Detecting tala Computationally in Polyphonic Context - A Novel Approach

In North-Indian-Music-System(NIMS),tablā is mostly used as percussive accompaniment for vocal-music in polyphonic-compositions. The human auditory system uses perceptual grouping of musical-elements and easily filters the tablā component, thereby decoding prominent rhythmic features like tāla, tempo from a polyphonic-composition. For Western music, lots of work have been reported for automated drum analysis of polyphonic-composition. However, attempts at computational analysis of tāla by separating the tablā-signal from mixed signal in NIMS have not been successful. Tablā is played with two components right and left. The right-hand component has frequency overlap with voice and other instruments. So, tāla analysis of polyphonic-composition, by accurately separating the tablā-signal from the mixture is a baffling task, therefore an area of challenge. In this work we propose a novel technique for successfully detecting tāla using left-tablā signal, producing meaningful results because the left-tablā normally doesn’t have frequency overlap with voice and other instruments. North-Indian-rhythm follows complex cyclic pattern, against linear approach of Western-rhythm. We have exploited this cyclic property along with stressed and non-stressed methods of playing tablāstrokes to extract a characteristic pattern from the left-tablā strokes, which, after matching with the grammar of tāla-system, determines the tāla and tempo of the composition. A large number of polyphonic(vocal+tablā+other-instruments) compositions has been analyzed with the methodology and the result clearly reveals the effectiveness of proposed techniques.