Confidence measures as a search guide in speech recognition

Sherif Abdou, Michael S Scordilis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Recently, there have been many efforts in developing confidence measures for speech recognition output. Usually, these measures are applied to the final result of the decoder. However, using these measures early in the search process can guide the search to more promising paths. In this paper we use a confidence metric, the Average Best_Base_Phoneme Rank, to dynamically tune the contribution of the language model score. The advantage of this metric is that it uses information already available during the decoder search. The performance of this guided search approach was tested in two experiments for the ATIS and WSJ data sets and results show significant reductions in recognition error rates.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume4
StatePublished - 2002
Event2002 IEEE International Conference on Acoustic, Speech, and Signal Processing - Orlando, FL, United States
Duration: May 13 2002May 17 2002

Other

Other2002 IEEE International Conference on Acoustic, Speech, and Signal Processing
CountryUnited States
CityOrlando, FL
Period5/13/025/17/02

Fingerprint

Information use
speech recognition
Speech recognition
confidence
decoders
phonemes
Experiments
output

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Abdou, S., & Scordilis, M. S. (2002). Confidence measures as a search guide in speech recognition. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 4)

Confidence measures as a search guide in speech recognition. / Abdou, Sherif; Scordilis, Michael S.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 4 2002.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abdou, S & Scordilis, MS 2002, Confidence measures as a search guide in speech recognition. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 4, 2002 IEEE International Conference on Acoustic, Speech, and Signal Processing, Orlando, FL, United States, 5/13/02.
Abdou S, Scordilis MS. Confidence measures as a search guide in speech recognition. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 4. 2002
Abdou, Sherif ; Scordilis, Michael S. / Confidence measures as a search guide in speech recognition. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 4 2002.
@inproceedings{ce0e5bc0b26f44018a27dd9d714a60f2,
title = "Confidence measures as a search guide in speech recognition",
abstract = "Recently, there have been many efforts in developing confidence measures for speech recognition output. Usually, these measures are applied to the final result of the decoder. However, using these measures early in the search process can guide the search to more promising paths. In this paper we use a confidence metric, the Average Best_Base_Phoneme Rank, to dynamically tune the contribution of the language model score. The advantage of this metric is that it uses information already available during the decoder search. The performance of this guided search approach was tested in two experiments for the ATIS and WSJ data sets and results show significant reductions in recognition error rates.",
author = "Sherif Abdou and Scordilis, {Michael S}",
year = "2002",
language = "English (US)",
volume = "4",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Confidence measures as a search guide in speech recognition

AU - Abdou, Sherif

AU - Scordilis, Michael S

PY - 2002

Y1 - 2002

N2 - Recently, there have been many efforts in developing confidence measures for speech recognition output. Usually, these measures are applied to the final result of the decoder. However, using these measures early in the search process can guide the search to more promising paths. In this paper we use a confidence metric, the Average Best_Base_Phoneme Rank, to dynamically tune the contribution of the language model score. The advantage of this metric is that it uses information already available during the decoder search. The performance of this guided search approach was tested in two experiments for the ATIS and WSJ data sets and results show significant reductions in recognition error rates.

AB - Recently, there have been many efforts in developing confidence measures for speech recognition output. Usually, these measures are applied to the final result of the decoder. However, using these measures early in the search process can guide the search to more promising paths. In this paper we use a confidence metric, the Average Best_Base_Phoneme Rank, to dynamically tune the contribution of the language model score. The advantage of this metric is that it uses information already available during the decoder search. The performance of this guided search approach was tested in two experiments for the ATIS and WSJ data sets and results show significant reductions in recognition error rates.

UR - http://www.scopus.com/inward/record.url?scp=0036288695&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036288695&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -