free-music-demixer VS lima

Hi HN,

Over the Christmas break I added multi-threading to the WASM Demucs module in freemusicdemixer

Demucs (v4 hybrid transformer) is a much higher quality model than the previous default, but it ran very slowly when limited to one worker: ~17 minutes for an average 4-minute song

I have since implemented multi-threading with WebWorkers.

If you raise the "MAX MEMORY" setting to 16 GB or 32 GB, your track will demix within 7-5 minutes, producing state-of-the-art results.

There is also support for the Demucs 6-source model which adds piano and guitar stems.

Please reach out and be loud about any bugs or UX issues you encounter!: https://github.com/sevagh/free-music-demixer/issues

* Post-processing step (bigger impact)

I tried to tackle the post-processing step in my C++ code (which would win ~1 dB in quality across all targets) but it's too tricky for now [2]. Maybe some other day.

1: https://github.com/sevagh/free-music-demixer/blob/main/examp...

2: https://github.com/sigsep/open-unmix-pytorch/blob/master/ope...

You can run a virtual machine (e.g. lima) from inside a nix-shell, exactly as you would do with a regular shell.

for others similarly curious, here's an example of the thing: https://github.com/noop-inc/template-java-spring-boot/blob/m...

they seem to be using the excellent lima <https://github.com/lima-vm/lima#readme> for booting on macOS; I run colima for its containerd and k8s support but strongly recommend both projects $(brew install lima colima)

Lima is what I use as well. It's quick and easy to just fire up a VM with default settings, but also very easy to configure with different file sharing options, port forwarding, different linux distributions, etc. (their examples are also pretty good IMO [1]).

In particular I use it to run an amd64 VM, which I need to run a stubborn service for work that doesn't run on arm CPUs.

[1] https://github.com/lima-vm/lima/tree/master/examples

Lima (1) is a project that packages Linux distros for MacOS and executes them via qemu in the backend. Maybe you could solve your problem by launching one of their vms and inspecting the command line it generates. You might find an option you were missing.

(1) https://github.com/lima-vm/lima

If you wish to delve into all the configuration possibilities for Lima VM, you can visit this resource.

I'd say Lima and Colima should be enough for most use cases:

https://lima-vm.io/

https://github.com/abiosoft/colima

Github: https://github.com/lima-vm/lima

Lima wraps QEMU in a simple CLI, with neat features for container users, such as filesystem sharing and automatic localhost port forwarding, as well as DNS and proxy propagation for enterprise networks. Rancher Desktop wraps Lima with k3s integration and GUI.

Talks: https://github.com/lima-vm/lima/blob/master/docs/talks.md