完整後設資料紀錄
DC 欄位語言
dc.contributor.authorChao, Yi-Hsiangen_US
dc.contributor.authorTsai, Wei-Hoen_US
dc.contributor.authorWang, Hsin-Minen_US
dc.contributor.authorChang, Ruei-Chuanen_US
dc.date.accessioned2014-12-08T15:10:45Z-
dc.date.available2014-12-08T15:10:45Z-
dc.date.issued2008-11-01en_US
dc.identifier.issn1558-7916en_US
dc.identifier.urihttp://dx.doi.org/10.1109/TASL.2008.2004297en_US
dc.identifier.urihttp://hdl.handle.net/11536/8229-
dc.description.abstractSpeaker verification can be viewed as a task of modeling and testing two hypotheses: the null hypothesis and the alternative hypothesis. Since the alternative hypothesis involves unknown impostors, it is usually hard to characterize a priori. In this paper, we propose improving the characterization of the alternative hypothesis by designing two decision functions based, respectively, on a weighted arithmetic combination and a weighted geometric combination of discriminative information derived from a set of pretrained background models. The parameters associated with the combinations are then optimized using two kernel discriminant analysis techniques, namely, the kernel Fisher discriminant (KFD) and support vector machine (SVM). The proposed approaches have two advantages over existing methods. The first is that they embed a trainable mechanism in the decision functions. The second is that they convert variable-length utterances into fixed-dimension characteristic vectors, which are easily processed by kernel discriminant analysis. The results of speaker-verification experiments conducted on two speech corpora show that the proposed methods outperform conventional likelihood ratio-based approaches.en_US
dc.language.isoen_USen_US
dc.subjectKernel Fisher Discriminant (KFD)en_US
dc.subjectlikelihood ratioen_US
dc.subjectspeaker verificationen_US
dc.subjectsupport vector machine (SVM)en_US
dc.titleUsing Kernel Discriminant Analysis to Improve the Characterization of the Alternative Hypothesis for Speaker Verificationen_US
dc.typeArticleen_US
dc.identifier.doi10.1109/TASL.2008.2004297en_US
dc.identifier.journalIEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSINGen_US
dc.citation.volume16en_US
dc.citation.issue8en_US
dc.citation.spage1675en_US
dc.citation.epage1684en_US
dc.contributor.department資訊工程學系zh_TW
dc.contributor.departmentDepartment of Computer Scienceen_US
dc.identifier.wosnumberWOS:000260463800027-
dc.citation.woscount2-
顯示於類別:期刊論文


文件中的檔案:

  1. 000260463800027.pdf

若為 zip 檔案,請下載檔案解壓縮後,用瀏覽器開啟資料夾中的 index.html 瀏覽全文。