Discussion on Music Visualization Technology and its Capability to Visualize Large Ensembles

This project serves as an exploration of the capabilities of music visualization, specifically as it pertains to limitations on source separation. Utilizing several different source separation and visualization techniques, I draw upon proof-of-concept visualization software I have constructed to inform a design structure that could be applied to large ensembles. Functionally, the two programs (one for static audio files and one for live audio) are very similar. Both take audio of one or more voices and construct a visual representation, but their implementations differ drastically. For static audio files, a .wav file is uploaded, the user specifies how many voices are in the piece, and the program constructs a visualization. For live audio, the user connects audio from each voice to a computer running Ableton Live with the program installed as a Max/MSP patch, feeding each input to the patch and producing the visualization