you know all this neural net nonsense? you can do it with audio too apparently...

  On 8/9/2015 at 12:41 PM, webby said:

Given that streaming services can automatically recognise copywrited audio if you upload it, why is it that we don't have a reverse sound search equivalent of reverse image search?

Shazam?

https://forum.watmm.com/topic/94451-aphex-twin-articletech-interview-in-latest-magpi-magazine-raspberry-pi-midimutant/

https://codepen.io/teropa/full/JLjXGK/

https://soniccharge.com/patternarium

Totally missed that Midimutant thing because I only read the EKT forum..

I have thought about building something like this myself, but I always think mapping all the synth params to code is like a hundred times more work than to actually make the patch evolver thing because for that you can just hook up any old genetic algorithm.