Comparing B-Spline and Spline Models for F0 Modelling
This article describes a new approach to estimate F 0 curves using B-spline and Spline models characterized by a knot sequence and associated control points. The free parameters of the model are the number of knots and their location. The free-knot placement, which is a NP-hard problem, is done using a global MLE (Maximum Likelihood Estimation) within a simulated-annealing strategy. Experiments are conducted in a speech processing context on a 7000 syllables french corpus. We estimate the two challenging models for increasing values of the number of free parameters. We show that a B-spline model provides a slightly better improvement than the Spline model in terms of RMS error.
KeywordsSpline Function Spline Model Quadratic Spline Quadratic Distance Melodic Contour
Unable to display preview. Download preview PDF.
- 1.Raux, A., Black, A.: A unit selection approach to f0 modeling and its application to emphasis. In: Proc. ASRU Conf., pp. 700–703 (2003)Google Scholar
- 2.Hirst, D., Cristo, A.D., Espesser, R.: Levels of representation and levels of analysis for the description of intonation systems. In: Horne, M. (ed.) Prosody: Theory and Experiment, vol. 14, pp. 51–87. Kluwer Academic Pusblisher, Dordrecht (2000)Google Scholar
- 4.Sakai, S., Glass, J.: Fundamental frequency modeling for corpus-based speech synthesis based on statistical learning techniques. In: Proc. ASRU Conf., pp. 712–717 (2003)Google Scholar
- 5.Barbot, N., Boeffard, O., Lolive, D.: F0 stylisation with a free-knot b-spline model and simulated-annealing optimization. In: Proc. Eurospeech Conf., pp. 325–328 (2005)Google Scholar
- 7.Mouline, S., Boeffard, O., Bagshaw, P.: Automatic adaptation of the momel f 0 stylisation algorithm to new corpora. In: Proc. of ICSLP (2004)Google Scholar