Abstract

Contrast is a very popular phenomenon in spoken language, and carries very important information to help understanding contents and structures of spoken language. In this paper, we propose an idea of automatic contrast detection as an effort for better speech understanding. We study the automatic tagging of three specific types of contrast: symmetric contrast, contrastive focus, and contrastive topic. We label the three types of contrasted words as contrast (C), and other words as noncontrast (C). The classification of contrast events is based on prosodic, spectral, and part-of-speech (POS) information sources. The integration of different knowledge sources is realized by a time-delay recursive neural network (TDRNN). The approach we proposed was testified on 235 spontaneous utterances consisting of 3500 words (samples). The contrast detection was speaker independent. The tests yielded an average of 87.9% classification rate.

Original languageEnglish (US)
Pages581-584
Number of pages4
StatePublished - Jan 1 2004
Event8th International Conference on Spoken Language Processing, ICSLP 2004 - Jeju, Jeju Island, Korea, Republic of
Duration: Oct 4 2004Oct 8 2004

Other

Other8th International Conference on Spoken Language Processing, ICSLP 2004
CountryKorea, Republic of
CityJeju, Jeju Island
Period10/4/0410/8/04

Fingerprint

spoken language
neural network
event
knowledge
Spoken Language
time

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Zhang, T., Hasegawa-Johnson, M. A., & Levinson, S. E. (2004). Automatic detection of contrast for speech understanding. 581-584. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.

Automatic detection of contrast for speech understanding. / Zhang, Tong; Hasegawa-Johnson, Mark Allan; Levinson, Stephen E.

2004. 581-584 Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.

Research output: Contribution to conferencePaper

Zhang, T, Hasegawa-Johnson, MA & Levinson, SE 2004, 'Automatic detection of contrast for speech understanding', Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of, 10/4/04 - 10/8/04 pp. 581-584.
Zhang T, Hasegawa-Johnson MA, Levinson SE. Automatic detection of contrast for speech understanding. 2004. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.
Zhang, Tong ; Hasegawa-Johnson, Mark Allan ; Levinson, Stephen E. / Automatic detection of contrast for speech understanding. Paper presented at 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.4 p.
@conference{7cffa14ef587431aa586d33af9cb6bf5,
title = "Automatic detection of contrast for speech understanding",
abstract = "Contrast is a very popular phenomenon in spoken language, and carries very important information to help understanding contents and structures of spoken language. In this paper, we propose an idea of automatic contrast detection as an effort for better speech understanding. We study the automatic tagging of three specific types of contrast: symmetric contrast, contrastive focus, and contrastive topic. We label the three types of contrasted words as contrast (C), and other words as noncontrast (C). The classification of contrast events is based on prosodic, spectral, and part-of-speech (POS) information sources. The integration of different knowledge sources is realized by a time-delay recursive neural network (TDRNN). The approach we proposed was testified on 235 spontaneous utterances consisting of 3500 words (samples). The contrast detection was speaker independent. The tests yielded an average of 87.9{\%} classification rate.",
author = "Tong Zhang and Hasegawa-Johnson, {Mark Allan} and Levinson, {Stephen E}",
year = "2004",
month = "1",
day = "1",
language = "English (US)",
pages = "581--584",
note = "8th International Conference on Spoken Language Processing, ICSLP 2004 ; Conference date: 04-10-2004 Through 08-10-2004",

}

TY - CONF

T1 - Automatic detection of contrast for speech understanding

AU - Zhang, Tong

AU - Hasegawa-Johnson, Mark Allan

AU - Levinson, Stephen E

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Contrast is a very popular phenomenon in spoken language, and carries very important information to help understanding contents and structures of spoken language. In this paper, we propose an idea of automatic contrast detection as an effort for better speech understanding. We study the automatic tagging of three specific types of contrast: symmetric contrast, contrastive focus, and contrastive topic. We label the three types of contrasted words as contrast (C), and other words as noncontrast (C). The classification of contrast events is based on prosodic, spectral, and part-of-speech (POS) information sources. The integration of different knowledge sources is realized by a time-delay recursive neural network (TDRNN). The approach we proposed was testified on 235 spontaneous utterances consisting of 3500 words (samples). The contrast detection was speaker independent. The tests yielded an average of 87.9% classification rate.

AB - Contrast is a very popular phenomenon in spoken language, and carries very important information to help understanding contents and structures of spoken language. In this paper, we propose an idea of automatic contrast detection as an effort for better speech understanding. We study the automatic tagging of three specific types of contrast: symmetric contrast, contrastive focus, and contrastive topic. We label the three types of contrasted words as contrast (C), and other words as noncontrast (C). The classification of contrast events is based on prosodic, spectral, and part-of-speech (POS) information sources. The integration of different knowledge sources is realized by a time-delay recursive neural network (TDRNN). The approach we proposed was testified on 235 spontaneous utterances consisting of 3500 words (samples). The contrast detection was speaker independent. The tests yielded an average of 87.9% classification rate.

UR - http://www.scopus.com/inward/record.url?scp=85009106568&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009106568&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85009106568

SP - 581

EP - 584

ER -