Reconnaissance rapide de mots isolés par quantification vectorielle multisections
Identifieur interne : 00E506 ( Main/Exploration ); précédent : 00E505; suivant : 00E507Reconnaissance rapide de mots isolés par quantification vectorielle multisections
Auteurs : A. Gourinda ; Jean-Paul Haton [France]Source :
English descriptors
- KwdEn :
Abstract
In this paper we present a fast word recognition algorithm based on Multisection Vector Quantization. A separate multisection codebook information systems designed for each word in the vocabulary by dividing the word into equal-length sections and by designing a codebook for each section. Unknown words are also divided into equal-length sections, each section is averaged and encoded with the Multisection codebooks, for speaker-dependent recognition of the french digits plus the words "oui" and "non" this approach achieved a recognition accuracy greater than 99 percent with only one distortion computation per input section for each vocabulary word. We give a generalization of this algorithm to continuous speech recognition.
Affiliations:
- France
- Grand Est, Lorraine (région)
- Nancy
- Centre national de la recherche scientifique, Institut national de recherche en informatique et en automatique, Laboratoire lorrain de recherche en informatique et ses applications, Université de Lorraine
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 000481
- to stream Crin, to step Curation: 000481
- to stream Crin, to step Checkpoint: 004113
- to stream Main, to step Merge: 00ED92
- to stream Main, to step Curation: 00E506
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="224">Reconnaissance rapide de mots isolés par quantification vectorielle multisections</title></titleStmt><publicationStmt><idno type="RBID">CRIN:gourinda87a</idno><date when="1987" year="1987">1987</date><idno type="wicri:Area/Crin/Corpus">000481</idno><idno type="wicri:Area/Crin/Curation">000481</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">000481</idno><idno type="wicri:Area/Crin/Checkpoint">004113</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">004113</idno><idno type="wicri:Area/Main/Merge">00ED92</idno><idno type="wicri:Area/Main/Curation">00E506</idno><idno type="wicri:Area/Main/Exploration">00E506</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Reconnaissance rapide de mots isolés par quantification vectorielle multisections</title><author><name sortKey="Gourinda, A" sort="Gourinda, A" uniqKey="Gourinda A" first="A." last="Gourinda">A. Gourinda</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><term>vector quantization</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en" wicri:score="3529">In this paper we present a fast word recognition algorithm based on Multisection Vector Quantization. A separate multisection codebook information systems designed for each word in the vocabulary by dividing the word into equal-length sections and by designing a codebook for each section. Unknown words are also divided into equal-length sections, each section is averaged and encoded with the Multisection codebooks, for speaker-dependent recognition of the french digits plus the words "oui" and "non" this approach achieved a recognition accuracy greater than 99 percent with only one distortion computation per input section for each vocabulary word. We give a generalization of this algorithm to continuous speech recognition.</div></front></TEI><affiliations><list><country><li>France</li></country><region><li>Grand Est</li><li>Lorraine (région)</li></region><settlement><li>Nancy</li></settlement><orgName><li>Centre national de la recherche scientifique</li><li>Institut national de recherche en informatique et en automatique</li><li>Laboratoire lorrain de recherche en informatique et ses applications</li><li>Université de Lorraine</li></orgName></list><tree><noCountry><name sortKey="Gourinda, A" sort="Gourinda, A" uniqKey="Gourinda A" first="A." last="Gourinda">A. Gourinda</name></noCountry><country name="France"><region name="Grand Est"><name sortKey="Haton, J P" sort="Haton, J P" uniqKey="Haton J" first="J.-P." last="Haton">Jean-Paul Haton</name></region></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 00E506 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 00E506 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Main
|étape= Exploration
|type= RBID
|clé= CRIN:gourinda87a
|texte= Reconnaissance rapide de mots isolés par quantification vectorielle multisections
}}
| This area was generated with Dilib version V0.6.33. |  |