CMMR/FRSM 2011 - 9-12 March, 2011, Utkal University, Bhubaneswar AN OBJECTIVE ASSESSMENT TOOL FOR ORNAMENTATION IN SINGING Chitralekha Gupta and Preeti Rao Department of Electrical Engineering Indian Institute of Technology Bombay, Mumbai 400076, India {chitralekha, prao}@ee.iitb.ac.in Abstract Important aspects of judging a singing performance include musical accuracy and voice quality. In the context of Indian classical music, not only is the correct sequence of notes important to musical accuracy but also the nature of pitch transitions between notes. These transitions are essentially related to alankars (ornaments) that embellish the inherent beauty of the genre. Thus a higher level of singing skill involves achieving the necessary expressiveness via correct rendering of ornamentation, and this ability can serve to distinguish a well trained singer from an amateur. In this work, we explore objective methods to assess the quality of ornamentation rendered by a singer with reference model or ideal singer of the same song. Methods are proposed for the perceptually relevant comparison of complex pitch movements, and validated with respect to subjective ratings by human experts. Such an objective assessment system can be a valuable feedback tool in the training of amateur singers and also be used for competitive singing platforms that aim to identify good singers from the masses. Keywords – Singing scoring, ornamentation, Indian music, polynomial curve fitting 1. Introduction Perceived pitch relates to the fundamental frequency of a sound. The pitch of singing voice is determined by the rate of vibration of the vocal cords. For males, the fundamental frequency can vary from 70 to 500 Hz and for females, from 150 to 700 Hz [1]. Melody is the temporal sequence of discrete notes (pitches). But for singing voice, this pitch contour is not just made up of discrete horizontal lines corresponding to distinct steady notes but with a continuously evolving curve of macro and micro-tonal pitch movements and ornamentations. The latter comprise the expressive elements of music. In Indian classical singing, quantitative assessment of progress of music learners based on improvisation, form and content is a topic under research [2]. There has been some objective analysis of the ornament meend, resembling a glide. Its proper rendition involves the accuracy of starting and ending notes, speed, and accent on intermediate notes [3,4]. Perceptual tests to differentiate between synthesized singing of vowel /a/ with a pitch movement of falling and rising intonation (concave, convex , linear) between two steady pitch states, 150 and 170 Hz, using a second degree equation, revealed that the different types of transitory movements are cognitively separable [5]. A methodology for automatic extraction of meend from the performances in Hindustani vocal music described in [6] also uses the fit to a second degree equation to detect the meend. Also automatic classification of meend attempted in [7] gives some important observations like frequency of occurrence of descending meends is maximum, and then that of rise-fall meends (meend with kan swar). The meends with intermediate touch notes are relatively less frequent. The duration of meend is generally between 300 - 500 ms. In this work, the ornament glide has been considered. The assessment is with respect to the “model” or ideal rendition of the same song. Considering the relatively easy availability of singers for popular Indian film music, we use Hindustani classical music based movie songs for testing our methods. The previous work reported on glide has been to model it computationally. In this paper, we attempt to use computational modeling to assess the degree of perceived closeness between a given rendition and a reference rendition taken to be that of the original playback singer of the song.