Skip navigation
Skip navigation

Background Specificity in Forensic Voice Comparison and Its Relation to the Bayesian Prior Probability

Wagner, Michael; Kinoshita, Yuko

Description

This study investigates the effect of background data specifici-ty on likelihood ratio and prior odds, and consequently on the posterior odds outcome. It is motivated by discussions on the correct choice of speaker recognition background, particularly in forensic voice comparison. We performed strictly controlled experiments with the ANDOSL database where background specificity is the sole independent variable. Results show that target and non-target scores are better separated with less...[Show more]

dc.contributor.advisorTyler, Michael D.
dc.contributor.authorWagner, Michael
dc.contributor.authorKinoshita, Yuko
dc.contributor.editorCarignan, Christopher
dc.coverage.spatialParramatta, Australia
dc.date.accessioned2019-12-04T04:38:34Z
dc.date.available2019-12-04T04:38:34Z
dc.date.createdDecember 6-9 2016
dc.identifier.isbn2207-1296
dc.identifier.urihttp://hdl.handle.net/1885/187303
dc.description.abstractThis study investigates the effect of background data specifici-ty on likelihood ratio and prior odds, and consequently on the posterior odds outcome. It is motivated by discussions on the correct choice of speaker recognition background, particularly in forensic voice comparison. We performed strictly controlled experiments with the ANDOSL database where background specificity is the sole independent variable. Results show that target and non-target scores are better separated with less spe-cific background, but that in turn priors must be adjusted down. Because the risk of class recognition instead of individ-ual recognition increases with lower background specificity, we suggest that the prior probability in the Bayes formula is factorised into one part that remains in the domain of the trier of fact – as is conventional – and another part that is related to the specificity of the assumed or agreed background.
dc.format.mimetypeapplication/pdf
dc.language.isoen_AU
dc.publisherThe Australasian Speech Science and Technology Association, Inc.
dc.relation.ispartofseriesSixteenth Australasian International Conference on Speech Science and Technology
dc.rights© 2016 ASSTA
dc.sourceProceedings of the Sixteenth Australasian International Conference on Speech Science and Technology
dc.source.urihttps://assta.org/wp-content/uploads/2019/09/SST2016_Proceedings.pdf
dc.titleBackground Specificity in Forensic Voice Comparison and Its Relation to the Bayesian Prior Probability
dc.typeConference paper
local.description.notesImported from ARIES
local.description.refereedYes
dc.date.issued2016
local.identifier.absfor200404 - Laboratory Phonetics and Speech Science
local.identifier.absfor090609 - Signal Processing
local.identifier.ariespublicationu5786633xPUB187
local.publisher.urlhttp://www.assta.org/
local.type.statusPublished Version
local.contributor.affiliationWagner, Michael, College of Engineering and Computer Science, ANU
local.contributor.affiliationKinoshita, Yuko, College of Asia and the Pacific, ANU
local.description.embargo2099-12-31
local.bibliographicCitation.startpage353
local.bibliographicCitation.lastpage356
local.identifier.absseo970120 - Expanding Knowledge in Languages, Communication and Culture
local.identifier.absseo970108 - Expanding Knowledge in the Information and Computing Sciences
local.identifier.absseo810107 - National Security
dc.date.updated2019-07-14T08:16:39Z
dcterms.accessRightsOpen Access via publisher website
CollectionsANU Research Publications

Download

File Description SizeFormat Image
Background Specificity in Forensic Voice Comparison and Its Relation to the Bayesian Prior Probability.pdf998.7 kBAdobe PDF    Request a copy


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  17 November 2022/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator