中英文献翻译语音识别speech recognition.docx

中英文献翻译语音识别speech recognition.docx

《中英文献翻译语音识别speech recognition.docx》由会员分享，可在线阅读，更多相关《中英文献翻译语音识别speech recognition.docx（15页珍藏版）》请在冰豆网上搜索。

中英文献翻译语音识别speechrecognition

中英文献翻译：

语音识别speechrecognition

SpeechRecognition

VictorZue,RonCole,&WayneWard

MITLaboratoryforComputerScience,Cambridge,Massachusetts,USA

OregonGraduateInstituteofScience&Technology,Portland,Oregon,USA

CarnegieMellonUniversity,Pittsburgh,Pennsylvania,USA

1DefiningtheProblem

Speechrecognitionistheprocessofconvertinganacousticsignal,capturedbyamicrophoneoratelephone,toasetofwords.Therecognizedwordscanbethefinalresults,asforapplicationssuchascommands&control,dataentry,anddocumentpreparation.Theycanalsoserveastheinputtofurtherlinguisticprocessinginordertoachievespeechunderstanding,asubjectcoveredinsection.

Speechrecognitionsystemscanbecharacterizedbymanyparameters,someofthemoreimportantofwhichareshowninFigure.Anisolated-wordspeechrecognitionsystemrequiresthatthespeakerpausebrieflybetweenwords,whereasacontinuousspeechrecognitionsystemdoesnot.Spontaneous,orextemporaneouslygenerated,speechcontainsdisfluencies,andismuchmoredifficulttorecognizethanspeechreadfromscript.Somesystemsrequirespeakerenrollment---ausermustprovidesamplesofhisorherspeechbeforeusingthem,whereasothersystemsaresaidtobespeaker-independent,inthatnoenrollmentisnecessary.Someoftheotherparametersdependonthespecifictask.Recognitionisgenerallymoredifficultwhenvocabulariesarelargeorhavemanysimilar-soundingwords.Whenspeechisproducedinasequenceofwords,languagemodelsorartificialgrammarsareusedtorestrictthecombinationofwords.

Thesimplestlanguagemodelcanbespecifiedasafinite-statenetwork,wherethepermissiblewordsfollowingeachwordaregivenexplicitly.Moregenerallanguagemodelsapproximatingnaturallanguagearespecifiedintermsofacontext-sensitivegrammar.

1

Onepopularmeasureofthedifficultyofthetask,combiningthevocabularysizeandthelanguagemodel,isperplexity,looselydefinedasthegeometricmeanofthenumberofwordsthatcanfollowawordafterthelanguagemodelhasbeenapplied（seesectionforadiscussionoflanguagemodelingingeneralandperplexityinparticular）.Finally,therearesomeexternalparametersthatcanaffectspeechrecognitionsystemperformance,includingthecharacteristicsoftheenvironmentalnoiseandthetypeandtheplacementofthemicrophone.

ParametersRange

SpeakingModeIsolatedwordstocontinuousspeech

SpeakingStyleReadspeechtospontaneousspeech

EnrollmentSpeaker-dependenttoSpeaker-independent

VocabularySmall（<20words）tolarge（>20,000words）

LanguageModelFinite-statetocontext-sensitive

PerplexitySmall（<10）tolarge（>100）

SNRHigh（>30dB）tolaw（<10dB）

TransducerVoice-cancellingmicrophonetotelephone

Table:

Typicalparametersusedtocharacterizethecapabilityofspeechrecognitionsystems

Speechrecognitionisadifficultproblem,largelybecauseofthemanysourcesofvariabilityassociatedwiththesignal.First,theacousticrealizationsofphonemes,thesmallestsoundunitsofwhichwordsarecomposed,arehighlydependentonthecontextinwhichtheyappear.Thesephoneticvariabilitiesareexemplifiedbytheacousticdifferencesofthephoneme，Atword

boundaries,contextualvariationscanbequitedramatic---makinggasshortagesoundlikegashshortageinAmericanEnglish,anddevoandaresoundlikedevandareinItalian.

Second,acousticvariabilitiescanresultfromchangesintheenvironmentaswellasinthepositionandcharacteristicsofthetransducer.Third,within-speakervariabilitiescanresultfromchangesinthespeaker'sphysicalandemotionalstate,speakingrate,orvoicequality.Finally,differencesinsociolinguisticbackground,dialect,andvocaltractsizeandshapecancontributetoacross-speakervariabilities.

Figureshowsthemajorcomponentsofatypicalspeechrecognitionsystem.Thedigitizedspeechsignalisfirsttransformedintoasetofusefulmeasurementsorfeaturesatafixedrate,

typicallyonceevery10--20msec（seesectionsand11.3forsignalrepresentationanddigitalsignalprocessing,respectively）.Thesemeasurementsarethenusedtosearchforthemostlikelywordcandidate,makinguseofconstraintsimposedbytheacoustic,lexical,andlanguagemodels.Throughoutthisprocess,trainingdataareusedtodeterminethevaluesofthemodelparameters.

Figure:

Componentsofatypicalspeechrecognitionsystem.

