InforLorV4, Main, Exploration, bibRecord, 000949

An investigation of likelihood normalization for robust ASR

Identifieur interne : 000949 ( Main/Exploration ); précédent : 000948; suivant : 000950

An investigation of likelihood normalization for robust ASR

Auteurs : Emmanuel Vincent [France] ; Aggelos Gkiokas [Grèce] ; Dominik Schnitzer [Autriche] ; Arthur Flexer [Autriche]

Source :

RBID : Hal:hal-01006142

Abstract

Noise-robust automatic speech recognition (ASR) systems rely on feature and/or model compensation. Existing compensation techniques typically operate on the features or on the parameters of the acoustic models themselves. By contrast, a number of normalization techniques have been defined in the field of speaker verification that operate on the resulting log-likelihood scores. In this paper, we provide a theoretical motivation for likelihood normalization due to the so-called "hubness" phenomenon and we evaluate the benefit of several normalization techniques on ASR accuracy for the 2nd CHiME Challenge task. We show that symmetric normalization (S-norm) reduces the relative error rate by 43% alone and by 10% after feature and model compensation.

Url:

https://hal.inria.fr/hal-01006142

Affiliations:

Links toward previous steps (curation, corpus...)

to stream Hal, to step Corpus: 000E16
to stream Hal, to step Curation: 000E16
to stream Hal, to step Checkpoint: 000884
to stream Main, to step Merge: 000950
to stream Main, to step Curation: 000949

