NCM-Based Raga Classification Using Musical Features

Abstract

This paper deals with the study of Carnatic raga identification using musical features. In Carnatic music, there are 72 melakartha ragas. Each raga is denoted by musical notes. The musical features of 72 main ragas are extracted. A number of features such as pitch, timbre, tonal, rhythmic features have been discussed with reference to their ability to distinguish different ragas. Due to the intricate nature of Carnatic music, the concept of neutrosophic logic is used to identify each raga. This is because the concept of neutrosophic logic lies in the neutralities present in between truth and false. This creates a component of indeterminacy, which will make raga identification more accurate and smooth. Neutrosophic Cognitive Maps (NCMs) are drawn based on the musical features and solved. Using neutrosophic logic, a reduced set of musical features is arrived for each raga which can be thought of features characterizing the raga. Each raga is classified using a set of musical features which are solutions of NCMs. This paper represents one of the first attempts to classify all 72 melakartha ragas of using neutrosophic logic.