SNR-normalisation for robust speech recognition

Author

Claes, Tom ; Van Compernolle, Dirk

Author_Institution

Katholieke Univ., Leuven, Heverlee, Belgium

Volume

1

fYear

1996

fDate

7-10 May 1996

Firstpage

331

Abstract

A new normalisation technique for speech recognition in adverse conditions is presented. Specifically the influence of additive noise in combination with convolutive distortions is considered. In the proposed method a masking constant is added to the outputs of a mel scale triangular filterbank. This is done for testing and training samples. The goal is to normalize the signal-to-noise ratio (SNR) in each frequency band by adapting the masking constant depending on the measured SNR or dynamic range in each band. This makes the extracted parameters less sensitive to the noise level, but also the influence of channel distortions is suppressed. The method is easy to implement and works on-line. Experimental results are given on the NOISEX-92 database and on real car data

Keywords

band-pass filters; convolution; filtering theory; noise; signal sampling; speech recognition; NOISEX-92 database; SNR-normalisation; additive noise; adverse conditions; channel distortions; convolutive distortions; dynamic range; experimental results; frequency band; masking constant; measured SNR; mel scale triangular filterbank; noise level; real car data; robust speech recognition; signal-to-noise ratio; testing samples; training samples; Additive noise; Data mining; Distortion measurement; Dynamic range; Filter bank; Frequency measurement; Noise robustness; Signal to noise ratio; Speech recognition; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on

Conference_Location

Atlanta, GA

ISSN

1520-6149

Print_ISBN

0-7803-3192-3

Type

conf

DOI

10.1109/ICASSP.1996.541099

Filename

541099