Paper Title
Automatic Raga Classification based on THAAT
Abstract
In North Indian classical music Ragas have been categorized by ten different Thaats. Each of the ragas
comprises of melodic combination of the notes based on Thaat to which it belongs. In this paper we classify Ragas into
various Thaats by detecting the key derived notes (Komal and Teevra Swar) present in the music sample. We used cepstral
analysis to find the fundamental frequency and Chebyshev’s inequality to detect the notes of the music signal. Each Thaat
has a typical combination of derived notes which forms the basis of classification.
Index terms - Raga, Thaat, Komalnote, Cepstral Analysis, Chebyshev’s inequality