Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system

Kazunori Komatani, Hiroshi G. Okuno

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

2 Citations (Scopus)

Abstract

We develop a method to detect erroneous interpretation results of user utterances by exploiting utterance histories of individual users in spoken dialogue systems that were deployed for the general public and repeatedly utilized. More specifically, we classify barge-in utterances into correctly and erroneously interpreted ones by using features of individual users' utterance histories such as their barge-in rates and estimated automatic speech recognition (ASR) accuracies. Online detection is enabled by making these features obtainable without any manual annotation or labeling. We experimentally compare classification accuracies for several cases when an ASR confidence measure is used alone or in combination with the features based on the user's utterance history. The error reduction rate was 15% when the utterance history was used.

Original languageEnglish
Title of host publicationProceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue
Pages289-296
Number of pages8
Publication statusPublished - 2010
Externally publishedYes
Event11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2010 - Tokyo
Duration: 2010 Sep 242010 Sep 25

Other

Other11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2010
CityTokyo
Period10/9/2410/9/25

Fingerprint

Spoken Dialogue Systems
Barges
Error Detection
Error detection
Speech recognition
Automatic Speech Recognition
Labeling
Confidence Measure
Error Reduction
Annotation
Classify
History

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction
  • Modelling and Simulation

Cite this

Komatani, K., & Okuno, H. G. (2010). Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system. In Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue (pp. 289-296)

Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system. / Komatani, Kazunori; Okuno, Hiroshi G.

Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue. 2010. p. 289-296.

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

Komatani, K & Okuno, HG 2010, Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system. in Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue. pp. 289-296, 11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2010, Tokyo, 10/9/24.
Komatani K, Okuno HG. Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system. In Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue. 2010. p. 289-296
Komatani, Kazunori ; Okuno, Hiroshi G. / Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system. Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue. 2010. pp. 289-296
@inproceedings{039e964571044743bd212a7ec08aae78,
title = "Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system",
abstract = "We develop a method to detect erroneous interpretation results of user utterances by exploiting utterance histories of individual users in spoken dialogue systems that were deployed for the general public and repeatedly utilized. More specifically, we classify barge-in utterances into correctly and erroneously interpreted ones by using features of individual users' utterance histories such as their barge-in rates and estimated automatic speech recognition (ASR) accuracies. Online detection is enabled by making these features obtainable without any manual annotation or labeling. We experimentally compare classification accuracies for several cases when an ASR confidence measure is used alone or in combination with the features based on the user's utterance history. The error reduction rate was 15{\%} when the utterance history was used.",
author = "Kazunori Komatani and Okuno, {Hiroshi G.}",
year = "2010",
language = "English",
isbn = "9781932432855",
pages = "289--296",
booktitle = "Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue",

}

TY - GEN

T1 - Online error detection of barge-in utterances by using individual users' utterance histories in spoken dialogue system

AU - Komatani, Kazunori

AU - Okuno, Hiroshi G.

PY - 2010

Y1 - 2010

N2 - We develop a method to detect erroneous interpretation results of user utterances by exploiting utterance histories of individual users in spoken dialogue systems that were deployed for the general public and repeatedly utilized. More specifically, we classify barge-in utterances into correctly and erroneously interpreted ones by using features of individual users' utterance histories such as their barge-in rates and estimated automatic speech recognition (ASR) accuracies. Online detection is enabled by making these features obtainable without any manual annotation or labeling. We experimentally compare classification accuracies for several cases when an ASR confidence measure is used alone or in combination with the features based on the user's utterance history. The error reduction rate was 15% when the utterance history was used.

AB - We develop a method to detect erroneous interpretation results of user utterances by exploiting utterance histories of individual users in spoken dialogue systems that were deployed for the general public and repeatedly utilized. More specifically, we classify barge-in utterances into correctly and erroneously interpreted ones by using features of individual users' utterance histories such as their barge-in rates and estimated automatic speech recognition (ASR) accuracies. Online detection is enabled by making these features obtainable without any manual annotation or labeling. We experimentally compare classification accuracies for several cases when an ASR confidence measure is used alone or in combination with the features based on the user's utterance history. The error reduction rate was 15% when the utterance history was used.

UR - http://www.scopus.com/inward/record.url?scp=84857752153&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857752153&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84857752153

SN - 9781932432855

SP - 289

EP - 296

BT - Proceedings of the SIGDIAL 2010 Conference: 11th Annual Meeting of the Special Interest Group onDiscourse and Dialogue

ER -