Title: Raga identification of Carnatic music based on the construction of Raga model
Authors: Rajeswari Sridhar; T.V. Geetha
Addresses: Department of Computer Science and Engineering, Anna University, Chennai 600025, India ' Department of Computer Science and Engineering, Anna University, Chennai 600025, India
Abstract: In this paper a Raga identification algorithm for Carnatic music is proposed which is based on designing and constructing a supervised static model called Raga model. This model is constructed using signal level features which are later mapped to musical features to convey the Raga lakshana characteristics. In this work the Raga lakshana characteristics namely Arohana, Avarohana, Graha, Amsa, Nyasa, Bahutva, are used for constructing the Raga model. The Raga identification process is a three-pronged approach which compares the signal level features, Raga lakshana of the input with the features available in the Raga model to determine the Raga. The final Raga is chosen as the one that is determined in at least two steps of the three pronged process and achieved an accuracy of nearly 80%.
Keywords: raga identification; Carnatic music; raga modelling; fundamental frequency; south India; Indian classical music.
DOI: 10.1504/IJSISE.2013.054795
International Journal of Signal and Imaging Systems Engineering, 2013 Vol.6 No.3, pp.172 - 181
Received: 19 Apr 2011
Accepted: 08 Oct 2011
Published online: 13 Sep 2013 *