Feature-Driven Classification of Musical Styles

Abstract ² 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 ² Musical style, Bayesian classifier. I. I NTRODUCTION USICAL style as well as the mechanisms underlying style recognition are relatively ill-defined [1]. Several definitions of musical style have been formulated so far. Cope >@GHILQHVPXVLFDOVW\OHDV³WKHLGHQWLILDEOHFKDUDFWHULVWLFVRIDFRPSRVHU¶VPXVLFZKLFKDre recognizably similar from one ZRUNWRDQRWKHU´$QRWKHUGHILQLWLRQRIPXVLFDOVW\OHLVJLYHQLQ>@³6W\OHLVDUHSOLFDWLRQRUSDWWHUQLQJHLWKHULQKXPDQbehavior or in the artifacts produced by human behavior, that results from a series of choices made within one set of FRQVWUDLQWV´,Q>@WKHRUHWLFDOFRPSUHKHQVLYHJXLGHOLQHVIRUstyle analysis are provided by dissecting musical style into three dimensions: large (groups of works, work, movement), middle (part, section, paragraph, sentence) and small (motive, subphrase, phrase). Whatever the definition, musical style (and its recognition) is something related to human nature: the average layperson can recognize the difference among simple stylistic features. But things change when it comes to computers. Automatic recognition of musical style is not an easy task. Even relatively simple stylistic manners of playing an instrument, VXFKDVSOD\LQJ³HQHUJHWLFDOO\´SOD\LQJ³O\ULFDOO\´RUSOD\LQJ³ZLWKV\QFRSDWLRQ´DUHGLIILFXOWWRGHWHFWUHOLDEO\E\DXWRmatic classification tools. Nevertheless, automatic classification of musical styles is gaining more and more importance since it may serve as a way to structure and organize the increasingly large number of