Le document en format XML

<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">An investigation of likelihood normalization for robust ASR</title>
<author><name sortKey="Vincent, Emmanuel" sort="Vincent, Emmanuel" uniqKey="Vincent E" first="Emmanuel" last="Vincent">Emmanuel Vincent</name>
<affiliation wicri:level="1"><hal:affiliation type="researchteam" xml:id="struct-205127" status="OLD"><idno type="RNSR">200118295L</idno>
<orgName>Analysis, perception and recognition of speech</orgName>
<orgName type="acronym">PAROLE</orgName>
<date type="end">2014-06-30</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/equipes/parole</ref>
</desc>
<listRelation><relation active="#struct-129671" type="direct"></relation>
<relation active="#struct-300009" type="indirect"></relation>
<relation active="#struct-423086" type="direct"></relation>
<relation active="#struct-206040" type="indirect"></relation>
<relation active="#struct-413289" type="indirect"></relation>
<relation name="UMR7503" active="#struct-441569" type="indirect"></relation>
</listRelation>
<tutelles><tutelle active="#struct-129671" type="direct"><org type="laboratory" xml:id="struct-129671" status="VALID"><idno type="RNSR">198618246Y</idno>
<orgName>INRIA Nancy - Grand Est</orgName>
<desc><address><addrLine>615 rue du Jardin Botanique 54600 Villers-lès-Nancy</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/nancy</ref>
</desc>
<listRelation><relation active="#struct-300009" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-300009" type="indirect"><org type="institution" xml:id="struct-300009" status="VALID"><orgName>Institut National de Recherche en Informatique et en Automatique</orgName>
<orgName type="acronym">Inria</orgName>
<desc><address><addrLine>Domaine de VoluceauRocquencourt - BP 10578153 Le Chesnay Cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/en/</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-423086" type="direct"><org type="department" xml:id="struct-423086" status="VALID"><orgName>Department of Natural Language Processing & Knowledge Discovery</orgName>
<orgName type="acronym">LORIA - NLPKD</orgName>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.loria.fr/la-recherche-en/departements/Knowledge-and-Language-Management</ref>
</desc>
<listRelation><relation active="#struct-206040" type="direct"></relation>
<relation active="#struct-300009" type="indirect"></relation>
<relation active="#struct-413289" type="indirect"></relation>
<relation name="UMR7503" active="#struct-441569" type="indirect"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-206040" type="indirect"><org type="laboratory" xml:id="struct-206040" status="VALID"><idno type="IdRef">067077927</idno>
<idno type="RNSR">198912571S</idno>
<idno type="IdUnivLorraine">[UL]RSI--</idno>
<orgName>Laboratoire Lorrain de Recherche en Informatique et ses Applications</orgName>
<orgName type="acronym">LORIA</orgName>
<date type="start">2012-01-01</date>
<desc><address><addrLine>Campus Scientifique BP 239 54506 Vandoeuvre-lès-Nancy Cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.loria.fr</ref>
</desc>
<listRelation><relation active="#struct-300009" type="direct"></relation>
<relation active="#struct-413289" type="direct"></relation>
<relation name="UMR7503" active="#struct-441569" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-413289" type="indirect"><org type="institution" xml:id="struct-413289" status="VALID"><idno type="IdRef">157040569</idno>
<idno type="IdUnivLorraine">[UL]100--</idno>
<orgName>Université de Lorraine</orgName>
<orgName type="acronym">UL</orgName>
<date type="start">2012-01-01</date>
<desc><address><addrLine>34 cours Léopold - CS 25233 - 54052 Nancy cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lorraine.fr/</ref>
</desc>
</org>
</tutelle>
<tutelle name="UMR7503" active="#struct-441569" type="indirect"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="ISNI">0000000122597504</idno>
<idno type="IdRef">02636817X</idno>
<orgName>Centre National de la Recherche Scientifique</orgName>
<orgName type="acronym">CNRS</orgName>
<date type="start">1939-10-19</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.cnrs.fr/</ref>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
<placeName><settlement type="city">Nancy</settlement>
<settlement type="city">Metz</settlement>
<region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
</placeName>
<orgName type="university">Université de Lorraine</orgName>
</affiliation>
</author>
<author><name sortKey="Gkiokas, Aggelos" sort="Gkiokas, Aggelos" uniqKey="Gkiokas A" first="Aggelos" last="Gkiokas">Aggelos Gkiokas</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-110738" status="VALID"><orgName>Institute for Language and Speech Processing</orgName>
<orgName type="acronym">ILSP</orgName>
<desc><address><addrLine>Institute for Language and Speech Processing / R.C. "Athena" Artemidos 6 & Epidavrou GR-151 25 MAROYSSI Greece</addrLine>
<country key="GR"></country>
</address>
<ref type="url">http://www.ilsp.gr/</ref>
</desc>
<listRelation><relation active="#struct-317224" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-317224" type="direct"><org type="institution" xml:id="struct-317224" status="INCOMING"><orgName>Institute for Language and Speech Processing</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Grèce</country>
</affiliation>
</author>
<author><name sortKey="Schnitzer, Dominik" sort="Schnitzer, Dominik" uniqKey="Schnitzer D" first="Dominik" last="Schnitzer">Dominik Schnitzer</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-93952" status="VALID"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<orgName type="acronym">OFAI</orgName>
<desc><address><addrLine>Freyung 6/6, A-1010 Vienna Austria</addrLine>
<country key="AT"></country>
</address>
<ref type="url">http://www.ofai.at/</ref>
</desc>
