Integrating acoustic and lexical features in topic segmentation of Chinese broadcast news using maximum entropy approach

Lei Xie, Yulian Yang, Zhi Qiang Liu, Wei Feng, Zihan Liu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

11 Scopus citations

Abstract

This paper studies how to integrate multi-modal features in automatic topic segmentation of Mandarin broadcast news. The multi-modal feature integration problem is formulated within the Maximum Entropy (MaxEnt) scheme for topic boundary classification by maximizing the entropy and respecting all known constraints (i.e., multiple features contributions). We particularly consider two types of features: (1) acoustic features, which reflect the editorial prosody of broadcast news, including pause duration, speaker change and speech type; and (2) lexical features extracted from speech recognition transcripts, which capture the semantic shifts of topics, including two local cohesiveness features and a new boundary indicator based on overall cohesiveness. Compared to local lexical features, the new overall cohesiveness feature maximizes the lexical cohesiveness of all topic fragments and reflects the fact that topic transitions in broadcast news are smooth and the distributional variations are subtle. Experiments show apparent performance improvement in topic segmentation of Chinese broadcast news by fusing acoustic and lexical features within the MaxEnt scheme.

Original languageEnglish
Title of host publicationICALIP 2010 - 2010 International Conference on Audio, Language and Image Processing, Proceedings
Pages407-413
Number of pages7
DOIs
StatePublished - 2010
Event2010 International Conference on Audio, Language and Image Processing, ICALIP 2010 - Shanghai, China
Duration: 23 Nov 201025 Nov 2010

Publication series

NameICALIP 2010 - 2010 International Conference on Audio, Language and Image Processing, Proceedings

Conference

Conference2010 International Conference on Audio, Language and Image Processing, ICALIP 2010
Country/TerritoryChina
CityShanghai
Period23/11/1025/11/10

Fingerprint

Dive into the research topics of 'Integrating acoustic and lexical features in topic segmentation of Chinese broadcast news using maximum entropy approach'. Together they form a unique fingerprint.

Cite this