This paper presents a novel approach to the estimation of the trajectories of spectral center-of-gravity using robust statistical models with penalized weighted spline smoothers. Most of the existing methods for tracking speech formant trajectories are based on dynamic programming algorithms with certain continuity constraints on the formant frequencies ([6, 8, 9]). The objective functions (or loss functions) in these approaches are usually ad hoc and have very complex expressions that are difficult to optimize. Also, many existing methods rely on the accuracy of the LPC spectral peaks and are not very robust against possible missing or spurious peaks. In this paper, instead of using the peaks of the LPC spectral functions, we propose a new approach to the estimation of the "center-of-gravities" in spectrogram using mixture models of spline smoothers ([5, 10]). There are two major advantages in this new approach: 1) it is robust against missing peaks and spurious peaks that may occur in LPC peak finding algorithms; 2) trajectory smoothness is guaranteed by the properties of spline regression models.
Bibliographic reference. Sun, Don X. (1995): "Robust estimation of spectral center-of-gravity trajectories using mixture spline models", In EUROSPEECH-1995, 749-752.