Data-Driven Stastical Modeling of Violin Bowing Gesture Parameter Contours
Esteban Maestre · University of Michigan Library Repository · 2009
We present a framework for modeling right-hand gestures in bowed-string instrument playing, applied to violin. Nearly non-intrusive sensing techniques allow for accurate acqui- sition of relevant timbre-related bowing gesture parameter cues. We model the temporal contour of bow transversal velocity, bow pressing force, and bow-bridge distance as se- quences of short segments, in particular Bcubic curve segments. Considering different articulations, dynamics, and contexts, a number of note classes is defined. Gesture pa- rameter contours of a performance database are analyzed at note-level by following a predefined grammar that dic- tates characteristics of curve segment sequences for each of the classes into consideration. Based on dynamic pro- gramming, gesture parameter contour analysis provides an optimal curve parameter vector for each note. The informa- tion present in such parameter vector is enough for recon- structing original gesture parameter contours with signifi- cant fidelity. From the resulting representation vectors, we construct a statistical model based on Gaussian mixtures, suitable for both analysis and synthesis of bowing gesture parameter contours. We show the potential of the model by synthesizing bowing gesture parameter contours from an annotated input score. Finally, we point out promising ap- plications and developments.