I don’t know if this is the correct forum for this question because I don’t know all the limitations of the VST architecture but I figured this was the best place to ask.
I’ve been working on a machine-learning quantize process specifically for drums that currently works directly with MIDI files. Since part of the analysis it does require analyzing long sections and making adjustments based on both past and future events, it doesn’t seem like a good candidate for a VST plug-in that is processing data in real time.
Does anyone know of a way to do this sort of thing without saving to a .MID file, analyzing and quantizing, and loading it back in? Like a truly ‘custom’ quantizer?