Keyword Spotting Based on STM Continuous Speech Recognition System
Identifieur interne : 00C344 ( Main/Curation ); précédent : 00C343; suivant : 00C345Keyword Spotting Based on STM Continuous Speech Recognition System
Auteurs : C.-T. Guan ; Y. Gong ; Y.-Q. Fu ; Jean-Paul Haton [France]Source :
English descriptors
- KwdEn :
Abstract
A keyword spotting system using the VINICS continuous speech recognition system based upon the recently proposed Stochastic Trajectory Modeling (STM) is presented. This keyword spotting system is a two pass classifier. The first component is a continuous speech recognizer serving as a word spotter using STM phonemic models. The second component is a postprocessor which rejects non-keywords in the framework of Linear Discrimination (LD) classification. Segmental probability scores of the spotter are also used and can improve the overall performances. Experiments were carried out on a speaker-dependent French database. An average FOM of 77.25========percnt; was obtained. This keyword spotting system is essentially vocabulary-independent and context-independent. New keywords can be added into the vocabulary in a very easy and flexible way.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: Pour aller vers cette notice dans l'étape Curation :001A25
- to stream Crin, to step Curation: Pour aller vers cette notice dans l'étape Curation :001A25
- to stream Crin, to step Checkpoint: Pour aller vers cette notice dans l'étape Curation :002B56
- to stream Main, to step Merge: Pour aller vers cette notice dans l'étape Curation :00CC01
Links to Exploration step
CRIN:guan95aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="466">Keyword Spotting Based on STM Continuous Speech Recognition System</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:guan95a</idno>
<date when="1995" year="1995">1995</date>
<idno type="wicri:Area/Crin/Corpus">001A25</idno>
<idno type="wicri:Area/Crin/Curation">001A25</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">001A25</idno>
<idno type="wicri:Area/Crin/Checkpoint">002B56</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">002B56</idno>
<idno type="wicri:Area/Main/Merge">00CC01</idno>
<idno type="wicri:Area/Main/Curation">00C344</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Keyword Spotting Based on STM Continuous Speech Recognition System</title>
<author><name sortKey="Guan, C T" sort="Guan, C T" uniqKey="Guan C" first="C.-T." last="Guan">C.-T. Guan</name>
</author>
<author><name sortKey="Gong, Y" sort="Gong, Y" uniqKey="Gong Y" first="Y." last="Gong">Y. Gong</name>
</author>
<author><name sortKey="Fu, Y Q" sort="Fu, Y Q" uniqKey="Fu Y" first="Y.-Q." last="Fu">Y.-Q. Fu</name>
</author>
<author><name sortKey="Haton, J P" sort="Haton, J P" uniqKey="Haton J" first="J.-P." last="Haton">Jean-Paul Haton</name>
<affiliation><country>France</country>
<placeName><settlement type="city">Nancy</settlement>
<region type="region" nuts="2">Grand Est</region>
<region type="region" nuts="2">Lorraine (région)</region>
</placeName>
<orgName type="laboratoire" n="5">Laboratoire lorrain de recherche en informatique et ses applications</orgName>
<orgName type="university">Université de Lorraine</orgName>
<orgName type="institution">Centre national de la recherche scientifique</orgName>
<orgName type="institution">Institut national de recherche en informatique et en automatique</orgName>
</affiliation>
</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="1890">A keyword spotting system using the VINICS continuous speech recognition system based upon the recently proposed Stochastic Trajectory Modeling (STM) is presented. This keyword spotting system is a two pass classifier. The first component is a continuous speech recognizer serving as a word spotter using STM phonemic models. The second component is a postprocessor which rejects non-keywords in the framework of Linear Discrimination (LD) classification. Segmental probability scores of the spotter are also used and can improve the overall performances. Experiments were carried out on a speaker-dependent French database. An average FOM of 77.25========percnt; was obtained. This keyword spotting system is essentially vocabulary-independent and context-independent. New keywords can be added into the vocabulary in a very easy and flexible way.</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 00C344 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Curation/biblio.hfd -nk 00C344 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Main |étape= Curation |type= RBID |clé= CRIN:guan95a |texte= Keyword Spotting Based on STM Continuous Speech Recognition System }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |