Evaluation of Bayes Decision Approach to Automatic Determination of Thresholds for Speaker Verification
Identifieur interne : 00CB90 ( Main/Merge ); précédent : 00CB89; suivant : 00CB91Evaluation of Bayes Decision Approach to Automatic Determination of Thresholds for Speaker Verification
Auteurs : Y. GongSource :
English descriptors
- KwdEn :
Abstract
Under Bayes statistical decision framework, this paper addresses statistical modelling and determination of thresholds for speaker verification systems. It is pointed out that speaker-dependent between-speaker score distribution is bi-modal, as opposed to common believe that the distribution is normal. Previous mono-modal modelling of between-speaker score distribution is then extended to bi-modal modelling. For a text-dependent application, experiments are reported which compare verification results with speaker-independent unique threshold, speaker-dependent mono-modal distributions and speaker-dependent bi-modal distributions. It is observed that speaker-dependent thresholds give dramatic error reduction, as compared to unique threshold and that bi-modal and mono-modal distribution models give very close verification results. For a 200 speaker database, using 1 sec of test speech, the resulting system resulted in a 0.65========percnt; mean verification error.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 001A27
- to stream Crin, to step Curation: 001A27
- to stream Crin, to step Checkpoint: 002B45
Links to Exploration step
CRIN:gong95cLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="503">Evaluation of Bayes Decision Approach to Automatic Determination of Thresholds for Speaker Verification</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:gong95c</idno>
<date when="1995" year="1995">1995</date>
<idno type="wicri:Area/Crin/Corpus">001A27</idno>
<idno type="wicri:Area/Crin/Curation">001A27</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">001A27</idno>
<idno type="wicri:Area/Crin/Checkpoint">002B45</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">002B45</idno>
<idno type="wicri:Area/Main/Merge">00CB90</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Evaluation of Bayes Decision Approach to Automatic Determination of Thresholds for Speaker Verification</title>
<author><name sortKey="Gong, Y" sort="Gong, Y" uniqKey="Gong Y" first="Y." last="Gong">Y. Gong</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>speech recognition</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="1099">Under Bayes statistical decision framework, this paper addresses statistical modelling and determination of thresholds for speaker verification systems. It is pointed out that speaker-dependent between-speaker score distribution is bi-modal, as opposed to common believe that the distribution is normal. Previous mono-modal modelling of between-speaker score distribution is then extended to bi-modal modelling. For a text-dependent application, experiments are reported which compare verification results with speaker-independent unique threshold, speaker-dependent mono-modal distributions and speaker-dependent bi-modal distributions. It is observed that speaker-dependent thresholds give dramatic error reduction, as compared to unique threshold and that bi-modal and mono-modal distribution models give very close verification results. For a 200 speaker database, using 1 sec of test speech, the resulting system resulted in a 0.65========percnt; mean verification error.</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 00CB90 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 00CB90 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Main |étape= Merge |type= RBID |clé= CRIN:gong95c |texte= Evaluation of Bayes Decision Approach to Automatic Determination of Thresholds for Speaker Verification }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |