💛 Donate to help keep AudioMuse-AI growing. Your support covers AI subscription for development (~€220/month), homelab electricity for testing (~€20/month), and makes the project sustainable.
Release Date: August 18th, 2026
AudioMuse AI v3.3.1 introduces several improvements, including faster audio analysis, dependency updates, and various bug fixes.
The analysis speedup comes from improving how audio is loaded: instead of loading it twice, audio is now loaded once and then resampled. This should save some seconds per analyzed song, depending on the hardware and song length.
The dependency updates not only improve stability and maintainability, but also introduce a newer version of cuML, extending the clustering algorithms available on GPU with support for Spectral Clustering.
Several other bug fixes and improvements were included, covering parameters shown in the Setup Manager, cron job retry logic, Music Map path handling, and CPU quota support for deployments and Docker Compose.
Important
The -nvidia image has been updated from CUDA 12.x to CUDA 13.3.1. Make sure your host has an up-to-date NVIDIA driver to use the new image properly.
What's Changed
- Fixture - #852 #785 #787 by @NeptuneHub in #858
- Analysis improvement by @NeptuneHub in #861
- Dependencies version update by @NeptuneHub in #862
- Setup wizard parameter grouping by @NeptuneHub in #864
Full Changelog: v3.3.0...v3.3.1