<listRelation><relation active="#struct-323078" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-323078" type="direct"><org type="institution" xml:id="struct-323078" status="INCOMING"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Autriche</country>
</affiliation>
</author>
<author><name sortKey="Flexer, Arthur" sort="Flexer, Arthur" uniqKey="Flexer A" first="Arthur" last="Flexer">Arthur Flexer</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-93952" status="VALID"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<orgName type="acronym">OFAI</orgName>
<desc><address><addrLine>Freyung 6/6, A-1010 Vienna Austria</addrLine>
<country key="AT"></country>
</address>
<ref type="url">http://www.ofai.at/</ref>
</desc>
<listRelation><relation active="#struct-323078" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-323078" type="direct"><org type="institution" xml:id="struct-323078" status="INCOMING"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Autriche</country>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">HAL</idno>
<idno type="RBID">Hal:hal-01006142</idno>
<idno type="halId">hal-01006142</idno>
<idno type="halUri">https://hal.inria.fr/hal-01006142</idno>
<idno type="url">https://hal.inria.fr/hal-01006142</idno>
<date when="2014-09-14">2014-09-14</date>
<idno type="wicri:Area/Hal/Corpus">000E16</idno>
<idno type="wicri:Area/Hal/Curation">000E16</idno>
<idno type="wicri:Area/Hal/Checkpoint">000884</idno>
<idno type="wicri:explorRef" wicri:stream="Hal" wicri:step="Checkpoint">000884</idno>
<idno type="wicri:Area/Main/Merge">000950</idno>
<idno type="wicri:Area/Main/Curation">000949</idno>
<idno type="wicri:Area/Main/Exploration">000949</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">An investigation of likelihood normalization for robust ASR</title>
<author><name sortKey="Vincent, Emmanuel" sort="Vincent, Emmanuel" uniqKey="Vincent E" first="Emmanuel" last="Vincent">Emmanuel Vincent</name>
<affiliation wicri:level="1"><hal:affiliation type="researchteam" xml:id="struct-205127" status="OLD"><idno type="RNSR">200118295L</idno>
<orgName>Analysis, perception and recognition of speech</orgName>
<orgName type="acronym">PAROLE</orgName>
<date type="end">2014-06-30</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/equipes/parole</ref>
</desc>
<listRelation><relation active="#struct-129671" type="direct"></relation>
<relation active="#struct-300009" type="indirect"></relation>
<relation active="#struct-423086" type="direct"></relation>
<relation active="#struct-206040" type="indirect"></relation>
<relation active="#struct-413289" type="indirect"></relation>
<relation name="UMR7503" active="#struct-441569" type="indirect"></relation>
</listRelation>
<tutelles><tutelle active="#struct-129671" type="direct"><org type="laboratory" xml:id="struct-129671" status="VALID"><idno type="RNSR">198618246Y</idno>
<orgName>INRIA Nancy - Grand Est</orgName>
<desc><address><addrLine>615 rue du Jardin Botanique 54600 Villers-lès-Nancy</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/nancy</ref>
</desc>
<listRelation><relation active="#struct-300009" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-300009" type="indirect"><org type="institution" xml:id="struct-300009" status="VALID"><orgName>Institut National de Recherche en Informatique et en Automatique</orgName>
<orgName type="acronym">Inria</orgName>
<desc><address><addrLine>Domaine de VoluceauRocquencourt - BP 10578153 Le Chesnay Cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.inria.fr/en/</ref>
</desc>
</org>
</tutelle>
<tutelle active="#struct-423086" type="direct"><org type="department" xml:id="struct-423086" status="VALID"><orgName>Department of Natural Language Processing & Knowledge Discovery</orgName>
<orgName type="acronym">LORIA - NLPKD</orgName>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.loria.fr/la-recherche-en/departements/Knowledge-and-Language-Management</ref>
</desc>
<listRelation><relation active="#struct-206040" type="direct"></relation>
<relation active="#struct-300009" type="indirect"></relation>
<relation active="#struct-413289" type="indirect"></relation>
<relation name="UMR7503" active="#struct-441569" type="indirect"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-206040" type="indirect"><org type="laboratory" xml:id="struct-206040" status="VALID"><idno type="IdRef">067077927</idno>
<idno type="RNSR">198912571S</idno>
<idno type="IdUnivLorraine">[UL]RSI--</idno>
<orgName>Laboratoire Lorrain de Recherche en Informatique et ses Applications</orgName>
<orgName type="acronym">LORIA</orgName>
<date type="start">2012-01-01</date>
<desc><address><addrLine>Campus Scientifique BP 239 54506 Vandoeuvre-lès-Nancy Cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.loria.fr</ref>
</desc>
<listRelation><relation active="#struct-300009" type="direct"></relation>
<relation active="#struct-413289" type="direct"></relation>
<relation name="UMR7503" active="#struct-441569" type="direct"></relation>
</listRelation>
</org>
</tutelle>
<tutelle active="#struct-413289" type="indirect"><org type="institution" xml:id="struct-413289" status="VALID"><idno type="IdRef">157040569</idno>
<idno type="IdUnivLorraine">[UL]100--</idno>
<orgName>Université de Lorraine</orgName>
<orgName type="acronym">UL</orgName>
<date type="start">2012-01-01</date>
<desc><address><addrLine>34 cours Léopold - CS 25233 - 54052 Nancy cedex</addrLine>
<country key="FR"></country>
</address>
<ref type="url">http://www.univ-lorraine.fr/</ref>
</desc>
</org>
</tutelle>
<tutelle name="UMR7503" active="#struct-441569" type="indirect"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="ISNI">0000000122597504</idno>
<idno type="IdRef">02636817X</idno>
<orgName>Centre National de la Recherche Scientifique</orgName>
<orgName type="acronym">CNRS</orgName>
<date type="start">1939-10-19</date>
