The effectiveness of corpus-induced dependency grammars for post-processing speech

M. P. Harper; C. M. White; W. Wang; M. T. Johnson; R. A. Helzerman

The effectiveness of corpus-induced dependency grammars for post-processing speech

M. P. Harper, C. M. White, W. Wang, M. T. Johnson, R. A. Helzerman

Research output: Contribution to conference › Paper › peer-review

5 Citations (SciVal)

Abstract

This paper investigates the impact of Constraint Dependency Grammars (CDG) on the accuracy of an integrated speech recognition and CDG parsing system. We compare a conventional CDG with CDGs that are induced from annotated sentences and template-expanded sentences. The grammars are evaluated on parsing speed, precision/coverage, and improvement of word and sentence accuracy of the integrated system. Sentence-derived CDGs significantly improve recognition accuracy over the conventional CDG but are less general. Expanding the sentences with templates provides us with a mechanism for increasing the coverage of the grammar with only minor reductions in recognition accuracy.

Original language	English
Pages	102-109
Number of pages	8
State	Published - 2000
Event	1st Meeting of the North American Chapter of the Association for Computational Linguistics, NAACL 2000 - Seattle, United States Duration: Apr 29 2000 → May 4 2000

Conference

Conference	1st Meeting of the North American Chapter of the Association for Computational Linguistics, NAACL 2000
Country/Territory	United States
City	Seattle
Period	4/29/00 → 5/4/00

Bibliographical note

Funding Information:
This research was supported by grants from Intel, Purdue Research Foundation, and National Science Foundation IRI 97-04358, CDA 96-17388, and #9980054-BCS.

Publisher Copyright:
© ANLP 2000. All rights reserved.

ASJC Scopus subject areas

Language and Linguistics
Computer Science Applications
Linguistics and Language

Cite this

@conference{ec00d55ddfd244c5ada0ca20a8226af0,

title = "The effectiveness of corpus-induced dependency grammars for post-processing speech",

abstract = "This paper investigates the impact of Constraint Dependency Grammars (CDG) on the accuracy of an integrated speech recognition and CDG parsing system. We compare a conventional CDG with CDGs that are induced from annotated sentences and template-expanded sentences. The grammars are evaluated on parsing speed, precision/coverage, and improvement of word and sentence accuracy of the integrated system. Sentence-derived CDGs significantly improve recognition accuracy over the conventional CDG but are less general. Expanding the sentences with templates provides us with a mechanism for increasing the coverage of the grammar with only minor reductions in recognition accuracy.",

author = "Harper, {M. P.} and White, {C. M.} and W. Wang and Johnson, {M. T.} and Helzerman, {R. A.}",

note = "Funding Information: This research was supported by grants from Intel, Purdue Research Foundation, and National Science Foundation IRI 97-04358, CDA 96-17388, and #9980054-BCS. Publisher Copyright: {\textcopyright} ANLP 2000. All rights reserved.; 1st Meeting of the North American Chapter of the Association for Computational Linguistics, NAACL 2000 ; Conference date: 29-04-2000 Through 04-05-2000",

year = "2000",

language = "English",

pages = "102--109",

}

TY - CONF

T1 - The effectiveness of corpus-induced dependency grammars for post-processing speech

AU - Harper, M. P.

AU - White, C. M.

AU - Wang, W.

AU - Johnson, M. T.

AU - Helzerman, R. A.

N1 - Funding Information: This research was supported by grants from Intel, Purdue Research Foundation, and National Science Foundation IRI 97-04358, CDA 96-17388, and #9980054-BCS. Publisher Copyright: © ANLP 2000. All rights reserved.

PY - 2000

Y1 - 2000

N2 - This paper investigates the impact of Constraint Dependency Grammars (CDG) on the accuracy of an integrated speech recognition and CDG parsing system. We compare a conventional CDG with CDGs that are induced from annotated sentences and template-expanded sentences. The grammars are evaluated on parsing speed, precision/coverage, and improvement of word and sentence accuracy of the integrated system. Sentence-derived CDGs significantly improve recognition accuracy over the conventional CDG but are less general. Expanding the sentences with templates provides us with a mechanism for increasing the coverage of the grammar with only minor reductions in recognition accuracy.

AB - This paper investigates the impact of Constraint Dependency Grammars (CDG) on the accuracy of an integrated speech recognition and CDG parsing system. We compare a conventional CDG with CDGs that are induced from annotated sentences and template-expanded sentences. The grammars are evaluated on parsing speed, precision/coverage, and improvement of word and sentence accuracy of the integrated system. Sentence-derived CDGs significantly improve recognition accuracy over the conventional CDG but are less general. Expanding the sentences with templates provides us with a mechanism for increasing the coverage of the grammar with only minor reductions in recognition accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84871612068&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871612068&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84871612068

SP - 102

EP - 109

T2 - 1st Meeting of the North American Chapter of the Association for Computational Linguistics, NAACL 2000

Y2 - 29 April 2000 through 4 May 2000

ER -

The effectiveness of corpus-induced dependency grammars for post-processing speech

Abstract

Conference

Bibliographical note

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this