Investigation into the Identification of Ragas: A Detailed Analysis

This paper endeavors to examine and ascertain the optimal AIML methodologies for the discernment of Ragas in Indian Classical Music. Following an extensive review of pertinent literature, we have meticulously devised a methodology for the comprehensive processing of our data sets, aiming to yield optimal results. The overarching goal of our paper is to delve into diverse AI/ML techniques, initially applied to the data sets under consideration with constrained values, and subsequently broaden the data set scope, to encompass all 5040 Ragas in Indian Classical Music, thereby enhancing system efficiency. Our investigation offers valuable insights into potential enhancements to the data set, addressing its limitations and augmenting the concept's efficacy for prospective applications. The data set necessitates meticulous pre processing, paving the way for the strategic implementation of techniques such as K-nearest neighbors (KNN) and Convolutional Neural Networks (CNN) on the Hindustani Music Dataset(HMD) and Classical Music data set (CMD)separately, with the ultimate objective of extracting optimal outcomes.