Hybrids of supervised and unsupervised models for extractive speech summarization

Shih Hsiang Lin; Yueng Tien Lo; Yao Ming Yeh; Berlin Chen

Hybrids of supervised and unsupervised models for extractive speech summarization

Shih Hsiang Lin, Yueng Tien Lo, Yao Ming Yeh, Berlin Chen

Department of Computer Science and Information Engineering

Research output: Contribution to journal › Conference article › peer-review

4 Citations (Scopus)

Abstract

Speech summarization, distilling important information and removing redundant and incorrect information from spoken documents, has become an active area of intensive research in the recent past. In this paper, we consider hybrids of supervised and unsupervised models for extractive speech summarization. Moreover, we investigate the use of the unsupervised summarizer to improve the performance of the supervised summarizer when manual labels are not available for training the latter. A novel training data selection and relabeling approach designed to leverage the inter-document or/and the inter-sentence similarity information is explored as well. Encouraging results were initially demonstrated.

Original language	English
Pages (from-to)	1507-1510
Number of pages	4
Journal	Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Publication status	Published - 2009 Nov 26
Event	10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009 - Brighton, United Kingdom Duration: 2009 Sep 6 → 2009 Sep 10

Keywords

Hybrid summarizer
Speech summarization
Unsupervised training

ASJC Scopus subject areas

Human-Computer Interaction
Signal Processing
Software
Sensory Systems

Cite this

@article{cb511afaef1748878f3dff21c0ada3ff,

title = "Hybrids of supervised and unsupervised models for extractive speech summarization",

abstract = "Speech summarization, distilling important information and removing redundant and incorrect information from spoken documents, has become an active area of intensive research in the recent past. In this paper, we consider hybrids of supervised and unsupervised models for extractive speech summarization. Moreover, we investigate the use of the unsupervised summarizer to improve the performance of the supervised summarizer when manual labels are not available for training the latter. A novel training data selection and relabeling approach designed to leverage the inter-document or/and the inter-sentence similarity information is explored as well. Encouraging results were initially demonstrated.",

keywords = "Hybrid summarizer, Speech summarization, Unsupervised training",

author = "Lin, {Shih Hsiang} and Lo, {Yueng Tien} and Yeh, {Yao Ming} and Berlin Chen",

year = "2009",

month = nov,

day = "26",

language = "English",

pages = "1507--1510",

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

issn = "2308-457X",

note = "10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009 ; Conference date: 06-09-2009 Through 10-09-2009",

}

TY - JOUR

T1 - Hybrids of supervised and unsupervised models for extractive speech summarization

AU - Lin, Shih Hsiang

AU - Lo, Yueng Tien

AU - Yeh, Yao Ming

AU - Chen, Berlin

PY - 2009/11/26

Y1 - 2009/11/26

N2 - Speech summarization, distilling important information and removing redundant and incorrect information from spoken documents, has become an active area of intensive research in the recent past. In this paper, we consider hybrids of supervised and unsupervised models for extractive speech summarization. Moreover, we investigate the use of the unsupervised summarizer to improve the performance of the supervised summarizer when manual labels are not available for training the latter. A novel training data selection and relabeling approach designed to leverage the inter-document or/and the inter-sentence similarity information is explored as well. Encouraging results were initially demonstrated.

AB - Speech summarization, distilling important information and removing redundant and incorrect information from spoken documents, has become an active area of intensive research in the recent past. In this paper, we consider hybrids of supervised and unsupervised models for extractive speech summarization. Moreover, we investigate the use of the unsupervised summarizer to improve the performance of the supervised summarizer when manual labels are not available for training the latter. A novel training data selection and relabeling approach designed to leverage the inter-document or/and the inter-sentence similarity information is explored as well. Encouraging results were initially demonstrated.

KW - Hybrid summarizer

KW - Speech summarization

KW - Unsupervised training

UR - http://www.scopus.com/inward/record.url?scp=70450172177&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450172177&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:70450172177

SP - 1507

EP - 1510

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

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

SN - 2308-457X

T2 - 10th Annual Conference of the International Speech Communication Association, INTERSPEECH 2009

Y2 - 6 September 2009 through 10 September 2009

ER -

Hybrids of supervised and unsupervised models for extractive speech summarization

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this