Time-aligned SVD analysis for speaker identification

Abstract

It is described in this paper for the application of pitch/tone information in the parametric trajectory model. Pitch as a dynamic feature and its contours----tone as a segmental-level feature are deserved their own particular characteristics, which match case of parametric trajectory model better compared with MFCC and energy. Here we give an improved pitch extraction algorithm and especially the "total" and ''parallel'' integration methods to combine these information with the base model. In the experiment of Mandarin connected digit recognition, we achieve 22.87% and 33.54% error reduction respectively for them, moreover when combined with these two methods, 38.72% error reduction is obtained.

Linear Software Models for Well-composed Systems

“…Although diagonality has seldom been calculated within modularity, formulas have appeared in other contexts. Clemins in (Clemins, et al 2002) used for speaker identification, the Frobenius normthe sum of the squaresof all off-diagonal elements to measure diagonality. Our offdiag definition is better suited to modularity, as it directly reflects distance to the diagonal, while the Frobenius norm just sums Boolean elements.…”