Content-based Music Indexing and Organization

While electronic music archives are gaining popularity, access to and <br> navigation within these archives is usually limited to text-based queries <br> or manually predefined genre category browsing.<br> We present a system that automatically organizes a music collection <br> according to the perceived sound similarity resembling genres or styles of <br> music.<br> Audio signals are processed according to psychoacoustic models to obtain a <br> time-invariant representation of its characteristics.<br> Subsequent clustering provides an intuitive interface where similar pieces <br> of music are grouped together on a map display.<br> <br> Category: H.5.5 - Information systems - Sound and Music Computing<br> Category: H.3.1 - Information Systems - Information Storage and Retrieval<br> [content analysis and indexing]<br> <br> Keywords: Music Retrieval, Genre, Rhythm, Psychoacoustic Models, <br> Clustering, Self-Organizing Map, Neural Networks<br> <br>