Show simple item record

dc.contributor.authorChristensen, Heidi
dc.contributor.authorKolluru, BalaKrishna
dc.contributor.authorGotoh, Yoshihiko
dc.contributor.authorRenals, Steve
dc.date.accessioned2006-05-09T14:36:48Z
dc.date.available2006-05-09T14:36:48Z
dc.date.issued2005
dc.identifier.citationIn Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP-05), Philadelphia, PA, USA, March 2005.en
dc.identifier.urihttp://hdl.handle.net/1842/948
dc.description.abstractThis paper presents an automatic system for structuring and preparing a news broadcast for applications such as speech summarization, browsing, archiving and information retrieval. This process comprises transcribing the audio using an automatic speech recognizer and subsequently segmenting the text into utterances and topics. A maximum entropy approach is used to build statistical models for both utterance and topic segmentation. The experimental work addresses the effect on performance of the topic boundary detector of three factors: the information sources used, the quality of the ASR transcripts, and the quality of the utterance boundary detector. The results show that the topic segmentation is not affected severely by transcripts errors, whereas errors in the utterance segmentation are more devastating.en
dc.format.extent84359 bytes
dc.format.extent117020 bytes
dc.format.mimetypeapplication/octet-stream
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherIEEE Signal Processing Society Press.en
dc.titleMaximum entropy segmentation of broadcast newsen
dc.typeConference Paperen


Files in this item

This item appears in the following Collection(s)

Show simple item record