Detection of Target Speakers in Audio Databases

Simple item page
dc.citation.bibtexNameinproceedingsen_US
dc.citation.conferenceNameIEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)en_US
dc.contributor.authorMagrin-Chagnolleau, Ivanen_US
dc.contributor.authorRosenberg, Aaronen_US
dc.contributor.authorParthasarathy, S.en_US
dc.contributor.orgDigital Signal Processing (http://dsp.rice.edu/)en_US
dc.date.accessioned2007-10-31T00:52:29Z
dc.date.available2007-10-31T00:52:29Z
dc.date.issued1999-01-15en
dc.date.modified2004-11-05en_US
dc.date.note2004-01-14en_US
dc.date.submitted1999-01-15en_US
dc.descriptionConference Paperen_US
dc.description.abstractThe problem of speaker detection in audio databases is addressed in this paper. Gaussian mixture modeling is used to build target speaker and background models. A detection algorithm based on a likelihood ratio calculation is applied to estimate target speaker segments. Evaluation procedures are defined in detail for this task. Results are given for different subsets of the HUB4 broadcast news database. For one target speaker, with the data restricted to high quality speech segments, the segment miss rate is approximately 7%. For unrestricted data, the segment miss rate is approximately 27%. In both cases the segment false alarm rate is 4 or 5 per hour. For two target speakers with unrestricted data, the segment miss rate is approximately 63% with about 27 segment false alarms per hour. The decrease in performance for two target speakers is largely associated with short speech segments in the two target speaker test data which are undetectable in the current configuration of the detection algorithm.en_US
dc.identifier.citationI. Magrin-Chagnolleau, A. Rosenberg and S. Parthasarathy, "Detection of Target Speakers in Audio Databases," 1999.
dc.identifier.doihttp://dx.doi.org/10.1109/ICASSP.1999.759797en_US
dc.identifier.urihttps://hdl.handle.net/1911/20077
dc.language.isoeng
dc.subjectTemporary*
dc.subject.keywordTemporaryen_US
dc.subject.otherSignal Processing Applicationsen_US
dc.titleDetection of Target Speakers in Audio Databasesen_US
dc.typeConference paper
dc.type.dcmiText
Files
Original bundle
Now showing 1 - 3 of 3
Thumbnail Image
Name:
Mag1999Non5Detection.PDF
Size:
57.04 KB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
Mag1999Non5Detection.PPT
Size:
98.5 KB
Format:
Microsoft Powerpoint
Download
No Thumbnail Available
Name:
Mag1999Non5Detection.PS
Size:
86.38 KB
Format:
Postscript Files
Download
Collections
ECE Publications
DSP Publications