Disfluency correction of spontaneous speech using conditional random fields with variable-length features

Jui Feng Yeh; Chung Hsien Wu; Wei Yen Wu

Disfluency correction of spontaneous speech using conditional random fields with variable-length features

Jui Feng Yeh, Chung Hsien Wu, Wei Yen Wu

Department of Computer Science and Information Engineering

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

2 Citations (Scopus)

Abstract

This paper presents an approach to detecting and correcting edit disfluency based on conditional random fields with variable-length features. The variable-length features consist of word, chunk and sentence features. Conditional random fields (CRF) are adopted to model the properties of the edit disfluency, including repair, repetition and restart, for edit disfluency detection. For the evaluation of the proposed method, Mandarin conversational dialogue corpus (MCDC) is used. The detection error rate of edit word is 17.3%. Compared with DF-gram, Maximum Entropy and the approach combining language model and alignment model, the proposed approach achieves 11.7%, 8% and 3.9% improvements, respectively. The experimental results show that the proposed model outperforms other methods and efficiently detects and corrects edit disfluency in spontaneous speech.

Original language	English
Title of host publication	International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007
Pages	549-552
Number of pages	4
Publication status	Published - 2007 Dec 1
Event	8th Annual Conference of the International Speech Communication Association, Interspeech 2007 - Antwerp, Belgium Duration: 2007 Aug 27 → 2007 Aug 31

Publication series

Name	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume	1
ISSN (Electronic)	1990-9772

Other

Other	8th Annual Conference of the International Speech Communication Association, Interspeech 2007
Country	Belgium
City	Antwerp
Period	07-08-27 → 07-08-31

All Science Journal Classification (ASJC) codes

Computer Science Applications
Software
Modelling and Simulation
Linguistics and Language
Communication

Fingerprint Dive into the research topics of 'Disfluency correction of spontaneous speech using conditional random fields with variable-length features'. Together they form a unique fingerprint.

Cite this

Yeh, J. F., Wu, C. H., & Wu, W. Y. (2007). Disfluency correction of spontaneous speech using conditional random fields with variable-length features. In International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007 (pp. 549-552). (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH; Vol. 1).

Yeh, Jui Feng ; Wu, Chung Hsien ; Wu, Wei Yen. / Disfluency correction of spontaneous speech using conditional random fields with variable-length features. International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007. 2007. pp. 549-552 (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH).

@inproceedings{f5f74ef5b61a489491aed8957e02d588,

title = "Disfluency correction of spontaneous speech using conditional random fields with variable-length features",

abstract = "This paper presents an approach to detecting and correcting edit disfluency based on conditional random fields with variable-length features. The variable-length features consist of word, chunk and sentence features. Conditional random fields (CRF) are adopted to model the properties of the edit disfluency, including repair, repetition and restart, for edit disfluency detection. For the evaluation of the proposed method, Mandarin conversational dialogue corpus (MCDC) is used. The detection error rate of edit word is 17.3%. Compared with DF-gram, Maximum Entropy and the approach combining language model and alignment model, the proposed approach achieves 11.7%, 8% and 3.9% improvements, respectively. The experimental results show that the proposed model outperforms other methods and efficiently detects and corrects edit disfluency in spontaneous speech.",

author = "Yeh, {Jui Feng} and Wu, {Chung Hsien} and Wu, {Wei Yen}",

year = "2007",

month = dec,

day = "1",

language = "English",

isbn = "9781605603162",

series = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",

pages = "549--552",

booktitle = "International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007",

note = "8th Annual Conference of the International Speech Communication Association, Interspeech 2007 ; Conference date: 27-08-2007 Through 31-08-2007",

}

Yeh, JF, Wu, CH & Wu, WY 2007, Disfluency correction of spontaneous speech using conditional random fields with variable-length features. in International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, vol. 1, pp. 549-552, 8th Annual Conference of the International Speech Communication Association, Interspeech 2007, Antwerp, Belgium, 07-08-27.

Disfluency correction of spontaneous speech using conditional random fields with variable-length features. / Yeh, Jui Feng; Wu, Chung Hsien; Wu, Wei Yen.

International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007. 2007. p. 549-552 (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH; Vol. 1).

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

TY - GEN

T1 - Disfluency correction of spontaneous speech using conditional random fields with variable-length features

AU - Yeh, Jui Feng

AU - Wu, Chung Hsien

AU - Wu, Wei Yen

PY - 2007/12/1

Y1 - 2007/12/1

N2 - This paper presents an approach to detecting and correcting edit disfluency based on conditional random fields with variable-length features. The variable-length features consist of word, chunk and sentence features. Conditional random fields (CRF) are adopted to model the properties of the edit disfluency, including repair, repetition and restart, for edit disfluency detection. For the evaluation of the proposed method, Mandarin conversational dialogue corpus (MCDC) is used. The detection error rate of edit word is 17.3%. Compared with DF-gram, Maximum Entropy and the approach combining language model and alignment model, the proposed approach achieves 11.7%, 8% and 3.9% improvements, respectively. The experimental results show that the proposed model outperforms other methods and efficiently detects and corrects edit disfluency in spontaneous speech.

AB - This paper presents an approach to detecting and correcting edit disfluency based on conditional random fields with variable-length features. The variable-length features consist of word, chunk and sentence features. Conditional random fields (CRF) are adopted to model the properties of the edit disfluency, including repair, repetition and restart, for edit disfluency detection. For the evaluation of the proposed method, Mandarin conversational dialogue corpus (MCDC) is used. The detection error rate of edit word is 17.3%. Compared with DF-gram, Maximum Entropy and the approach combining language model and alignment model, the proposed approach achieves 11.7%, 8% and 3.9% improvements, respectively. The experimental results show that the proposed model outperforms other methods and efficiently detects and corrects edit disfluency in spontaneous speech.

UR - http://www.scopus.com/inward/record.url?scp=56149085175&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=56149085175&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:56149085175

SN - 9781605603162

T3 - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

SP - 549

EP - 552

BT - International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007

T2 - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007

Y2 - 27 August 2007 through 31 August 2007

ER -

Yeh JF, Wu CH, Wu WY. Disfluency correction of spontaneous speech using conditional random fields with variable-length features. In International Speech Communication Association - 8th Annual Conference of the International Speech Communication Association, Interspeech 2007. 2007. p. 549-552. (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH).