We worked with Olivia Erlanger to create a data sonification piece from housing foreclosure data in the Detroit area. Our web genius Mike Sperone wrote a php "web scraper" for us that would pull in the most recent 24 foreclosed on properties from zillow.com, the data of which consisted of the address, price, longitude/latitude along with some other descriptors.
Olivia was interested in how the data might be used to "chop and screw" songs which were popular during pivotal years leading to the financial crisis in 2008.
The first challenge was to reverse engineer a chopped and screwed track - define key elements in the music, and come up with ways to use data from the foreclosures to parametrically chop and skrew audio clips in real time. We looked to "The Originator" himself, DJ Screw for inspiration.
One obvious signature of the genre is the slowed down tempo. We achieve this in our tracks by mapping the price of the foreclosed properties to the playback speed of the track. Lower prices generate slower playback speeds for the songs and vice versa. To create the stuttering effects we used computer musician Miller Puckette's transient detection algorithm (bonk~) to tell us whether we are on beat or not. When we are on beat, we ask the program to decide whether to continue playing the song or to go back and repeat the measure. How likely we are to go back and repeat a measure is determined by the location of the property in Detroit using it's longitudinal and latitudinal value.
The piece runs in a continuous loop, generating new music from continuously updated information on Zillow. It is currently on view at What Pipeline in Detroit. Embedded below is a 42 minute excerpt of the piece.