“The Way It Sounds”: Timbre Models for Analysis and Retrieval of Music Signals

By Jean-Julien Aucouturier, François Pachet, and Mark Sandler.

IEEE Transactions on Multimedia 7(6), 1028–1035 (December 2005). DOI: 10.1109/TMM.2005.858380.

Read the paper (PDF) · Publisher version (DOI)

About the paper

The paper models the timbre of entire recordings, where multiple instruments and voices can overlap. It uses statistical models of short audio frames to compare songs by their overall sound, find sections with different textures, and identify recurring patterns.

Abstract

Electronic Music Distribution is in need of robust and automatically extracted music descriptors. An important attribute of a piece of polyphonic music is what is commonly referred to as “the way it sounds”. While there has been a large quantity of research done to model the timbre of individual instruments, little work has been done to analyze “real world” timbre mixtures such as the ones found in popular music. In this paper, we present our research about such “polyphonic timbres”. We describe an effective way to model the textures found in a given music signal, and show that such timbre models provide new solutions to many issues traditionally encountered in music signal processing and music information retrieval. Notably, we describe their applications for music similarity, segmentation and pattern induction.

Related work

Music Similarity Measures: What’s the Use? introduces a timbre similarity measure for music browsing. Improving Timbre Similarity: How High’s the Sky? examines how far that family of measures can be improved.

Hubness in Music Similarity traces the research on songs that appear unusually often as nearest neighbors in timbre-based music retrieval.

A Scale-Free Distribution of False Positives for a Large Class of Audio Similarity Measures (2008) examines how often the same songs turn up as false matches across timbre similarity models.

Back to Publications