The end goal is procedural music generation. That is a LONG way off however, you have to start somewhere.
This project has been rolling around in my head for a while now but only in theory.
I was provided an opportunity to put pen to paper for a final project for a 'signals and systems' class for my EE undergrad.
- Intakes .wav files
- Reads in and preforms laplace transformations to get a live frequency-amplitude plot
- collect frequency data on the song as it progresses.
- Taking a tool from 'Fabformat' (on my main page) generate parametric cubic splines from the songs.
- See how good that playback is.
- Mess with shifting the equations from #2 to see how that changes the song.
- Once its possible to read in and analyze music, Having structured patterns of data would allow an opportunity to work with Machine Learning. It would be cool but out of my wheelhouse right now.