Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

A Corpus-Based Approach for Robust ASR in Reverberant Environments

Laurent Couvreur (1), Christophe Couvreur (1), Christophe Ris (2)

(1) Faculté Polytechnique de Mons, Belgium
(2) Lernout & Hauspie Speech Products, Belgium

In this paper, we discuss the use of artificial room reverberation to increase the performance of automatic speech recognition (ASR) systems in reverberant enclosures. Our approach consists in training acoustic models on artificially reverberated speech material. In order to obtain the desired reverberated speech training database, we propose to use a reverberating filter whose impulse response is designed to match two high-level acoustic properties of the target reverberant operating environment, namely the early-to-late energy ratio and the reverberation time. Speech recognition experiments in simulated reverberant environments show that recognizers trained on speech reverberated with the proposed method outperform systems trained on clean speech, even when channel normalization methods like CMS and logRASTA-PLP are used. The extension of our approach to multi-style training is also considered.


Full Paper

Bibliographic reference.  Couvreur, Laurent / Couvreur, Christophe / Ris, Christophe (2000): "A corpus-based approach for robust ASR in reverberant environments", In ICSLP-2000, vol.1, 397-400.