scikit-clean VS airsonos

Among all these feel-good stories, how about one with a bit different ending?

During my masters, I created a ML library that dealt with noise in dataset. I implemented bunch of papers, but unlike your usual research code, I spent a long time obsessing about it's API, performance, created documentation, CI- the whole shebang [1]. But then, like avg research code, I moved on and promptly forgot about it.

One day about a year ago the cofounder of a very new, small startup working on something similar texted me about the project on linkedin. We chatted for a bit, but as a guy who thinks he's too cool for linkedin, I next logged in and saw his last message about wanting to collaborate about 3/4 months after the fact.

Well they raised $25 million dollars a few months ago :(

[1] https://github.com/Shihab-Shahriar/scikit-clean

The repository https://github.com/stephen/airsonos has the code and is surprisingly accessible.

If it's just Airplay for music, you can try Airsonos: https://github.com/stephen/airsonos

This one is pretty simple. All the communication with the sonos device happens in the clear and the protocols are actually pretty well documented.

Getting good at using wireshark is a good place to start.

This project is pretty dead but I remember using it a few years ago https://github.com/stephen/airsonos

IIUC I need to run https://github.com/stephen/airsonos in order to get my Sonos system to be an option on Airplay https://s.natalian.org/2021-10-02/music.png option?

You could try running AirSonos https://github.com/stephen/airsonos to stream audio via AirPlay.