<desc><address><country key="FR"></country>
</address>
<ref type="url">http://www.cnrs.fr/</ref>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>France</country>
<placeName><settlement type="city">Nancy</settlement>
<settlement type="city">Metz</settlement>
<region type="region" nuts="2">Grand Est</region>
<region type="old region" nuts="2">Lorraine (région)</region>
</placeName>
<orgName type="university">Université de Lorraine</orgName>
</affiliation>
</author>
<author><name sortKey="Gkiokas, Aggelos" sort="Gkiokas, Aggelos" uniqKey="Gkiokas A" first="Aggelos" last="Gkiokas">Aggelos Gkiokas</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-110738" status="VALID"><orgName>Institute for Language and Speech Processing</orgName>
<orgName type="acronym">ILSP</orgName>
<desc><address><addrLine>Institute for Language and Speech Processing / R.C. "Athena" Artemidos 6 & Epidavrou GR-151 25 MAROYSSI Greece</addrLine>
<country key="GR"></country>
</address>
<ref type="url">http://www.ilsp.gr/</ref>
</desc>
<listRelation><relation active="#struct-317224" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-317224" type="direct"><org type="institution" xml:id="struct-317224" status="INCOMING"><orgName>Institute for Language and Speech Processing</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Grèce</country>
</affiliation>
</author>
<author><name sortKey="Schnitzer, Dominik" sort="Schnitzer, Dominik" uniqKey="Schnitzer D" first="Dominik" last="Schnitzer">Dominik Schnitzer</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-93952" status="VALID"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<orgName type="acronym">OFAI</orgName>
<desc><address><addrLine>Freyung 6/6, A-1010 Vienna Austria</addrLine>
<country key="AT"></country>
</address>
<ref type="url">http://www.ofai.at/</ref>
</desc>
<listRelation><relation active="#struct-323078" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-323078" type="direct"><org type="institution" xml:id="struct-323078" status="INCOMING"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Autriche</country>
</affiliation>
</author>
<author><name sortKey="Flexer, Arthur" sort="Flexer, Arthur" uniqKey="Flexer A" first="Arthur" last="Flexer">Arthur Flexer</name>
<affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-93952" status="VALID"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<orgName type="acronym">OFAI</orgName>
<desc><address><addrLine>Freyung 6/6, A-1010 Vienna Austria</addrLine>
<country key="AT"></country>
</address>
<ref type="url">http://www.ofai.at/</ref>
</desc>
<listRelation><relation active="#struct-323078" type="direct"></relation>
</listRelation>
<tutelles><tutelle active="#struct-323078" type="direct"><org type="institution" xml:id="struct-323078" status="INCOMING"><orgName>Austrian Research Institute for Artificial Intelligence</orgName>
<desc><address><country key="FR"></country>
</address>
</desc>
</org>
</tutelle>
</tutelles>
</hal:affiliation>
<country>Autriche</country>
</affiliation>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Noise-robust automatic speech recognition (ASR) systems rely on feature and/or model compensation. Existing compensation techniques typically operate on the features or on the parameters of the acoustic models themselves. By contrast, a number of normalization techniques have been defined in the field of speaker verification that operate on the resulting log-likelihood scores. In this paper, we provide a theoretical motivation for likelihood normalization due to the so-called "hubness" phenomenon and we evaluate the benefit of several normalization techniques on ASR accuracy for the 2nd CHiME Challenge task. We show that symmetric normalization (S-norm) reduces the relative error rate by 43% alone and by 10% after feature and model compensation.</div>
</front>
</TEI>
<affiliations><list><country><li>Autriche</li>
<li>France</li>
<li>Grèce</li>
</country>
<region><li>Grand Est</li>
<li>Lorraine (région)</li>
</region>
<settlement><li>Metz</li>
<li>Nancy</li>
</settlement>
<orgName><li>Université de Lorraine</li>
</orgName>
</list>
<tree><country name="France"><region name="Grand Est"><name sortKey="Vincent, Emmanuel" sort="Vincent, Emmanuel" uniqKey="Vincent E" first="Emmanuel" last="Vincent">Emmanuel Vincent</name>
</region>
</country>
<country name="Grèce"><noRegion><name sortKey="Gkiokas, Aggelos" sort="Gkiokas, Aggelos" uniqKey="Gkiokas A" first="Aggelos" last="Gkiokas">Aggelos Gkiokas</name>
</noRegion>
</country>
<country name="Autriche"><noRegion><name sortKey="Schnitzer, Dominik" sort="Schnitzer, Dominik" uniqKey="Schnitzer D" first="Dominik" last="Schnitzer">Dominik Schnitzer</name>
</noRegion>
<name sortKey="Flexer, Arthur" sort="Flexer, Arthur" uniqKey="Flexer A" first="Arthur" last="Flexer">Arthur Flexer</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 000949 | SxmlIndent | more

HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000949 | 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=     Hal:hal-01006142
   |texte=   An investigation of likelihood normalization for robust ASR
}}

This area was generated with Dilib version V0.6.33.
Data generation: Mon Jun 10 21:56:28 2019. Site generation: Fri Feb 25 15:29:27 2022

	Serveur d'exploration sur la recherche en informatique en Lorraine
	Attention, ce site est en cours de développement ! Attention, site généré par des moyens informatiques à partir de corpus bruts. Les informations ne sont donc pas validées.

Serveur d'exploration sur la recherche en informatique en Lorraine

An investigation of likelihood normalization for robust ASR

An investigation of likelihood normalization for robust ASR

Source :

Abstract

Links toward previous steps (curation, corpus...)

Le document en format XML

Pour manipuler ce document sous Unix (Dilib)

Pour mettre un lien sur cette page dans le réseau Wicri