Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription
Identifieur interne : 001639 ( Main/Exploration ); précédent : 001638; suivant : 001640Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription
Auteurs : Denis Jouvet [France] ; Dominique Fohr [France]Source :
- Lecture Notes in Computer Science [ 0302-9743 ]
Abstract
Abstract: This paper analysis the behavior of forward and backward-based decoders used for speech transcription. Experiments have showed that backward-based decoding leads to similar recognition performance as forward-based decoding, which is consistent with the fact that both systems handle similar information through the acoustic, lexical and language models. However, because of heuristics, search algorithms used in decoding explore only a limited portion of the search space. As forward-based and backward-based approaches do not process the speech signal in the same temporal way, they explore different portions of the search space; leading to complementary systems that can be efficiently combined using the ROVER approach. The speech transcription results achieved by combining forward-based and backward-based systems are significantly better than the results obtained by combining the same amount of forward-only or backward-only systems. This confirms the complementary of the forward and backward approaches and thus the usefulness of their combination.
Url:
DOI: 10.1007/978-3-642-40585-3_12
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000005
- to stream Istex, to step Curation: 000005
- to stream Istex, to step Checkpoint: 000265
- to stream Main, to step Merge: 001651
- to stream Main, to step Curation: 001639
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription</title>
<author><name sortKey="Jouvet, Denis" sort="Jouvet, Denis" uniqKey="Jouvet D" first="Denis" last="Jouvet">Denis Jouvet</name>
</author>
<author><name sortKey="Fohr, Dominique" sort="Fohr, Dominique" uniqKey="Fohr D" first="Dominique" last="Fohr">Dominique Fohr</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:003535B66680484637167A5E4F2D2EB92E16FD02</idno>
<date when="2013" year="2013">2013</date>
<idno type="doi">10.1007/978-3-642-40585-3_12</idno>
<idno type="url">https://api.istex.fr/ark:/67375/HCB-0DH79VJ9-3/fulltext.pdf</idno>
<idno type="wicri:Area/Istex/Corpus">000005</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000005</idno>
<idno type="wicri:Area/Istex/Curation">000005</idno>
<idno type="wicri:Area/Istex/Checkpoint">000265</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000265</idno>
<idno type="wicri:doubleKey">0302-9743:2013:Jouvet D:analysis:and:combination</idno>
<idno type="wicri:Area/Main/Merge">001651</idno>
<idno type="wicri:Area/Main/Curation">001639</idno>
<idno type="wicri:Area/Main/Exploration">001639</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription</title>
<author><name sortKey="Jouvet, Denis" sort="Jouvet, Denis" uniqKey="Jouvet D" first="Denis" last="Jouvet">Denis Jouvet</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Speech Group, LORIA Inria, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="4"><country xml:lang="fr">France</country>
<wicri:regionArea>Université de Lorraine, LORIA, UMR 7503, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
<orgName type="university">Université de Lorraine</orgName>
</affiliation>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>CNRS, LORIA, UMR 7503, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Fohr, Dominique" sort="Fohr, Dominique" uniqKey="Fohr D" first="Dominique" last="Fohr">Dominique Fohr</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>Speech Group, LORIA Inria, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="4"><country xml:lang="fr">France</country>
<wicri:regionArea>Université de Lorraine, LORIA, UMR 7503, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
<orgName type="university">Université de Lorraine</orgName>
</affiliation>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>CNRS, LORIA, UMR 7503, F-54600, Villers-lès-Nancy</wicri:regionArea>
<placeName><region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
<settlement type="city">Villers-lès-Nancy</settlement>
</placeName>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s" type="main" xml:lang="en">Lecture Notes in Computer Science</title>
<idno type="ISSN">0302-9743</idno>
<idno type="eISSN">1611-3349</idno>
<idno type="ISSN">0302-9743</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: This paper analysis the behavior of forward and backward-based decoders used for speech transcription. Experiments have showed that backward-based decoding leads to similar recognition performance as forward-based decoding, which is consistent with the fact that both systems handle similar information through the acoustic, lexical and language models. However, because of heuristics, search algorithms used in decoding explore only a limited portion of the search space. As forward-based and backward-based approaches do not process the speech signal in the same temporal way, they explore different portions of the search space; leading to complementary systems that can be efficiently combined using the ROVER approach. The speech transcription results achieved by combining forward-based and backward-based systems are significantly better than the results obtained by combining the same amount of forward-only or backward-only systems. This confirms the complementary of the forward and backward approaches and thus the usefulness of their combination.</div>
</front>
</TEI>
<affiliations><list><country><li>France</li>
</country>
<region><li>Grand Est</li>
<li>Lorraine (région)</li>
</region>
<settlement><li>Villers-lès-Nancy</li>
</settlement>
<orgName><li>Université de Lorraine</li>
</orgName>
</list>
<tree><country name="France"><region name="Grand Est"><name sortKey="Jouvet, Denis" sort="Jouvet, Denis" uniqKey="Jouvet D" first="Denis" last="Jouvet">Denis Jouvet</name>
</region>
<name sortKey="Fohr, Dominique" sort="Fohr, Dominique" uniqKey="Fohr D" first="Dominique" last="Fohr">Dominique Fohr</name>
<name sortKey="Fohr, Dominique" sort="Fohr, Dominique" uniqKey="Fohr D" first="Dominique" last="Fohr">Dominique Fohr</name>
<name sortKey="Fohr, Dominique" sort="Fohr, Dominique" uniqKey="Fohr D" first="Dominique" last="Fohr">Dominique Fohr</name>
<name sortKey="Jouvet, Denis" sort="Jouvet, Denis" uniqKey="Jouvet D" first="Denis" last="Jouvet">Denis Jouvet</name>
<name sortKey="Jouvet, Denis" sort="Jouvet, Denis" uniqKey="Jouvet D" first="Denis" last="Jouvet">Denis Jouvet</name>
</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 001639 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001639 | 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= ISTEX:003535B66680484637167A5E4F2D2EB92E16FD02 |texte= Analysis and Combination of Forward and Backward Based Decoders for Improved Speech Transcription }}
This area was generated with Dilib version V0.6.33. |