Abstract:
Music plays a vital role in our day-to-day life, especially in today’s digital age.
Computational musicology is an interdisciplinary area in which computational methods
are used to analyse musical structures: notes, chords, rhythms, and patterns thereof. While
western classical music is extensively explored, North Indian classical music remains to
be explored computationally. Meanwhile, our recent review of the literature revealed
that Raag identification is among the frequent data mining tasks applied to North Indian
music. However, recognition of their rhythmic structures is also important as it serves in a
multitude of applications e.g., intelligent music archival, enhanced navigation and
retrieval of music, and informed music listening. Rhythm in North Indian classical music
revolves around the theme of Taal - the cycle of beats of specific syllables and beats. It is
the most basic information for listeners to follow the rhythmic structure of music. Laya is
the speed of Taal and may vary between Vilambit (slow), Madhya (medium),
and Drut (fast). Taken together, the main aim of the proposed study is to apply data
mining for the recognition of Taal and Laya in North Indian classical music. A dataset of
151 excerpts (2mins; 44.1 kHz; stereo; .wav) from CompMusic Hindustani test corpus,
belonging to four popular Taals is used in the study. For each Taal, there are excerpts in
three Layas. Acoustic features pertaining to fluctuation, beat spectrum, onsets, event
density, tempo, metre, metroid, and pulse clarity will be extracted using MATLAB
MIRToolbox. The performance of frequently adopted algorithms e.g., k-Nearest Neighbor
and Support Vector Machine is to be compared in the study with the aim of developing a
classifier with higher accuracy. Even though the findings of the study would be limited by
the consideration of a smaller dataset, the study would make a promising contribution
through computationally exploring rhythmic patterns of a great musical tradition
Keywords: Music data mining, Taal and Laya recognition, North Indian classical music,
Rhythmic analysis, Computational musicology
