Automatic Genre Classification of Musical Signals
- PDF / 469,201 Bytes
- 12 Pages / 600.03 x 792 pts Page_size
- 103 Downloads / 251 Views
Research Article Automatic Genre Classification of Musical Signals Jayme Garcia Arnal Barbedo and Amauri Lopes Departamento de Comunicac¸o˜es, Faculdade de Engenharia El´etrica e de Computac¸a˜ o (FEEC), Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6101, Campinas 13083-852, Brazil Received 28 November 2005; Revised 26 June 2006; Accepted 29 June 2006 Recommended by George Tzanetakis We present a strategy to perform automatic genre classification of musical signals. The technique divides the signals into 21.3 milliseconds frames, from which 4 features are extracted. The values of each feature are treated over 1-second analysis segments. Some statistical results of the features along each analysis segment are used to determine a vector of summary features that characterizes the respective segment. Next, a classification procedure uses those vectors to differentiate between genres. The classification procedure has two main characteristics: (1) a very wide and deep taxonomy, which allows a very meticulous comparison between different genres, and (2) a wide pairwise comparison of genres, which allows emphasizing the differences between each pair of genres. The procedure points out the genre that best fits the characteristics of each segment. The final classification of the signal is given by the genre that appears more times along all signal segments. The approach has shown very good accuracy even for the lowest layers of the hierarchical structure. Copyright © 2007 J. G. A. Barbedo and A. Lopes. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1.
INTRODUCTION
The advances experienced in the last decades in areas as information, communication, and media technologies have made available a large amount of all kinds of data. This is particularly true for music, whose databases have grown exponentially since the advent of the first perceptual coders early in the 90’s. This situation demands for tools able to ease searching, retrieving, and handling such a huge amount of data. Among those tools, automatic musical genre classifiers (AMGC) can have a particularly important role, since they could be able to automatically index and retrieve audio data in a human-independent way. This is very useful because a large portion of the metadata used to describe music content is inconsistent or incomplete. Music search and retrieval is the most important application of AGC, but it is not the only one. There are several other technologies that can benefit from AGC. For example, it would be possible to create an automatic equalizer able to choose which frequency bands should be attenuated or reinforced according to the label assigned to the signal being considered. AGC could also be used to automatically select radio stations playing a particular genre of music. The research field of automatic music genre classification has got increasing importance in the last f
Data Loading...
