Very nice. Is the function that determines whether a chord inversion is likely to be correct based on any sort of well-known algorithm? I don't have the musical fluency to comprehend what the intervals all correspond to. Seems like a neat solution, though.
Totally original. I came up with it by looking at lots of examples and seeing patterns emerge. It's not perfect, but it turns out that checking a select few indicators is sufficient in the vast majority of cases.
I am usually fiddling with or listening to livecoded music, which evolves in time from silence to (sometimes) pleasing sounds, so it was refreshing to hear more consonant and familiar sounds. I thoroughly enjoy playing with this, but I would have to learn more music theory to really 'get it'. Amazing!
Sounds a lot like a high schooler just getting into improvisation. There's little art going on here. What about phrases? Motivic development? There's more to improvisation than just scales and rhythm.
And I say that was a composition major.
How do you generate your rhythms? Is there structure or just randomness?
Configurable randomness. A particular beat has probabilities of containing eighth notes, triplets, or sixteenth notes based on the "Rhythmic Variety" setting, and those notes have a probability of being rests based on the "Rests" setting.