Hi there! I hadn’t even thought of it that deeply. I was just pondering semi-generative microsound/lowercase/glitch/granular music - there is a software called “Forester 2.0” that is controllable, but it’s based on the Max/MSP framework and requires that to run.
I don’t make video game music (anymore), so Wwise is not what I’m looking to use. Just kinda wondered what would happen if the size of the file could be reduced by removing the sound bank and opening it up to user material.