Final project for Deep Listening (ARTS-4410) at RPI in Summer 2019
In this project, I wanted to use algorithmic composition to create classical piano music by chopping up existing pieces by famous composers. I did this with the help of the piano-midi.de dataset, a collection of hundreds of classical compositions by composers such as Mozart, Bach, Beethoven, and more, in MIDI format. I fetched all of these MIDI files then created a program that parses a user-defined subset of this data using python-midi and successively chooses random tracks to splice together chunks of 20-50 notes. To keep things flowing organically, tracks are spliced together only at places where the pitches match up.