfix

Release Date: April 17, 2026

AudioMuse AI v1.0.1 introduce bugfix to the flask container to revert the high ram load (gunicorn 2 workers was loading all the index twice).

What's Changed

• Flask bugfix by @NeptuneHub in #430

Full Changelog: v1.0.0...v1.0.1

AudioMuse AI v1.0.0: Setup Wizard

Release Date: April 17, 2026

AudioMuse AI v1.0.0 introduces a new Setup Wizard UI that simplifies installation and configuration by moving Application Settings into the web interface instead of environment variables.

From this version onward, the only required environment variables are only the infrastructural one related to Redis and PostgreSQL connections. All other parameters are configured during the first startup directly through the AudioMuse web page, significantly reducing setup complexity and minimizing configuration errors.

The Setup Wizard is not limited to the initial setup and can be accessed at any time from the administration menu to update settings on the fly. Each change is automatically applied by automatically restarting the Flask and Worker containers, ensuring updates take effect immediately without manual intervention.

IMPORTANT: For legacy systems, existing environment variables are automatically imported into the database during the first startup to preserve current configurations. After this migration, environment variables are no longer used except for Redis and PostgreSQL, which remain exclusively managed through env vars.

What's Changed

• • Authentication improvement + RQ fix by @NeptuneHub in #400
• fix(ui): dark-mode restore warning and scrollable sidebar by @rendyhd in #411
• Setup Wizard UI by @NeptuneHub in #413
• Remove duplicate lines by @ballfather in #416
• refactor: mini fixes by @ballfather in #412
• fix(docker): add .gitattributes to enforce LF for container files by @rendyhd in #421
• Docker: Proper (and faster) shutdown sequence by @ballfather in #414
• Documentation update with Wizard UI instruction by @NeptuneHub in #420

New Contributors

• @ballfather made their first contribution in #416

Full Changelog: v0.9.6...v1.0.0

AudioMuse AI v0.9.6: Security update - AUTH_ENABLED true by default

Release Date: April 6, 2026

AudioMuse AI v0.9.6 is a security update that set the env var AUTH_ENABLED to true by default. This work with this additional env var:

1. AUDIOMUSE_USER - mandatory
2. AUDIOMUSE_PASSWORD - mandaotry
3. API_TOKEN - optional, only if you use an external plugin
4. JWT_SECRET - optional

you can read more here: AUTH DOCS

If you don't want to use it (not reccommanded) you can just set AUTH_ENABLED to false.

To help this transition the .env.example and all the deployments example are updated to support this additioanl env var.

What's Changed

• Auth enabled by default by @NeptuneHub in #399

Full Changelog: v0.9.5...v0.9.6

AudioMuse AI v0.9.5: Mood improvement and backup functionality

Release Date: April 5, 2026

AudioMuse AI v0.9.5 enhances how moods are used across the app. Mood can now be selected as an input for both Similar Song and Song Alchemy, making recommendations more flexible and expressive.

A major addition in this release is the new Backup & Restore functionality, now available in the Administration menu. This sits alongside the existing Cleaning and Scheduled Tasks tools, giving you better control over system maintenance and data management.

What's Changed

• Similar song from mood by @NeptuneHub in #395
• Brust graph for mood by @NeptuneHub in #396
• Backup and restore by @NeptuneHub in #397
• mood selector improvement by @NeptuneHub in #398

Full Changelog: v0.9.4...v0.9.5

AudioMuse AI v0.9.4: Path to Happiness

Release Date: April 4, 2026

With AudioMuse-AI v0.9.4, we take a step further in turning your music library into a space you can actively explore rather than simply browse.

The Song Path feature has been expanded to move beyond connections between songs, now allowing you to navigate through moods such as Happy, Sad, Relaxed, Aggressive, and Danceable. This opens the possibility to shape personal journeys across your collection, tracing emotional paths and discovering new relationships within your music.

At the same time, Song Alchemy becomes more persistent and meaningful. You can now save its results as anchors, which can later be reused either as starting points for new Song Alchemy or as input for your Song Path functionality. This adds continuity to your exploration and gives you more control over how you move through your library.

Together, these improvements reinforce the vision of Music Map: your library as a world, and AudioMuse-AI as the set of instruments that allows you to navigate it in new and expressive ways.

This release also includes a minor update to Navidrome playlist creation, which are now set as public by default, making them visible across users and better suited for shared environments.

What's Changed

• Revise AI model testing information in PARAMETERS.md by @rendyhd in #376
• Devel -> Main : create public playlist on navidrome - #377 by @NeptuneHub in #378
• Song Path - Added mood by @NeptuneHub in #387
• Alchemy and song path Anchor by @NeptuneHub in #389
• Avoid multiple batch task in parallel - #391 by @NeptuneHub in #393

Full Changelog: v0.9.3...v0.9.4

AudioMuse AI v0.9.3:

Release Date: March 16, 2026

AudioMuse-AI v0.9.3 fixes an issue where the search functionality could fail in some edge cases. It also improves the Instant Playlist feature by enhancing the prompt with additional metadata fields such as year, album artist, and, when supported from the music server, rating.

What's Changed

• Add album_artist, year, rating, file_path column to data layer across all providers by @rendyhd in #306
• Fix Search over null field by @NeptuneHub in #373

Full Changelog: v0.9.2...v0.9.3

AudioMuse AI v0.9.2: minor fix

Release Date: March 13, 2026

AudioMuse-AI v0.9.2 introduce some minor fix

What's Changed

• Devel -> Main by @NeptuneHub in #369

Full Changelog: v0.9.1...v0.9.2

AudioMuse AI v0.9.1: fix noavx2 image

Release Date: March 13, 2026

AudioMuse-AI v0.9.1 fix the noavx2 image bug #368

Full Changelog: v0.9.0...v0.9.1

AudioMuse AI v0.9.0: Distilled CLAP (DCLAP)

Release Date: March 12, 2026

AudioMuse-AI v0.9.0 introduces the AudioMuse-AI DCLAP model, which replaces CLAP for text search functionality.
DCLAP is significantly faster and has been tested at about 5–6x faster on a Raspberry Pi 5 (8GB RAM + SSD), making semantic text search practical even on lower-power or older hardware.

As with the previous implementation, text-search embeddings can still be disabled by setting the env varCLAP_ENABLED to false.

⚠️ Important: this version requires a clean database and a new full analysis of your audio library.
Databases generated with previous versions are not compatible.

This release also replaces the MTG Essentia Musicnn model with the original Musicnn model.
The original model does not include the classification head (used for attributes such as danceability, happiness, sadness, etc.), as these are now computed using DCLAP embeddings. Additionally, the original Musicnn model is distributed under the ISC license, which is more permissive and better aligned with the goals of this project.

What's Changed

• Musicnn model by @NeptuneHub in #357
• feat: replace Flask dev server with Gunicorn WSGI for production by @12somyasahu in #356
• app.py resolving merge conflict by @NeptuneHub in #358
• Improve search behavior on "playlist from similar song" feature by @sfredo in #335
• DCLAP by @NeptuneHub in #359

New Contributors

• @12somyasahu made their first contribution in #356
• @sfredo made their first contribution in #335

Full Changelog: v0.8.14...v0.9.0

AudioMuse AI v0.8.14: Fix latest tag

Release Date: March 8, 2026

AudioMuse-AI v0.8.14 release is needed just to fix the latest tag that for error was moved on the model release.

Full Changelog: v0.8.13...v0.8.14