Ross sent me this link: “SGER: A Music Search Engine Based on Aesthetic Similarity”
It’s a grant the National Science Foundation recently funded. From the abstract:
This SGER project aims to develop a prototype music search engine based on identifying aesthetic similarities. This engine will utilize power-law metrics to extract statistical proportions of music-theoretic and other attributes of music pieces (e.g., Pitch, Duration, Pitch Distance, Duration Distance, Melodic Intervals, Harmonic Intervals, Melodic Bigrams, etc.).
The engine searches for pieces that are aesthetically similar to the input piece using a mean squared error (MSE) approach. Preliminary testing has been done using the Classical Music Archives corpus (14,695 MIDI pieces), combined with 500+ MIDI pieces from other styles (e.g. Jazz, Rock, Country, etc.). Similar metrics have already been validated on aesthetic attributes of textual materials. Text results (author attribution, style identification, and pleasantness prediction) indicated an high level of accuracy.
Sounds cool!
This sounds eerily similar to the Music Genome Project – available at Pandora.com.
Ah yes, MIDI – the music format that launched a million cheesy web sites.
Analyzing the musical DNA of Midi files is technically a lot simpler than audio. Audio will be a much tougher (and more worthwhile) nut to crack.
I had the exact same thought as Pete….this sounds very similar to what Pandora does.
Any idea how it’s different?
Relative to your article on music search, you might want to check out http://www.owlmusicsearch.com. We launched in October. We power audio search for http://search.creativecommons.org as well as direclty through the aforementioned direct url. We provide a music discovery experience based on music itself and a query language based on audio features and not text or keywords.
Your stupid.