<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-03874256</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-21T01:42:03+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">A Benchmarking on Cloud based Speech-To-Text Services for French Speech and Background Noise Effect</title>
            <author role="aut">
              <persName>
                <forename type="first">Binbin</forename>
                <surname>Xu</surname>
              </persName>
              <email type="md5">104a834bf0153300a21810fc45e1217c</email>
              <email type="domain">inria.fr</email>
              <idno type="idhal" notation="string">binbin-xu</idno>
              <idno type="idhal" notation="numeric">747446</idno>
              <idno type="halauthorid" notation="string">17852-747446</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-0822-5250</idno>
              <idno type="ARXIV">https://arxiv.org/a/xu_b_3</idno>
              <affiliation ref="#struct-1100796"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Tao</forename>
                <surname>Chongyang</surname>
              </persName>
              <idno type="halauthorid">2659656-0</idno>
              <affiliation ref="#struct-1100796"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Youssef</forename>
                <surname>Raqui</surname>
              </persName>
              <email type="md5">82474ec896ba90551ed429bb06d99dff</email>
              <email type="domain">diappymed.com</email>
              <idno type="idhal" notation="numeric">1195143</idno>
              <idno type="halauthorid" notation="string">2242474-1195143</idno>
              <affiliation ref="#struct-1072785"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Sylvie</forename>
                <surname>Ranwez</surname>
              </persName>
              <email type="md5">ec1662047a6c7320459b3b527501c713</email>
              <email type="domain">mines-ales.fr</email>
              <idno type="idhal" notation="string">sylvie-ranwez</idno>
              <idno type="idhal" notation="numeric">20659</idno>
              <idno type="halauthorid" notation="string">29379-20659</idno>
              <idno type="RESEARCHERID">http://www.researcherid.com/rid/F-5975-2013</idno>
              <idno type="ORCID">https://orcid.org/0000-0002-8707-0313</idno>
              <idno type="IDREF">https://www.idref.fr/165809418</idno>
              <idno type="RESEARCHERID">http://www.researcherid.com/rid/http://www.researcherid.com/rid/F-5975-2013</idno>
              <affiliation ref="#struct-1100796"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Céline</forename>
                <surname>Rouveirol</surname>
              </persName>
              <email type="md5">b93c50ff9185ef754937e6da2f10f4d3</email>
              <email type="domain">lipn.univ-paris13.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2022-11-27 21:42:07</date>
              <date type="whenModified">2026-02-12 08:38:02</date>
              <date type="whenReleased">2022-12-13 08:18:49</date>
              <date type="whenProduced">2021-07-01</date>
              <date type="whenEndEmbargoed">2022-11-27</date>
              <ref type="file" target="https://hal.science/hal-03874256v1/document">
                <date notBefore="2022-11-27"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-03874256v1/file/APIA2021_paper_12.pdf" id="file-3874256-3407817">
                <date notBefore="2022-11-27"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="735329">
                <persName>
                  <forename>Céline</forename>
                  <surname>Rouveirol</surname>
                </persName>
                <email type="md5">b93c50ff9185ef754937e6da2f10f4d3</email>
                <email type="domain">lipn.univ-paris13.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-03874256</idno>
            <idno type="halUri">https://hal.science/hal-03874256</idno>
            <idno type="halBibtex">xu:hal-03874256</idno>
            <idno type="halRefHtml">&lt;i&gt;Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle (APIA 2021)&lt;/i&gt;, Jul 2021, Bordeaux, France. p. 102-107</idno>
            <idno type="halRef">Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle (APIA 2021), Jul 2021, Bordeaux, France. p. 102-107</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-3874256-3407817"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="UNIV-PARIS13">Université Paris-Nord - Paris XIII </idno>
            <idno type="stamp" n="EM-ALES">Ecole des Mines d'Alès</idno>
            <idno type="stamp" n="UNIV-MONTPELLIER">Université de Montpellier</idno>
            <idno type="stamp" n="INSTITUTS-TELECOM">composantes instituts telecom </idno>
            <idno type="stamp" n="SORBONNE-PARIS-NORD">Université Sorbonne Paris Nord</idno>
