On the Initialization of Dynamic Models for Speech Features

Conference: Sprachkommunikation 2010 - 9. ITG-Fachtagung
10/06/2010 - 10/08/2010 at Bochum, Deutschland

Proceedings: Sprachkommunikation 2010

Pages: 4Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Krueger, Alexander; Leutnant, Volker; Haeb-Umbach, Reinhold (Department of Communications Engineering, University of Paderborn, 33098 Paderborn, Germany)
Ackermann, Marcel; Bloemer, Johannes (Department of Computer Science, University of Paderborn, 33102 Paderborn, Germany)

In this work, a novel approach for the initialization of switching linear dynamic models (SLDMs) as dynamic models for the trajectory of speech features is proposed. Borrowing ideas from the "k-means++"-algorithm, the goal of this approach is to find distinctly different SLDMs, modelling the complex dynamics of the speech features, already at the initialization stage of a subsequently following "expectation-maximization (EM)"-algorithm. Experimental results comparing differently initialized SLDMs in a model-based speech feature enhancement scheme show the superiority of the proposed initialization routine in terms of a reduced word error rate on an automatic speech recognition task.