Showing posts with label AI. Show all posts
Showing posts with label AI. Show all posts

Sunday, August 31, 2025

From the treasure chest: Python scripts for specific use cases

 

         

Here are some interesting scripts that aim to help with organizing a music library and specific use cases. Not all may be plex specific, the scripts just caught my eye. Caveat: I haven't tried them out myself unless noted. If you try them, please leave a comment to help others. I plan to update this list from time to time.

Chromatix: Plexamp is an excellent mobile app which can run on desktops - but is not good at that. Chromatix is my favorite when I listen to a specific album. It has a nice UI with a lot of features - not the DJ's and the other stuff dependent on sonic analysis though. In use.

Plex-Music-Organizer: Exporting Plex music metadata and update tags from CSV within the database

Audiograde: Picks a random album, plays it to help rating its songs.

There are new projects to create playlists with the help of AI being released on github every day it seems. These are some interesting ones:


Plex-mcp-server: transforms libraries into AI-queryable databases.

Plexmix: AI generated playlists based on the library in plex. Generates mood, environment and instrument embeddings to filter by. 

plex-open-playlister: generates intelligent playlists using locally installed LLMs (Ollama). Mix of semantic, signal-based and traditional approaches (lastfm queries).

AudioMuse AI: Feature-rich sonic analysis and AI-powered clustering to create smart, tempo, mood and environment based playlists within Jellyfin and Navidrome. A plugin for Musicbrainz Picard fetches tempo and key inter alia.  No plugin for plex yet.




Friday, January 31, 2025

Daylist - Playlists

     


Plex and plexamp already have a number of options for automatically creating playlists: Radios or sound journeys created based on certain criteria. 

The Meloday script, which can be found here, takes a time of day or mood-based approach. Meloday creates playlists for the respective time of day, be it morning, afternoon or late at night. It does this by using tracks that have already been played at those times, and adding tracks with sonic similarities to this framework. 

It even uses artificial intelligence, but not to generate the playlist itself, just to reformulate the names and descriptions of the new playlists.

There are many enthusiastic voices on reddit\/plexamp, even Elan Feingold, one of the founders of Plex, has commented positively - so it's worth giving it a try.