            <idno type="stamp" n="EUROMOV-DHM" corresp="UNIV-MONTPELLIER">EuroMov Digital Health in Motion</idno>
            <idno type="stamp" n="UM-2015-2021" corresp="UNIV-MONTPELLIER">Université de Montpellier (2015-2021)</idno>
            <idno type="stamp" n="UM-EPE" corresp="UNIV-MONTPELLIER">Université de Montpellier - EPE</idno>
            <idno type="stamp" n="APIA-2021" corresp="UNIV-PARIS13">APIA 2021 </idno>
            <idno type="stamp" n="IA">Intelligence Artificielle</idno>
            <idno type="stamp" n="IMT-MINES-ALES" corresp="INSTITUT-MINES-TELECOM">IMT Mines Alès</idno>
            <idno type="stamp" n="INSTITUT-MINES-TELECOM">Institut Mines Telecom</idno>
            <idno type="stamp" n="STAPS">STAPS</idno>
            <idno type="stamp" n="COLL-STAPS">COLL-STAPS</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="3">National</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">A Benchmarking on Cloud based Speech-To-Text Services for French Speech and Background Noise Effect</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Binbin</forename>
                    <surname>Xu</surname>
                  </persName>
                  <email type="md5">104a834bf0153300a21810fc45e1217c</email>
                  <email type="domain">inria.fr</email>
                  <idno type="idhal" notation="string">binbin-xu</idno>
                  <idno type="idhal" notation="numeric">747446</idno>
                  <idno type="halauthorid" notation="string">17852-747446</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-0822-5250</idno>
                  <idno type="ARXIV">https://arxiv.org/a/xu_b_3</idno>
                  <affiliation ref="#struct-1100796"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Tao</forename>
                    <surname>Chongyang</surname>
                  </persName>
                  <idno type="halauthorid">2659656-0</idno>
                  <affiliation ref="#struct-1100796"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Youssef</forename>
                    <surname>Raqui</surname>
                  </persName>
                  <email type="md5">82474ec896ba90551ed429bb06d99dff</email>
                  <email type="domain">diappymed.com</email>
                  <idno type="idhal" notation="numeric">1195143</idno>
                  <idno type="halauthorid" notation="string">2242474-1195143</idno>
                  <affiliation ref="#struct-1072785"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Sylvie</forename>
                    <surname>Ranwez</surname>
                  </persName>
                  <email type="md5">ec1662047a6c7320459b3b527501c713</email>
                  <email type="domain">mines-ales.fr</email>
                  <idno type="idhal" notation="string">sylvie-ranwez</idno>
                  <idno type="idhal" notation="numeric">20659</idno>
                  <idno type="halauthorid" notation="string">29379-20659</idno>
                  <idno type="RESEARCHERID">http://www.researcherid.com/rid/F-5975-2013</idno>
                  <idno type="ORCID">https://orcid.org/0000-0002-8707-0313</idno>
                  <idno type="IDREF">https://www.idref.fr/165809418</idno>
                  <idno type="RESEARCHERID">http://www.researcherid.com/rid/http://www.researcherid.com/rid/F-5975-2013</idno>
                  <affiliation ref="#struct-1100796"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>Conférence Nationale sur les Applications Pratiques de l’Intelligence Artificielle (APIA 2021)</title>
                  <date type="start">2021-07-01</date>
                  <date type="end">2021-07-02</date>
                  <settlement>Bordeaux</settlement>
                  <country key="FR">France</country>
                </meeting>
                <imprint>
                  <biblScope unit="pp">p. 102-107</biblScope>
                  <date type="datePub">2021</date>
                </imprint>
              </monogr>
              <ref target="https://arxiv.org/abs/2105.03409" type="seeAlso"/>
              <ref type="publisher">https://pfia2021.fr/conferences/apia/actes_APIA_CH_PFIA2021.pdf</ref>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Google Cloud</term>
                <term xml:lang="en">IBM Watson</term>
                <term xml:lang="en">French language</term>
                <term xml:lang="en">Benchmarking</term>
                <term xml:lang="en">Speech-To-Text</term>
                <term xml:lang="fr">Reconnaissance vocale</term>
                <term xml:lang="fr">Evaluation comparative</term>
                <term xml:lang="fr">Langue Française</term>
                <term xml:lang="fr">Google Cloud Speech-To-Text</term>
                <term xml:lang="fr">Microsoft Azure Cognitive Services</term>
                <term xml:lang="fr">Amazon Transcribe</term>
                <term xml:lang="fr">IBM Watson Speech to Text</term>
