On-line spotting of ASR errors by means of prosody

J. Hirschberg; D. Litman; M.G.J. Swerts

On-line spotting of ASR errors by means of prosody

J. Hirschberg
, D. Litman
, M.G.J. Swerts

Research output: Contribution to journal › Article › Academic › peer-review

1 Downloads (Pure)

Abstract

Given the state of the art of current speech technology, errors are unavoidable in present spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. We identify methods to distinguish between correctly and incorrectly recognized utterances, scored by hand for semantic concept accuracy, using acoustic/prosodic characteristics. The analysis was performed on data collected during independent experiments done with an interactive voice response system that provides travel information over the phone.

Original language	English
Pages (from-to)	142-149
Number of pages	8
Journal	IPO Annual Progress Report
Volume	34
Publication status	Published - 1999

Cite this

@article{197868f41f9d4fd7ab4862040e5f97cf,

title = "On-line spotting of ASR errors by means of prosody",

abstract = "Given the state of the art of current speech technology, errors are unavoidable in present spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. We identify methods to distinguish between correctly and incorrectly recognized utterances, scored by hand for semantic concept accuracy, using acoustic/prosodic characteristics. The analysis was performed on data collected during independent experiments done with an interactive voice response system that provides travel information over the phone.",

author = "J. Hirschberg and D. Litman and M.G.J. Swerts",

year = "1999",

language = "English",

volume = "34",

pages = "142--149",

journal = "IPO Annual Progress Report",

issn = "0921-2566",

}

TY - JOUR

T1 - On-line spotting of ASR errors by means of prosody

AU - Hirschberg, J.

AU - Litman, D.

AU - Swerts, M.G.J.

PY - 1999

Y1 - 1999

N2 - Given the state of the art of current speech technology, errors are unavoidable in present spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. We identify methods to distinguish between correctly and incorrectly recognized utterances, scored by hand for semantic concept accuracy, using acoustic/prosodic characteristics. The analysis was performed on data collected during independent experiments done with an interactive voice response system that provides travel information over the phone.

AB - Given the state of the art of current speech technology, errors are unavoidable in present spoken dialogue systems. Therefore, one of the main concerns in dialogue design is how to decide whether or not the system has understood the user correctly. We identify methods to distinguish between correctly and incorrectly recognized utterances, scored by hand for semantic concept accuracy, using acoustic/prosodic characteristics. The analysis was performed on data collected during independent experiments done with an interactive voice response system that provides travel information over the phone.

M3 - Article

SN - 0921-2566

VL - 34

SP - 142

EP - 149

JO - IPO Annual Progress Report

JF - IPO Annual Progress Report

ER -

On-line spotting of ASR errors by means of prosody

Abstract

Fingerprint

Cite this