Very quickly I learned that there was a reason AI couldn't actually listen to music. It was because the technology wasn't there yet. While progress had been made in terms of Music Generation (i.e Suno), listening to music was a whole different beast.
My initial hope was that an AI could give an opinion on all the different versions of Chopin Op. 25 No. 6 and why Josef Lhevinne's is the best, or give me Violinist recommendations if I enjoyed the ferocity of Ivry Gitlis. I wanted an AI to understand the lyricity of Peter Schreier, or the caprice of Maurice Andre.
But that is not quite possible.
I quickly adjusted my scope from an AI appreciating music, to an AI being able to understand the notes and rhythms. This is how I learned of the concept of Music Information Retrieval.
Music information retrieval (MIR) is the interdisciplinary science of retrieving information from music. Those involved in MIR may have a background in academic musicology, psychoacoustics, psychology, signal processing, informatics, machine learning, optical music recognition, computational intelligence, or some combination of these. -- from Wikipedia.
That is also how I came up with the name Mirgenta, which is a portmanteau of MIR and Agent (not exactly, but Mirgenta sounds cooler than Miragent). Also, the Magenta color scheme is an ode to its near-homophone. Coincidentally (or maybe not by accident), Google Deepmind works on a MIR product called Magenta. https://magenta.withgoogle.com/