HIERARCHICAL GAUSSIAN TREE WITH INERTIA RATIO MAXIMIZATION FOR THE CLASSIFICATION OF LARGE MUSICAL INSTRUMENT DATABASES

1. ABSTRACT This paper addresses the problem of classifying large databases of musical instrument sounds. We propose an efficient algorithm for selecting the most appropriate features for a given classification task. This algorithm, called IRMFSP, is based on the maximization of the ratio of the between-class inertia to the total inertia combined with a step-wise feature space orthogonalization. The IRMFSP algorithm is then compared successfully to the widely used feature selection algorithm CFS. We then show the limits of usual flat (all classes considered on a same level) classifiers for large database classification and propose the use of hierarchical classifiers. Finally, we show the applicability of our system for large database classification. Especially considered is the application when our classification system is trained on a given database and used for the classification of another database possibly recorded in co mpletely different conditions.

[1]  Xavier Rodet,et al.  Studies and Improvements in Automatic Classification of Musical Sound Samples , 2003, ICMC.

[2]  Stephen McAdams,et al.  Instrument Sound Description in the Context of MPEG-7 , 2000, ICMC.

[3]  Perfecto Herrera-Boyer,et al.  Automatic Classification of Musical Instrument Sounds , 2003 .

[4]  J C Brown.  Computer identification of musical instruments using pattern recognition with cepstral coefficients as features. , 1999, The Journal of the Acoustical Society of America.

[5]  Stephen R. Garner,et al.  WEKA: The Waikato Environment for Knowledge Analysis , 1996 .

[6]  Kristoffer Jensen,et al.  Binary Decision Tree Classification of Musical Sounds , 1999, ICMC.

[7]  Lluís A. Belanche Muñoz,et al.  Feature selection algorithms: a survey and experimental evaluation , 2002, 2002 IEEE International Conference on Data Mining, 2002. Proceedings..

[8]  Malcolm Slaney,et al.  Construction and evaluation of a robust multifeature speech/music discriminator , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[9]  Xavier Rodet,et al.  Automatically selecting signal descriptors for SoundClassification , 2002, ICMC.

[10]  C.-C. Jay Kuo,et al.  Hierarchical classification of audio data for archiving and retrieving , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[11]  Keith Dana Martin,et al.  Sound-source recognition: a theory and computational model , 1999 .

[12]  Stephen McAdams,et al.  Caractérisation du timbre des sons complexes.II. Analyses acoustiques et quantification psychophysique , 1994 .

[13]  Mark A. Hall,et al.  Correlation-based Feature Selection for Discrete and Numeric Class Machine Learning , 1999, ICML.

[14]  François Pachet,et al.  The CUIDADO Project , 2002, ISMIR.

[15]  Antti Eronen,et al.  Comparison of features for musical instrument recognition , 2001, Proceedings of the 2001 IEEE Workshop on the Applications of Signal Processing to Audio and Acoustics (Cat. No.01TH8575).