Reliable utterance segment recognition by integrating a grammar with statistical language constraints

Hajime Tsukada, Hirofumi Yamamoto, Toshiyuki Takezawa, Yoshinori Sagisaka

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper proposes a novel approach to the recognition of complete utterances and partial segments of utterances. This approach ensures a high level of confidence in the results. The proposed method is based on the cooperative use of a conventional n-gram constraint and additional grammatical constraints which take deviations from the grammar into account with a multi-pass search strategy. The partial utterance segments are obtained with high confidence as the segments that satisfy both n-gram and grammatical constraints. For improved efficiency, the context-free grammar expressing the grammatical constraints is approximated by a finite-state automaton. We consider all kinds of deviations from the grammar such as insertions, deletions and substitutions when applying the grammatical constraints. As a result, we can achieve a more robust application of grammatical constraints compared to a conventional word-skipping robust parser that can only handle one type of deviation, that is, insertions. Our experiments confirm that the proposed method can recognize partial segments of utterances more reliably than conventional continuous speech recognition methods using only n-grams. In addition, our results indicate that allowing more deviations from the grammatical constraints leads to better performance than the conventional word-skipping robust parser approach.

Original languageEnglish
Pages (from-to)299-309
Number of pages11
JournalSpeech Communication
Volume26
Issue number4
Publication statusPublished - 1998 Dec
Externally publishedYes

Fingerprint

Continuous speech recognition
Context free grammars
Finite automata
Grammar
grammar
Substitution reactions
Language
N-gram
language
confidence
Deviation
Experiments
substitution
Partial
Insertion
Confidence
efficiency
experiment
Finite State Automata
Context-free Grammar

Keywords

  • Context-free grammar
  • Finite-state automaton
  • Finite-state transducer
  • Language model
  • Robust parsing
  • Speech recognition

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Experimental and Cognitive Psychology
  • Linguistics and Language

Cite this

Reliable utterance segment recognition by integrating a grammar with statistical language constraints. / Tsukada, Hajime; Yamamoto, Hirofumi; Takezawa, Toshiyuki; Sagisaka, Yoshinori.

In: Speech Communication, Vol. 26, No. 4, 12.1998, p. 299-309.

Research output: Contribution to journalArticle

Tsukada, Hajime ; Yamamoto, Hirofumi ; Takezawa, Toshiyuki ; Sagisaka, Yoshinori. / Reliable utterance segment recognition by integrating a grammar with statistical language constraints. In: Speech Communication. 1998 ; Vol. 26, No. 4. pp. 299-309.
@article{8c9e2a87cfd04f7b9956b8723af4c083,
title = "Reliable utterance segment recognition by integrating a grammar with statistical language constraints",
abstract = "This paper proposes a novel approach to the recognition of complete utterances and partial segments of utterances. This approach ensures a high level of confidence in the results. The proposed method is based on the cooperative use of a conventional n-gram constraint and additional grammatical constraints which take deviations from the grammar into account with a multi-pass search strategy. The partial utterance segments are obtained with high confidence as the segments that satisfy both n-gram and grammatical constraints. For improved efficiency, the context-free grammar expressing the grammatical constraints is approximated by a finite-state automaton. We consider all kinds of deviations from the grammar such as insertions, deletions and substitutions when applying the grammatical constraints. As a result, we can achieve a more robust application of grammatical constraints compared to a conventional word-skipping robust parser that can only handle one type of deviation, that is, insertions. Our experiments confirm that the proposed method can recognize partial segments of utterances more reliably than conventional continuous speech recognition methods using only n-grams. In addition, our results indicate that allowing more deviations from the grammatical constraints leads to better performance than the conventional word-skipping robust parser approach.",
keywords = "Context-free grammar, Finite-state automaton, Finite-state transducer, Language model, Robust parsing, Speech recognition",
author = "Hajime Tsukada and Hirofumi Yamamoto and Toshiyuki Takezawa and Yoshinori Sagisaka",
year = "1998",
month = "12",
language = "English",
volume = "26",
pages = "299--309",
journal = "Speech Communication",
issn = "0167-6393",
publisher = "Elsevier",
number = "4",

}

TY - JOUR

T1 - Reliable utterance segment recognition by integrating a grammar with statistical language constraints

AU - Tsukada, Hajime

AU - Yamamoto, Hirofumi

AU - Takezawa, Toshiyuki

AU - Sagisaka, Yoshinori

PY - 1998/12

Y1 - 1998/12

N2 - This paper proposes a novel approach to the recognition of complete utterances and partial segments of utterances. This approach ensures a high level of confidence in the results. The proposed method is based on the cooperative use of a conventional n-gram constraint and additional grammatical constraints which take deviations from the grammar into account with a multi-pass search strategy. The partial utterance segments are obtained with high confidence as the segments that satisfy both n-gram and grammatical constraints. For improved efficiency, the context-free grammar expressing the grammatical constraints is approximated by a finite-state automaton. We consider all kinds of deviations from the grammar such as insertions, deletions and substitutions when applying the grammatical constraints. As a result, we can achieve a more robust application of grammatical constraints compared to a conventional word-skipping robust parser that can only handle one type of deviation, that is, insertions. Our experiments confirm that the proposed method can recognize partial segments of utterances more reliably than conventional continuous speech recognition methods using only n-grams. In addition, our results indicate that allowing more deviations from the grammatical constraints leads to better performance than the conventional word-skipping robust parser approach.

AB - This paper proposes a novel approach to the recognition of complete utterances and partial segments of utterances. This approach ensures a high level of confidence in the results. The proposed method is based on the cooperative use of a conventional n-gram constraint and additional grammatical constraints which take deviations from the grammar into account with a multi-pass search strategy. The partial utterance segments are obtained with high confidence as the segments that satisfy both n-gram and grammatical constraints. For improved efficiency, the context-free grammar expressing the grammatical constraints is approximated by a finite-state automaton. We consider all kinds of deviations from the grammar such as insertions, deletions and substitutions when applying the grammatical constraints. As a result, we can achieve a more robust application of grammatical constraints compared to a conventional word-skipping robust parser that can only handle one type of deviation, that is, insertions. Our experiments confirm that the proposed method can recognize partial segments of utterances more reliably than conventional continuous speech recognition methods using only n-grams. In addition, our results indicate that allowing more deviations from the grammatical constraints leads to better performance than the conventional word-skipping robust parser approach.

KW - Context-free grammar

KW - Finite-state automaton

KW - Finite-state transducer

KW - Language model

KW - Robust parsing

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=0032288381&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032288381&partnerID=8YFLogxK

M3 - Article

VL - 26

SP - 299

EP - 309

JO - Speech Communication

JF - Speech Communication

SN - 0167-6393

IS - 4

ER -