Not entirely automatic, but I found that Navidrome (at least when you use a third party client like Feishen or Symfonium) has smart playlists. I use that to write some simple queries like:
By genre:
By listening habits:
and so on.
Obviously, the genre smart playlists assume your music has Genres tags. I use Picard (with the Last.fm genre plugin) for that.
Ripping a tag from a pillow that says “Under Penalty of Law: Not to be Removed By Anyone but the Consumer”
Laughs in Navidrome
Take this down nephew
git clone this before it gets taken down
I was wondering why these types of open source projects always push to Github, despite the latter always complying with DMCA. (I get that Github provides discoverabilty features, but it just isn’t worth it to have all your work taken down).
On a similar note, has anyone tried out https://radicle.xyz/? It’s supposed to actually make use of git’s peer to peer nature (and not the client server model that everyone adopts with git) and ideally provide discoverability features.
The said I’ve only read the faq and haven’t actually tried it myself. Basically I’m wondering if it’s worth doing a deep dive on this technology
MusicBrainz Picard
It’s a GUI tool that gives you the ability to tag songs in bulk. It uses the MusicBrainz database to provide relevant metadata, and also provides user plugins (like the Last.fm plugin I mentioned)
I get my songs in batches (through Bandcamp and youtube-dlp) every couple of months, so GUI is fine for me.
However, I’m seeing a CLI alternative called https://beets.io/ ? Apparently it is a CLI tool that also queries the MusicBrainz database (or Beatport, or Discogs), and apparently integrates well with Lidarr. I never tried it, however.