Raga Classification Based on MFCC and Variants

Abstract

Analysis and classification of raga is the need of time especially in music industry. With the presence of abundance of multimedia data on internet, it is imperative to develop appropriate tools to classify ragas. In this work, an attempt has been made to use Mel Frequency Cepstral Coefficients (MFCC) for classification. MFCC pattern present in the signal along with its statistical distribution can be characterized using MFCC based co-occurrence matrix. This matrix captures both tonal and temporal aspects of melody. Ragas differ in terms of distribution of spectral power which was the motivation to use MFCC. Support Vector Machine (SVM) has been utilized as the classifier. Database consists of 30 ragas from Hindustani music dataset (HMD) part of Indian art Music Raga recognition dataset which is the largest publicly available database. Experimental result indicates the effectiveness of the proposed scheme which is giving accuracy of 86.6%