Feature-Driven Classification of Musical Styles

Authors: A. Buzzanca, G. Castellano, A.M. Fanelli

Abstract:

In this paper we address the problem of musical style classification, which has a number of applications like indexing in musical databases or automatic composition systems. Starting from MIDI files of real-world improvisations, we extract the melody track and cut it into overlapping segments of equal length. From these fragments, some numerical features are extracted as descriptors of style samples. We show that a standard Bayesian classifier can be conveniently employed to build an effective musical style classifier, once this set of features has been extracted from musical data. Preliminary experimental results show the effectiveness of the developed classifier that represents the first component of a musical audio retrieval system

Keywords: Musical style, Bayesian classifier.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1061453

References:

[1] M. Crump, ¶Çé│A principal components approach to the perception of musical style¶Çé┤, Banff Annual Seminar in Cognitive Science (BASICS), Banff, Alberta May 10, 2002.
[2] S. Brook, Barry, ¶Çé│Style and Content Analysis in Music¶Çé┤ The simplified Plaine and Easie Code in The Analysis of Communication Content, George Gerbner et Al. New York, Wiley 1969
[3] D. Cope, ¶Çé│Computers and Musical style¶Çé┤, Madison, Wisconsin: A-R Editions, Inc. 1991
[4] J. La Rue, ¶Çé│Guidelines for style analysis¶Çé┤, New York: W. W. Norton & Company 1970;
[5] Leonard Meyer, ¶Çé│Style and Music¶Çé┤ Philadelphia: University of Pennsylvania Press, 1989.
[6] G. Buzzanca, ¶Çé│A supervised learning approach to musical style recognition¶Çé┤ Additional proceedings of the Second International Conference ICMAI, Edinburgh, Scotland. 2002
[7] MIDI2TXT v1.14 midi binaries to text mnemonic by Guenter Nagler 1995.
[8] D. J. C. MacKay, ¶Çé│Information-based objective functions for active data selection. Neural Computation 4¶Çé┤ (4), 590- 6041992c 1992a
[9] D. J. C. MacKay, ¶Çé│Bayesian methods for backpropagation networks¶Çé┤. In E. Domany, J. L. van Hemmen, and K. Schulten (Eds.), Models of Neural Networks III, Chapter 6. New York: Springer-Verlag 1994a
[10] C. M. Bishop, ¶Çé│Neural Networks for pattern Recognition¶Çé┤. Clarendon Press, 1995.