We address the issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages. Help message generation for OOG utterances is a challenge because language understanding based on automatic speech recognition (ASR) of OOG utterances is usually erroneous; important words are often misrecognized or missing from such utterances. Our grammar verification method uses a weighted finite-state transducer, to accurately identify the grammar rule that the user intended to use for the utterance, even if important words are missing from the ASR results. We then use a ranking algorithm, RankBoost, to rank help message candidates in order of likely usefulness. Its features include the grammar verification results and the utterance history representing the user's experience.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Kazunori KOMATANI, Yuichiro FUKUBAYASHI, Satoshi IKEDA, Tetsuya OGATA, Hiroshi G. OKUNO, "Selecting Help Messages by Using Robust Grammar Verification for Handling Out-of-Grammar Utterances in Spoken Dialogue Systems" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 12, pp. 3359-3367, December 2010, doi: 10.1587/transinf.E93.D.3359.
Abstract: We address the issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages. Help message generation for OOG utterances is a challenge because language understanding based on automatic speech recognition (ASR) of OOG utterances is usually erroneous; important words are often misrecognized or missing from such utterances. Our grammar verification method uses a weighted finite-state transducer, to accurately identify the grammar rule that the user intended to use for the utterance, even if important words are missing from the ASR results. We then use a ranking algorithm, RankBoost, to rank help message candidates in order of likely usefulness. Its features include the grammar verification results and the utterance history representing the user's experience.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.3359/_p
Copy
@ARTICLE{e93-d_12_3359,
author={Kazunori KOMATANI, Yuichiro FUKUBAYASHI, Satoshi IKEDA, Tetsuya OGATA, Hiroshi G. OKUNO, },
journal={IEICE TRANSACTIONS on Information},
title={Selecting Help Messages by Using Robust Grammar Verification for Handling Out-of-Grammar Utterances in Spoken Dialogue Systems},
year={2010},
volume={E93-D},
number={12},
pages={3359-3367},
abstract={We address the issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages. Help message generation for OOG utterances is a challenge because language understanding based on automatic speech recognition (ASR) of OOG utterances is usually erroneous; important words are often misrecognized or missing from such utterances. Our grammar verification method uses a weighted finite-state transducer, to accurately identify the grammar rule that the user intended to use for the utterance, even if important words are missing from the ASR results. We then use a ranking algorithm, RankBoost, to rank help message candidates in order of likely usefulness. Its features include the grammar verification results and the utterance history representing the user's experience.},
keywords={},
doi={10.1587/transinf.E93.D.3359},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Selecting Help Messages by Using Robust Grammar Verification for Handling Out-of-Grammar Utterances in Spoken Dialogue Systems
T2 - IEICE TRANSACTIONS on Information
SP - 3359
EP - 3367
AU - Kazunori KOMATANI
AU - Yuichiro FUKUBAYASHI
AU - Satoshi IKEDA
AU - Tetsuya OGATA
AU - Hiroshi G. OKUNO
PY - 2010
DO - 10.1587/transinf.E93.D.3359
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2010
AB - We address the issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages. Help message generation for OOG utterances is a challenge because language understanding based on automatic speech recognition (ASR) of OOG utterances is usually erroneous; important words are often misrecognized or missing from such utterances. Our grammar verification method uses a weighted finite-state transducer, to accurately identify the grammar rule that the user intended to use for the utterance, even if important words are missing from the ASR results. We then use a ranking algorithm, RankBoost, to rank help message candidates in order of likely usefulness. Its features include the grammar verification results and the utterance history representing the user's experience.
ER -