Bayesian confidence scoring and adaptation techniques for speech recognition

Tae Yoon Kim, Hanseok Ko

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Bayesian combining of confidence measures is proposed for speech recognition. Bayesian combining is achieved by the estimation of joint pdf of confidence feature vector in correct and incorrect hypothesis classes. In addition, the adaptation of a confidence score using the pdf is presented. The proposed methods reduced the classification error rate by 18% from the conventional single feature based confidence scoring method in isolated word Out-of-Vocabulary rejection test.

Original languageEnglish
Pages (from-to)1756-1759
Number of pages4
JournalIEICE Transactions on Communications
VolumeE88-B
Issue number4
DOIs
Publication statusPublished - 2005 Sep 9

Fingerprint

Speech recognition

Keywords

  • Adaptation
  • Confidence measure
  • OOV rejection
  • Speech recognition

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Networks and Communications

Cite this

Bayesian confidence scoring and adaptation techniques for speech recognition. / Kim, Tae Yoon; Ko, Hanseok.

In: IEICE Transactions on Communications, Vol. E88-B, No. 4, 09.09.2005, p. 1756-1759.

Research output: Contribution to journalArticle

@article{bc52113d0ae84c9f9be63ed770465143,
title = "Bayesian confidence scoring and adaptation techniques for speech recognition",
abstract = "Bayesian combining of confidence measures is proposed for speech recognition. Bayesian combining is achieved by the estimation of joint pdf of confidence feature vector in correct and incorrect hypothesis classes. In addition, the adaptation of a confidence score using the pdf is presented. The proposed methods reduced the classification error rate by 18{\%} from the conventional single feature based confidence scoring method in isolated word Out-of-Vocabulary rejection test.",
keywords = "Adaptation, Confidence measure, OOV rejection, Speech recognition",
author = "Kim, {Tae Yoon} and Hanseok Ko",
year = "2005",
month = "9",
day = "9",
doi = "10.1093/ietcom/e88-b.4.1756",
language = "English",
volume = "E88-B",
pages = "1756--1759",
journal = "IEICE Transactions on Communications",
issn = "0916-8516",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "4",

}

TY - JOUR

T1 - Bayesian confidence scoring and adaptation techniques for speech recognition

AU - Kim, Tae Yoon

AU - Ko, Hanseok

PY - 2005/9/9

Y1 - 2005/9/9

N2 - Bayesian combining of confidence measures is proposed for speech recognition. Bayesian combining is achieved by the estimation of joint pdf of confidence feature vector in correct and incorrect hypothesis classes. In addition, the adaptation of a confidence score using the pdf is presented. The proposed methods reduced the classification error rate by 18% from the conventional single feature based confidence scoring method in isolated word Out-of-Vocabulary rejection test.

AB - Bayesian combining of confidence measures is proposed for speech recognition. Bayesian combining is achieved by the estimation of joint pdf of confidence feature vector in correct and incorrect hypothesis classes. In addition, the adaptation of a confidence score using the pdf is presented. The proposed methods reduced the classification error rate by 18% from the conventional single feature based confidence scoring method in isolated word Out-of-Vocabulary rejection test.

KW - Adaptation

KW - Confidence measure

KW - OOV rejection

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=24144499416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24144499416&partnerID=8YFLogxK

U2 - 10.1093/ietcom/e88-b.4.1756

DO - 10.1093/ietcom/e88-b.4.1756

M3 - Article

AN - SCOPUS:24144499416

VL - E88-B

SP - 1756

EP - 1759

JO - IEICE Transactions on Communications

JF - IEICE Transactions on Communications

SN - 0916-8516

IS - 4

ER -