Greetings Lemmings!
I am not new to self hosting; been at it for a few years. And I have neglected a very important part of my home lab; MUSIC!
So I stream music from Navidrome to all of my devices.
My music library has grown organically over the years, for probably the past 17 years. And a few times I had manually organized and fixed up some of the tags. But ultimately I ended up with a mess of a library.
I am working on cleaning that up. Though I absolutely should have cleaned it up before I created 2 backup scripts.
In short, I have a script that is called from a systemd service on a timer that runs my backup script that essentially more or less uses rsync to mirror the files in LiveMusicDir to MusicArchive1. This happens on my docker host where my Navidrome lives. The Music is on an NFS share hosted outside of the docker host.
The next step I run another similar setup on my desktop; systemd service running a script on a timer that uses rsync and other dependencies to track changes.
What I plan to do is organize the source. I am using beets, and learning as I go.
What methods do you use for managing a large music library?
I luckily do, I have a 3090 with 24GB of vram that copper modded. Hermes you say, may have to investigate.
I have 3 copies of my library
Yeah, it’s all the rage now. Qwen3.6 35B or Google Gemma 26B should work fine for the task. Llmfan on hugging face uses the heretic framework to “abliterate” them and remove any safe guards that might prevent working with “pirated” content.
You can run hermes in a container or vm if you’re worried about the ai hallucinating, though I haven’t seen that happen. Use as high of a Q quant value as you can and run llama.cpp for speed. Or just try a free cloud model with hermes and see if it works.
The agent installed a bunch of mp3 scanning tools, did an inventory of my library and generated a list of actions for me to approve before it ran. Feels like the future.
If I run Hermes, or anything AI, I’ll want to investigate how to containerize it. Which may be difficult with one GPU.
Podman/docker works fine and doesn’t need the entire gpu like a vm. Hermes has a supported method for just that.