At the intersection of art and science, a new application developed at the College of Charleston could be music to the ears of many iPhone users.
College of Charleston Computer Science professor Bill Manaris and several of his students have developed a new hybrid internet radio / music discovery platform called Armonique (http://www.armonique.org).
Unlike other music discovery systems such as Pandora, Armonique identifies aesthetic similarities from over 250 statistical patterns in the music. These patterns are based on Pythagorean principles of harmonic proportion.
At the core of their research is a complex piece of computing wizardry called an Artificial Neural Network (ANN). Computer programs sort through a database of music – like your iTunes library – and extract measurable values from each song, things like pitch, duration, melodic and harmonic intervals, etc. The ANN then takes that information and plugs it into various mathematical models that help it classify songs based on aesthetic similarity.
Other platforms use humans to tag songs, a less effective way of determining song similarity.
With this information, Armonique is able to develop a music playlist that is rated highly in experiments by listeners using this new platform.
With a library of thousands of songs, Armonique is the prototype of future music discovery engines.
And now there is an app for that.
A group of Manaris’ students, incuding J.R. Armstrong, Thomas Zalonis, and Perry Spyropoulos, have just developed an application of Armonique that can be downloaded for free on the Apple iPhone platform. This will allow iPhone owners to enjoy a constant stream of music generated to be appealing to each individual listener.
This research was funded by the National Science Foundation (NSF). Music has been provided by Magnatune and the Classical Music Archives.
More on Armonique can be found at http://www.youtube.com/watch?v=EDbvm2MY5RI
