8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Evaluation of Model-Based Feature Enhancement on the AURORA-4 Task

Veronique Stouten, Hugo van Hamme, Jacques Duchateau, Patrick Wambacq

Katholieke Universiteit Leuven, Belgium

In this paper we focus on the challenging task of noise robustness for large vocabulary Continuous Speech Recognition (LVCSR) systems in non-stationary noise environments. We have extended our Model-Based Feature Enhancement (MBFE) algorithm - that we earlier successfully applied to small vocabulary CSR in the AURORA-2 framework - to cope with the new demands that are imposed by the large vocabulary size in the AURORA-4 task. To incorporate a priori knowledge of the background noise, we combine scalable Hidden Markov Models (HMMs) of the cepstral feature vectors of both clean speech and noise, using a Vector Taylor Series approximation in the power spectral domain. Then, a global MMSE-estimate of the clean speech is calculated based on this combined HMM. This technique is easily embeddable in the feature extraction module of a recogniser and is intrinsically suited for the removal of non-stationary additive noise. Our approach is validated on the AURORA-4 task, revealing a significant gain in noise robustness over the baseline.

Full Paper

Bibliographic reference.  Stouten, Veronique / Hamme, Hugo van / Duchateau, Jacques / Wambacq, Patrick (2003): "Evaluation of model-based feature enhancement on the AURORA-4 task", In EUROSPEECH-2003, 349-352.