Signal modeling enhancements for automatic speech recognition

Author

Nossair, Zaki B. ; Silsbee, Peter L. ; Zahorian, Stephen A.

Author_Institution

Dept. of Electr. & Comput. Eng., Old Dominion Univ., Norfolk, VA, USA

Volume

1

fYear

1995

fDate

9-12 May 1995

Firstpage

824

Abstract

Experiments in modeling speech signals for phoneme classification are described. Enhancements to standard speech processing methods include basis vector representations of dynamic feature trajectories, morphological smoothing (dilation) of spectral features, and the use of many closely spaced, short analysis windows. Results are reported from experiments using the TIMIT database of up to 71.0% correct classification of 16 presegmented vowels in a noise-free environment, and 54.5% correct classification in a 10 dB signal-to-noise ratio environment

Keywords

modelling; smoothing methods; spectral analysis; speech enhancement; speech recognition; TIMIT database; automatic speech recognition; basis vector representations; dilation; dynamic feature trajectories; morphological smoothing; noise-free environment; phoneme classification; short analysis windows; signal modeling enhancements; spectral features; standard speech processing methods; Automatic speech recognition; Cepstral analysis; Finite impulse response filter; Frequency; Sampling methods; Signal to noise ratio; Speech analysis; Speech recognition; Testing; Working environment noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on

Conference_Location

Detroit, MI

ISSN

1520-6149

Print_ISBN

0-7803-2431-5

Type

conf

DOI

10.1109/ICASSP.1995.479821

Filename

479821