Searching and organizing growing digital music collections requires a computational model of music similarity. This paper describes a system for performing flexible music similarity queries using SVM active learning. We evaluated the success of our system by classifying 1210 pop songs according to
mood and
style (from an online music guide) and by the performing artist. In comparing a number of representations for songs, we found the statistics of mel-frequency cepstral coefficients to perform best in precision-at-20 comparisons. We also show that by choosing training examples intelligently, active learning requires half as many labeled examples to achieve the same accuracy as a standard scheme.
Keywords Support vector machines - Active learning - Music classification
Michael Mandel is a PhD candidate at Columbia University. He received his BS degree in Computer Science from the Massachusetts Institute of Technology in 2004 and his MS from Columbia University in Electrical Engineering in 2006. In addition to music recommendation and music similarity, he is interested in computational models of sound and hearing and machine learning.
Graham Poliner received his BS degree in Electrical Engineering from the Georgia Institute of Technology in 2002 and his MS degree in Electrical Engineering from Columbia University in 2004 where he is currently a PhD candidate. His research interests include the application of signal processing and machine learning techniques toward music information retrieval.
Daniel Ellis is an associate professor in the Electrical Engineering Department at Columbia University in the City of New York. His Laboratory for Recognition and Organization of Speech and Audio (LabROSA) is concerned with all aspects of extracting high-level information from audio, including speech recognition, music description, and environmental sound processing. Ellis has a PhD in Electrical Engineering from MIT, where he was a research assistant at the Media Lab, and he spent several years as a research scientist at the International Computer Science Institute in Berkeley, CA. He also runs the AUDITORY email list of 1700 worldwide researchers in perception and cognition of sound.