The following project was done in anticipation of publishing a research paper in the renowned International symposium on Frontiers of Research in Speech and Music (FRSM) conference. The paper will be published as part of the Advances in Intelligent Systems and Computing (AISC), Springer Proceedings Series (due July 2021). The papers will be indexed in ISI Proceedings, EI-Compendex, DBLP, SCOPUS, Google Scholar and Springerlink.
In music Information Retrieval (MIR), Raga Classification plays a vital role in discerning the features of Indian classical music and in a multitude of other tasks like database organisation of music files to music recommendation systems. The project encompasses a variety of techniques like ANN, LSTM and XGBoost models for the task of Raga Identification. The work is initially carried out on a set of 10 ragas and then extended to 20 ragas. Both the tasks showed state of the art results with an accuracy of 99.66% and 98.93% for a set of ten and twenty raagas set respectively.
The process was carried out on the raagas pertaining to Carnatic music, a division of Indian classical music. The data samples for the same were obtained from a standard data set belonging to the Dunya Website. Only those ragas which had around 15 songs were chosen. This was done to ensure homogeneity of the dataset.