A short dive into signal processing and machine learning for music
I spent a week at Stanford studying the landscape of Music Information Retrieval. I think I was the only person in the room without a MS or a PhD on the way - I was honored to attend.
The week was capped off by a presentation from Douglas Eck, who brought a great presentation about how to think of machine learning in the arts and some cutting edge research going on in the Magenta project at Google.
During the week, I built an open source script that uses a deep learning-powered chromagram detection algorithm to convert an audio file into a MIDI file. More detail on Github.