Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems

Kazunori Komatani; Satoshi Ikeda; Yuichiro Fukubayashi; Tetsuya Ogata; Hiroshi G. Okuno

doi:10.3115/1708376.1708422

Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems

Kazunori Komatani^*, Satoshi Ikeda, Yuichiro Fukubayashi, Tetsuya Ogata, Hiroshi G. Okuno

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

We address an issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages for novice users. Help generation for OOG utterances is a challenging problem because language understanding (LU) results based on automatic speech recognition (ASR) results for such utterances are always erroneous as important words are often misrecognized or missed from such utterances. We first develop grammar verification for OOG utterances on the basis of a Weighted Finite-State Transducer (WFST). It robustly identifies a grammar rule that a user intends to utter, even when some important words are missed from the ASR result. We then adopt a ranking algorithm, RankBoost, whose features include the grammar verification results and the utterance history representing the user's experience.

Original language	English
Title of host publication	Proceedings of the SIGDIAL 2009 Conference
Subtitle of host publication	10th Annual Meeting of the Special Interest Group on Discourse and Dialogue
Publisher	Association for Computational Linguistics (ACL)
Pages	314-321
Number of pages	8
ISBN (Print)	9781932432640
DOIs	https://doi.org/10.3115/1708376.1708422
Publication status	Published - 2009
Externally published	Yes
Event	10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2009 - London, United Kingdom Duration: 2009 Sept 11 → 2009 Sept 12

Publication series

Name	Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue

Conference

Conference	10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2009
Country/Territory	United Kingdom
City	London
Period	09/9/11 → 09/9/12

ASJC Scopus subject areas

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

Access to Document

10.3115/1708376.1708422

Cite this

Komatani, K., Ikeda, S., Fukubayashi, Y., Ogata, T., & Okuno, H. G. (2009). Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems. In Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue (pp. 314-321). (Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue). Association for Computational Linguistics (ACL). https://doi.org/10.3115/1708376.1708422

Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems. / Komatani, Kazunori; Ikeda, Satoshi; Fukubayashi, Yuichiro et al.
Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue. Association for Computational Linguistics (ACL), 2009. p. 314-321 (Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Komatani, K, Ikeda, S, Fukubayashi, Y, Ogata, T & Okuno, HG 2009, Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems. in Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue. Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, Association for Computational Linguistics (ACL), pp. 314-321, 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2009, London, United Kingdom, 09/9/11. https://doi.org/10.3115/1708376.1708422

Komatani K, Ikeda S, Fukubayashi Y, Ogata T, Okuno HG. Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems. In Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue. Association for Computational Linguistics (ACL). 2009. p. 314-321. (Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue). doi: 10.3115/1708376.1708422

Komatani, Kazunori ; Ikeda, Satoshi ; Fukubayashi, Yuichiro et al. / Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems. Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue. Association for Computational Linguistics (ACL), 2009. pp. 314-321 (Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue).

@inproceedings{93145bb79bcc4a1aa6b2fc3d5457a707,

title = "Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems",

abstract = "We address an issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages for novice users. Help generation for OOG utterances is a challenging problem because language understanding (LU) results based on automatic speech recognition (ASR) results for such utterances are always erroneous as important words are often misrecognized or missed from such utterances. We first develop grammar verification for OOG utterances on the basis of a Weighted Finite-State Transducer (WFST). It robustly identifies a grammar rule that a user intends to utter, even when some important words are missed from the ASR result. We then adopt a ranking algorithm, RankBoost, whose features include the grammar verification results and the utterance history representing the user's experience.",

author = "Kazunori Komatani and Satoshi Ikeda and Yuichiro Fukubayashi and Tetsuya Ogata and Okuno, {Hiroshi G.}",

year = "2009",

doi = "10.3115/1708376.1708422",

language = "English",

isbn = "9781932432640",

series = "Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue",

publisher = "Association for Computational Linguistics (ACL)",

pages = "314--321",

booktitle = "Proceedings of the SIGDIAL 2009 Conference",

note = "10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2009 ; Conference date: 11-09-2009 Through 12-09-2009",

}

TY - GEN

T1 - Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems

AU - Komatani, Kazunori

AU - Ikeda, Satoshi

AU - Fukubayashi, Yuichiro

AU - Ogata, Tetsuya

AU - Okuno, Hiroshi G.

PY - 2009

Y1 - 2009

N2 - We address an issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages for novice users. Help generation for OOG utterances is a challenging problem because language understanding (LU) results based on automatic speech recognition (ASR) results for such utterances are always erroneous as important words are often misrecognized or missed from such utterances. We first develop grammar verification for OOG utterances on the basis of a Weighted Finite-State Transducer (WFST). It robustly identifies a grammar rule that a user intends to utter, even when some important words are missed from the ASR result. We then adopt a ranking algorithm, RankBoost, whose features include the grammar verification results and the utterance history representing the user's experience.

AB - We address an issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages for novice users. Help generation for OOG utterances is a challenging problem because language understanding (LU) results based on automatic speech recognition (ASR) results for such utterances are always erroneous as important words are often misrecognized or missed from such utterances. We first develop grammar verification for OOG utterances on the basis of a Weighted Finite-State Transducer (WFST). It robustly identifies a grammar rule that a user intends to utter, even when some important words are missed from the ASR result. We then adopt a ranking algorithm, RankBoost, whose features include the grammar verification results and the utterance history representing the user's experience.

UR - http://www.scopus.com/inward/record.url?scp=78650023182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650023182&partnerID=8YFLogxK

U2 - 10.3115/1708376.1708422

DO - 10.3115/1708376.1708422

M3 - Conference contribution

AN - SCOPUS:78650023182

SN - 9781932432640

T3 - Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue

SP - 314

EP - 321

BT - Proceedings of the SIGDIAL 2009 Conference

PB - Association for Computational Linguistics (ACL)

T2 - 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL 2009

Y2 - 11 September 2009 through 12 September 2009

ER -

Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this