文章基本信息

标题：CLASSIFICATION OF NORTH INDIAN MUSICAL INSTRUMENTS USING SPECTRAL FEATURES
本地全文：下载
作者：Mona Kumari ; Priyank Kumar ; Sandeep Solanki 等
期刊名称：Computer Sciences and Telecommunications
印刷版ISSN：1512-1232
出版年度：2010
卷号：29
期号：6
页码：11-24
出版社：Internet Academy
摘要：Musical instrument recognition finds its application in the field of Musical Information Retrieval such as automatic tagging, cataloguing of music for search engine, automatic musical transcription, genre classification etc. The present work has been emphasized on the identification of different North Indian Musical Instruments such as Flute, Sitar, Dholak, Bhapang, and Mandar. The spectral features were obtained from framed music signal and in order to classify them with each instrument spectral features such as spectral centroid, Spectral Spread, Spectral Skewness, Spectral Kurtosis, Spectral Slope, Spectral Rolloff, are obtained from MFCC feature extraction. Auto Correlation was also obtained for better classification of the instruments. It has been found that MFCC with 13 coefficients gives a better classification of monophonic instruments (Indian musical instruments) than any other feature. However, it was observed that Auto Correlation with 12 coefficients classified the instruments reasonably well. This technique is being attempted for the first time in classifying North Indian musical instruments.
关键词：Musical instrument; feature extraction; spectral feature; MFCC; ; Autocorrelation