[Frotag]

I've tried self-hosting with navidrome [0] / plex / jellyfin but the thing I miss most is music discovery via radios / discover weekly. I've tried replicating it a few times with embedding vectors + vector search but at best it finds songs in the (sub)-genre with the tempo / mood being pretty different.

Maybe I just need better data, been meaning to try again when that spotify crawl by annas-archive gets released. I've just been using musicbrainz [1] and youtube. Model-wise I've tried off-the-shelf ones like [2] and [3] and training auto-encoders like VAEs / MAEs [4]

[0] - https://www.navidrome.org/

[1] - https://musicbrainz.org/

[2] - https://github.com/LAION-AI/CLAP

[3] - https://github.com/SonyCSLParis/music2latent

[4] - https://arxiv.org/pdf/2207.06405

[hyphmngo]

i've found the essentia models to be pretty good for vector similarity. much better results than CLAP https://essentia.upf.edu/models.html#discogs-effnet

have a service running here if you're into electronic music and want to poke around https://cosine.club/

[urbandw311er]

With the recent data leak of Spotify‘s entire playlist database, you might be able to build something considerably better for music discovery now!

[josh-sematic]

This is the main factor that makes me leery to try going the full self host route.