Share Email Print

Proceedings Paper

Explicit noise hypotheses in speech recognition
Author(s): Richard K. Fox; John R. Josephson
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Noise is typically present in the input signal for perception problems. Noise arises in speech recognition due to both background sounds, and unintentional derivations from the intended utterance on the part of the speaker. The task of speech recognition is to correctly identify the words (or meaning) carried by the speech signal. Thus the speech recognizer must be able to successfully handle noise. We describe here a method of explicitly identifying and labeling noise elements in a speech signal. NOISE hypotheses are generated, and considered for acceptance, as part of an abductive inference strategy for speech processing. An abductive problem solver is able to treat noise within a unified inferential framework, treating noise hypotheses similarly to other hypotheses, weighing the explanatory alternatives in a context-sensitive manner, and with no need to resort to indirect methods to achieve noise tolerance.

Paper Details

Date Published: 29 October 1993
PDF: 11 pages
Proc. SPIE 2032, Neural and Stochastic Methods in Image and Signal Processing II, (29 October 1993); doi: 10.1117/12.162043
Show Author Affiliations
Richard K. Fox, Univ. of Texas Pan American (United States)
John R. Josephson, The Ohio State Univ. (United States)

Published in SPIE Proceedings Vol. 2032:
Neural and Stochastic Methods in Image and Signal Processing II
Su-Shing Chen, Editor(s)

© SPIE. Terms of Use
Back to Top