              </keywords>
              <classCode scheme="halDomain" n="info.info-ai">Computer Science [cs]/Artificial Intelligence [cs.AI]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>This study presents a large scale benchmarking on cloudbased Speech-To-Text systems : Google Cloud Speech-To-Text, Microsoft Azure Cognitive Services, Amazon Transcribe, IBM Watson Speech to Text. For each systems, 40 158 clean and noisy speech files about 101 hours are tested. Effect of background noise on STT quality is also evaluated with 5 different Signal-to-noise ratios from 40 dB to 0 dB. Results showed that Microsoft Azure provided lowest transcription error rate 9.09% on clean speech, with high robustness to noisy environment. Google Cloud and Amazon Transcribe gave similar performance, but the latter is very limited for time-constraint usage. Though IBM Watson could work correctly in quiet conditions, it is highly sensible to noisy speech which could strongly limit its application in real life situations.</p>
            </abstract>
            <abstract xml:lang="fr">
              <p>Alors que les applications de reconnaissance vocale se sont imposées dans notre quotidien, il existe peu d'études à grande échelle pour comparer les performances des solutions de l'état de l'art. Ceci est d'autant plus vrai dans une langue autre que la langue anglaise. Cet article propose une telle analyse comparative basée sur 17 heures d'enregistrement en Français. Quatre systèmes sont analysés : Google Cloud Speech-To-Text, Microsoft Azure Cognitive Services, Amazon Transcribe, et IBM Watson Speech to Text. Chacun ayant été mis à l'épreuve de cinq niveaux de bruit de fond, c'est l'équivalent de 400 heures de discours qui sont analysées. Microsoft Azure Cognitive Services a montré les meilleurs résultats en terme de taux d'erreur et une bonne résistance au bruit, tandis que la sensibilité au bruit d'IBM Watson Speech to Text compromet son usage en situation réelle.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="laboratory" xml:id="struct-1100796" status="VALID">
          <idno type="RNSR">202123850M</idno>
          <idno type="ROR">https://ror.org/054b22910</idno>
          <orgName>EuroMov - Digital Health in Motion</orgName>
          <orgName type="acronym">Euromov DHM</orgName>
          <date type="start">2022-01-01</date>
          <desc>
            <address>
              <addrLine>Université de Montpellier UFR STAPS 700 avenue du Pic Saint Loup34090 Montpellier</addrLine>
              <country key="FR"/>
            </address>
          </desc>
          <listRelation>
            <relation active="#struct-6279" type="direct"/>
            <relation active="#struct-302102" type="indirect"/>
            <relation name="UR_UM_IMT_102" active="#struct-1100589" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-1072785" status="INCOMING">
          <orgName>DiappyMed</orgName>
        </org>
        <org type="institution" xml:id="struct-6279" status="VALID">
          <idno type="IdRef">032486111</idno>
          <idno type="ISNI">0000 0000 9734 247X</idno>
          <idno type="ROR">https://ror.org/03e8rf594</idno>
          <orgName>IMT MINES ALÈS</orgName>
          <date type="start">1843-09-22</date>
          <desc>
            <address>
              <addrLine>IMT - Mines Ales École des MinesSite de Clavières : 6 avenue de Clavières, 30319 Alès CedexSite de Croupillac : Rue Jules Renard, 30100 AlèsSite de Pau : Hélioparc (bâtiment Einstein 1), 2 avenue du Président-Pierre-Angot, 64053 Pau Cedex 9</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.mines-ales.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-302102" type="direct"/>
          </listRelation>
        </org>
        <org type="regroupinstitution" xml:id="struct-302102" status="VALID">
          <idno type="IdRef">192427156</idno>
          <idno type="ISNI">000000012202567X</idno>
          <idno type="ROR">https://ror.org/025vp2923</idno>
          <idno type="Wikidata">Q27962533</idno>
          <orgName>Institut Mines-Télécom [Paris]</orgName>
          <orgName type="acronym">IMT</orgName>
          <date type="start">2012-03-01</date>
          <desc>
            <address>
              <addrLine>19 Place Marguerite Perey, 91120 Palaiseau</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.imt.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-1100589" status="VALID">
          <idno type="ROR">https://ror.org/051escj72</idno>
          <orgName>Université de Montpellier</orgName>
          <orgName type="acronym">UM</orgName>
          <date type="start">2022-01-01</date>
          <desc>
            <address>
              <addrLine>163 rue Auguste Broussonnet - 34090 Montpellier</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.umontpellier.fr/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>