-
#AutomaticMixing
Inspiration
Curiosity for what kind of wacky shit we could do with spacial sound immersion through L-isa
What it does
Given a track input, we extract specific audio features (e.g. onset, root-mean-square, spectral centroid, among others), and using a set of rules, use this info to create a L-isa configuration specifically tailored to the input track.
How we built it
First we composed a track that we wanted to use, then use a series of Python scripts to extract the audio features. We then write the audio features as a wav file that we send to Max along with the original track, which then gets put through a set of rules to send a OSC message to the L-isa control system, and thus the processor.
Challenges we ran into
Finding a Python library to extract the features that worked on our systems, giving up on that and then setting up a Docker build for a development environment, communicating between Python and Max, trying to remember rust programming knowledge.
Accomplishments that we're proud of
We set out with a goal and accomplished it fully.
What we learned
How to extract audio features for a track and how useful they can be, things how cool L-isa/immersive soundscape design is.
What's next for Dy\nSpac
Probably nothing idk.
Log in or sign up for Devpost to join the conversation.