See: http://www.leedsdatamill.org/blog/oompah-py-big-brass-meets-big-data
Code and data needed to generate the score and parts for an approximately 10 minute composition for brass band representing foot-fall data collected in the city of Leeds.
The piece is called "The Rhythm of the City" and is scored for a traditional
British brass band with one percussionist on tubular bells. The code used to
generate it is called oompah.py and is a Python script initially hacked
together over the course of a LeedsDataMill sponsored hackathon weekend (and
subsequently finished as an exercise in musical exploration).
Each section of the brass band represents a location in the city of Leeds. The piece is split into 24 8 bar phrases (encompassing each hour of the day). As each location's footfall data changes - becoming quiter or busier - so the music played by the associated section within the band reflects this change.
The piece begins at midnight in a quiet mood but as the music progresses (punctuated by bells sounding the hours) the different parts of the city wake up to a climax during the working day before returning to a placid state of affairs as the evening closes the musical proceedings.
The software reads the raw footfall data and turns it into the musical score for the piece.
- Python 2.7+
- Lilypond 2.14.2+
The make command is your best bet:
make
Will generate a lilypond and PDF for the score and each part along with a
score.midi file so you get a sense of what it'll sound like:
make clean
Cleans up all the parts and Lilypond related junk.
Check out the instructions in the presentation directory for displaying
the presentation both authors gave on this project at PyconUK 2014.
All feedback is most welcome!