Extractive broadcast news summarization leveraging recurrent neural network language modeling techniques

Kuan Yu Chen, Shih Hung Liu, Berlin Chen, Hsin Min Wang, Ea Ee Jan, Wen Lian Hsu, Hsin Hsi Chen

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Extractive text or speech summarization manages to select a set of salient sentences from an original document and concatenate them to form a summary, enabling users to better browse through and understand the content of the document. A recent stream of research on extractive summarization is to employ the language modeling (LM) approach for important sentence selection, which has proven to be effective for performing speech summarization in an unsupervised fashion. However, one of the major challenges facing the LM approach is how to formulate the sentence models and accurately estimate their parameters for each sentence in the document to be summarized. In view of this, our work in this paper explores a novel use of recurrent neural network language modeling (RNNLM) framework for extractive broadcast news summarization. On top of such a framework, the deduced sentence models are able to render not only word usage cues but also long-span structural information of word co-occurrence relationships within broadcast news documents, getting around the need for the strict bag-of-words assumption. Furthermore, different model complexities and combinations are extensively analyzed and compared. Experimental results demonstrate the performance merits of our summarization methods when compared to several well-studied state-of-the-art unsupervised methods.

Original languageEnglish
Article number7111264
Pages (from-to)1322-1334
Number of pages13
JournalIEEE Transactions on Audio, Speech and Language Processing
Volume23
Issue number8
DOIs
Publication statusPublished - 2015 Aug 1

Fingerprint

sentences
news
Recurrent neural networks
cues
bags
occurrences
estimates

Keywords

  • Language modeling
  • long-span structural information
  • recurrent neural network
  • speech summarization

ASJC Scopus subject areas

  • Acoustics and Ultrasonics
  • Electrical and Electronic Engineering

Cite this

Extractive broadcast news summarization leveraging recurrent neural network language modeling techniques. / Chen, Kuan Yu; Liu, Shih Hung; Chen, Berlin; Wang, Hsin Min; Jan, Ea Ee; Hsu, Wen Lian; Chen, Hsin Hsi.

In: IEEE Transactions on Audio, Speech and Language Processing, Vol. 23, No. 8, 7111264, 01.08.2015, p. 1322-1334.

Research output: Contribution to journalArticle

Chen, Kuan Yu ; Liu, Shih Hung ; Chen, Berlin ; Wang, Hsin Min ; Jan, Ea Ee ; Hsu, Wen Lian ; Chen, Hsin Hsi. / Extractive broadcast news summarization leveraging recurrent neural network language modeling techniques. In: IEEE Transactions on Audio, Speech and Language Processing. 2015 ; Vol. 23, No. 8. pp. 1322-1334.
@article{313c921d645d42e9b12fb35361e267b9,
title = "Extractive broadcast news summarization leveraging recurrent neural network language modeling techniques",
abstract = "Extractive text or speech summarization manages to select a set of salient sentences from an original document and concatenate them to form a summary, enabling users to better browse through and understand the content of the document. A recent stream of research on extractive summarization is to employ the language modeling (LM) approach for important sentence selection, which has proven to be effective for performing speech summarization in an unsupervised fashion. However, one of the major challenges facing the LM approach is how to formulate the sentence models and accurately estimate their parameters for each sentence in the document to be summarized. In view of this, our work in this paper explores a novel use of recurrent neural network language modeling (RNNLM) framework for extractive broadcast news summarization. On top of such a framework, the deduced sentence models are able to render not only word usage cues but also long-span structural information of word co-occurrence relationships within broadcast news documents, getting around the need for the strict bag-of-words assumption. Furthermore, different model complexities and combinations are extensively analyzed and compared. Experimental results demonstrate the performance merits of our summarization methods when compared to several well-studied state-of-the-art unsupervised methods.",
keywords = "Language modeling, long-span structural information, recurrent neural network, speech summarization",
author = "Chen, {Kuan Yu} and Liu, {Shih Hung} and Berlin Chen and Wang, {Hsin Min} and Jan, {Ea Ee} and Hsu, {Wen Lian} and Chen, {Hsin Hsi}",
year = "2015",
month = "8",
day = "1",
doi = "10.1109/TASLP.2015.2432578",
language = "English",
volume = "23",
pages = "1322--1334",
journal = "IEEE Transactions on Audio, Speech and Language Processing",
issn = "1558-7916",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - Extractive broadcast news summarization leveraging recurrent neural network language modeling techniques

AU - Chen, Kuan Yu

AU - Liu, Shih Hung

AU - Chen, Berlin

AU - Wang, Hsin Min

AU - Jan, Ea Ee

AU - Hsu, Wen Lian

AU - Chen, Hsin Hsi

PY - 2015/8/1

Y1 - 2015/8/1

N2 - Extractive text or speech summarization manages to select a set of salient sentences from an original document and concatenate them to form a summary, enabling users to better browse through and understand the content of the document. A recent stream of research on extractive summarization is to employ the language modeling (LM) approach for important sentence selection, which has proven to be effective for performing speech summarization in an unsupervised fashion. However, one of the major challenges facing the LM approach is how to formulate the sentence models and accurately estimate their parameters for each sentence in the document to be summarized. In view of this, our work in this paper explores a novel use of recurrent neural network language modeling (RNNLM) framework for extractive broadcast news summarization. On top of such a framework, the deduced sentence models are able to render not only word usage cues but also long-span structural information of word co-occurrence relationships within broadcast news documents, getting around the need for the strict bag-of-words assumption. Furthermore, different model complexities and combinations are extensively analyzed and compared. Experimental results demonstrate the performance merits of our summarization methods when compared to several well-studied state-of-the-art unsupervised methods.

AB - Extractive text or speech summarization manages to select a set of salient sentences from an original document and concatenate them to form a summary, enabling users to better browse through and understand the content of the document. A recent stream of research on extractive summarization is to employ the language modeling (LM) approach for important sentence selection, which has proven to be effective for performing speech summarization in an unsupervised fashion. However, one of the major challenges facing the LM approach is how to formulate the sentence models and accurately estimate their parameters for each sentence in the document to be summarized. In view of this, our work in this paper explores a novel use of recurrent neural network language modeling (RNNLM) framework for extractive broadcast news summarization. On top of such a framework, the deduced sentence models are able to render not only word usage cues but also long-span structural information of word co-occurrence relationships within broadcast news documents, getting around the need for the strict bag-of-words assumption. Furthermore, different model complexities and combinations are extensively analyzed and compared. Experimental results demonstrate the performance merits of our summarization methods when compared to several well-studied state-of-the-art unsupervised methods.

KW - Language modeling

KW - long-span structural information

KW - recurrent neural network

KW - speech summarization

UR - http://www.scopus.com/inward/record.url?scp=84930943980&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930943980&partnerID=8YFLogxK

U2 - 10.1109/TASLP.2015.2432578

DO - 10.1109/TASLP.2015.2432578

M3 - Article

AN - SCOPUS:84930943980

VL - 23

SP - 1322

EP - 1334

JO - IEEE Transactions on Audio, Speech and Language Processing

JF - IEEE Transactions on Audio, Speech and Language Processing

SN - 1558-7916

IS - 8

M1 - 7111264

ER -