Minimum Bayes Risk Estimation and Decoding in Large Vocabulary Continuous Speech Recognition

William BYRNE

doi:10.1093/ietisy/e89-d.3.900

Minimum Bayes Risk Estimation and Decoding in Large Vocabulary Continuous Speech Recognition

William BYRNE

Full Text Views

0

Share
Cite this

Summary :

Minimum Bayes risk estimation and decoding strategies based on lattice segmentation techniques can be used to refine large vocabulary continuous speech recognition systems through the estimation of the parameters of the underlying hidden Markov models and through the identification of smaller recognition tasks which provides the opportunity to incorporate novel modeling and decoding procedures in LVCSR. These techniques are discussed in the context of going 'beyond HMMs', showing in particular that this process of subproblem identification makes it possible to train and apply small-domain binary pattern classifiers, such as Support Vector Machines, to large vocabulary continuous speech recognition.

Publication: IEICE TRANSACTIONS on Information Vol.E89-D No.3 pp.900-907

Publication Date: 2006/03/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1093/ietisy/e89-d.3.900

Type of Manuscript: Special Section INVITED PAPER (Special Section on Statistical Modeling for Speech Processing)

Category

Cite this

Copy

William BYRNE, "Minimum Bayes Risk Estimation and Decoding in Large Vocabulary Continuous Speech Recognition" in IEICE TRANSACTIONS on Information, vol. E89-D, no. 3, pp. 900-907, March 2006, doi: 10.1093/ietisy/e89-d.3.900.
Abstract: Minimum Bayes risk estimation and decoding strategies based on lattice segmentation techniques can be used to refine large vocabulary continuous speech recognition systems through the estimation of the parameters of the underlying hidden Markov models and through the identification of smaller recognition tasks which provides the opportunity to incorporate novel modeling and decoding procedures in LVCSR. These techniques are discussed in the context of going 'beyond HMMs', showing in particular that this process of subproblem identification makes it possible to train and apply small-domain binary pattern classifiers, such as Support Vector Machines, to large vocabulary continuous speech recognition.
URL: https://globals.ieice.org/en_transactions/information/10.1093/ietisy/e89-d.3.900/_p

Copy

@ARTICLE{e89-d_3_900,
author={William BYRNE, },
journal={IEICE TRANSACTIONS on Information},
title={Minimum Bayes Risk Estimation and Decoding in Large Vocabulary Continuous Speech Recognition},
year={2006},
volume={E89-D},
number={3},
pages={900-907},
abstract={Minimum Bayes risk estimation and decoding strategies based on lattice segmentation techniques can be used to refine large vocabulary continuous speech recognition systems through the estimation of the parameters of the underlying hidden Markov models and through the identification of smaller recognition tasks which provides the opportunity to incorporate novel modeling and decoding procedures in LVCSR. These techniques are discussed in the context of going 'beyond HMMs', showing in particular that this process of subproblem identification makes it possible to train and apply small-domain binary pattern classifiers, such as Support Vector Machines, to large vocabulary continuous speech recognition.},
keywords={},
doi={10.1093/ietisy/e89-d.3.900},
ISSN={1745-1361},
month={March},}

Copy

TY - JOUR
TI - Minimum Bayes Risk Estimation and Decoding in Large Vocabulary Continuous Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 900
EP - 907
AU - William BYRNE
PY - 2006
DO - 10.1093/ietisy/e89-d.3.900
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E89-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2006
AB - Minimum Bayes risk estimation and decoding strategies based on lattice segmentation techniques can be used to refine large vocabulary continuous speech recognition systems through the estimation of the parameters of the underlying hidden Markov models and through the identification of smaller recognition tasks which provides the opportunity to incorporate novel modeling and decoding procedures in LVCSR. These techniques are discussed in the context of going 'beyond HMMs', showing in particular that this process of subproblem identification makes it possible to train and apply small-domain binary pattern classifiers, such as Support Vector Machines, to large vocabulary continuous speech recognition.
ER -