Musical instrument recognition by pairwise classification strategies

Author

Essid, Slim ; Richard, Gaël ; David, Bertrand

Author_Institution

LTCI-CNRS, Paris

Volume

14

Issue

4

fYear

2006

fDate

7/1/2006 12:00:00 AM

Firstpage

1401

Lastpage

1412

Abstract

Musical instrument recognition is an important aspect of music information retrieval. In this paper, statistical pattern recognition techniques are utilized to tackle the problem in the context of solo musical phrases. Ten instrument classes from different instrument families are considered. A large sound database is collected from excerpts of musical phrases acquired from commercial recordings translating different instrument instances, performers, and recording conditions. More than 150 signal processing features are studied including new descriptors. Two feature selection techniques, inertia ratio maximization with feature space projection and genetic algorithms are considered in a class pairwise manner whereby the most relevant features are fetched for each instrument pair. For the classification task, experimental results are provided using Gaussian mixture models (GMMs) and support vector machines (SVMs). It is shown that higher recognition rates can be reached with pairwise optimized subsets of features in association with SVM classification using a radial basis function kernel

Keywords

Gaussian processes; acoustic signal processing; genetic algorithms; information retrieval; musical instruments; pattern recognition; radial basis function networks; support vector machines; Gaussian mixture models; descriptors; feature selection technique; feature space projection; genetic algorithms; inertia ratio maximization; music information retrieval; musical instrument recognition; pairwise classification strategies; radial basis function kernel; solo musical phrase; statistical pattern recognition; support vector machines; Genetic algorithms; Instruments; Music information retrieval; Pattern recognition; Signal processing algorithms; Source separation; Spatial databases; Support vector machine classification; Support vector machines; Timbre; Feature selection; Gaussian mixture model (GMM); genetic algorithms; inertia ratio maximization with feature space projection (IRMFSP); musical instrument recognition; pairwise classification; support vector machine (SVM);

fLanguage

English

Journal_Title

Audio, Speech, and Language Processing, IEEE Transactions on

Publisher

ieee

ISSN

1558-7916

Type

jour

DOI

10.1109/TSA.2005.860842

Filename

1643665