Computing relative redundancy to measure grammatical constraint in speech recognition tasks

24 March 2005

conference paper
Published by Institute of Electrical and Electronics Engineers (IEEE)

Vol. 3, 409-412
https://doi.org/10.1109/icassp.1978.1170406

Abstract

In this paper we present new and computationally efficient algorithms for computing some statistical properties of finite languages. In particular, the relative redundancy which measures grammatical constraint, is computed for several speech recognition task languages which have appeared in the literature.

Keywords

SPEECH RECOGNITION
ENTROPY
Q MEASUREMENT
NATURAL LANGUAGES
LABORATORIES
PARTICLE MEASUREMENTS
VOCABULARY
UNCERTAINTY
MARKOV PROCESSES

This publication has 1 reference indexed in Scilit:

Improving word recognition accuracy by means of syntax
Published by Institute of Electrical and Electronics Engineers (IEEE) ,2005

Cited by 14 articles