Speechrecognitionsystemsattempttomodelthesourcesofvariabilitydescribedaboveinseveralways.Atthelevelofsignalrepresentation,researchershavedevelopedrepresentationsthatemphasizeperceptuallyimportantspeaker-independentfeaturesofthesignal,andde-emphasizespeaker-dependentcharacteristics.Attheacousticphoneticlevel,speakervariabilityistypicallymodeledusingstatisticaltechniquesappliedtolargeamountsofdata.Speakeradaptationalgorithmshavealsobeendevelopedthatadaptspeaker-independentacousticmodelstothoseofthecurrentspeakerduringsystemuse,（seesection）.Effectsoflinguisticcontextattheacousticphoneticlevelaretypicallyhandledbytrainingseparatemodelsforphonemesindifferentcontexts;thisiscalledcontextdependentacousticmodeling.

Wordlevelvariabilitycanbehandledbyallowingalternatepronunciationsofwordsinrepresentationsknownaspronunciationnetworks.Commonalternatepronunciationsofwords,aswellaseffectsofdialectandaccentarehandledbyallowingsearchalgorithmstofindalternatepathsofphonemesthroughthesenetworks.Statisticallanguagemodels,basedonestimatesofthefrequencyofoccurrenceofwordsequences,areoftenusedtoguidethesearchthroughthemostprobablesequenceofwords.

ThedominantrecognitionparadigminthepastfifteenyearsisknownashiddenMarkov

models（HMM）.AnHMMisadoublystochasticmodel,inwhichthegenerationoftheunderlyingphonemestringandtheframe-by-frame,surfaceacousticrealizationsarebothrepresentedprobabilisticallyasMarkovprocesses,asdiscussedinsections,and11.2.Neuralnetworkshavealsobeenusedtoestimatetheframebasedscores;thesescoresarethenintegratedintoHMM-basedsystemarchitectures,inwhathascometobeknownashybridsystems,asdescribedinsection11.5.

Aninterestingfeatureofframe-basedHMMsystemsisthatspeechsegmentsareidentifiedduringthesearchprocess,ratherthanexplicitly.Analternateapproachistofirstidentifyspeechsegments,thenclassifythesegmentsandusethesegmentscorestorecognizewords.Thisapproachhasproducedcompetitiverecognitionperformanceinseveraltasks.

2StateoftheArt

Commentsaboutthestate-of-the-artneedtobemadeinthecontextofspecificapplicationswhichreflecttheconstraintsonthetask.Moreover,differenttechnologiesaresometimesappropriatefordifferenttasks.Forexample,whenthevocabularyissmall,theentirewordcanbemodeledasasingleunit.Suchanapproachisnotpracticalforlargevocabularies,wherewordmodelsmustbebuiltupfromsubwordunits.

PerformanceofspeechrecognitionsystemsistypicallydescribedintermsofworderrorrateE,definedas:

whereNisthetotalnumberofwordsinthetestset,andS,I,andDarethetotalnumberof

substitutions,insertions,anddeletions,respectively.

Thepastdecadehaswitnessedsignificantprogressinspeechrecognitiontechnology.Worderrorratescontinuetodropbyafactorof2everytwoyears.Substantialprogresshasbeenmadeinthebasictechnology,leadingtotheloweringofbarrierstospeakerindependence,continuousspeech,andlargevocabularies.Thereareseveralfactorsthathavecontributedtothisrapidprogress.First,thereisthecomingofageoftheHMM.HMMispowerfulinthat,withtheavailabilityoftrainingdata,theparametersofthemodelcanbetrainedautomaticallytogiveoptimalperformance.

Second,muchefforthasgoneintothedevelopmentoflargespeechcorporaforsystem

development,training,andtesting.Someofthesecorporaaredesignedforacousticphoneticresearch,whileothersarehighlytaskspecific.Nowadays,itisnotuncommontohavetensofthousandsofsentencesavailableforsystemtrainingandtesting.Thesecorporapermitresearcherstoquantifytheacousticcuesimportantforphoneticcontrastsandtodetermineparametersoftherecognizersinastatisticallymeaningfulway.Whilemanyofthesecorpora（e.g.,TIMIT,RM,ATIS,andWSJ;seesection12.3）wereoriginallycollectedunderthesponsorshipoftheU.S.DefenseAdvancedResearchProjectsAgency（ARPA）tospurhumanlanguagetechnologydevelopmentamongitscontractors,theyhaveneverthelessgainedworld-wideacceptance（e.g.,inCanada,France,Germany,Japan,andtheU.K.）asstandardsonwhichtoevaluatespeechrecognition.

Third,progresshasbeenbroughtaboutbytheestablishmentofstandardsforperformanceevaluation.Onlyadecadeago,researcherstrainedandtestedtheirsystemsusinglocallycollecteddata,andhadnotbeenverycarefulindelineatingtrainingandtestingsets.Asaresult,itwasverydifficulttocompareperformanceacrosssystems,andasystem'sperformancetypicallydegradedwhenitwaspresentedwithpreviouslyunseendata.Therecentavailabilityofalargebodyofdatainthepublicdomain,coupledwiththespecificationofevaluationstandards,hasresultedinuniformdocumentationoftestresults,thuscontributingtogreaterreliabilityinmonitoringprogress（corpusdevelopmentactivitiesandevaluationmethodologiesaresummarizedinchapters12and13respectively）.

Finally,advancesincomputertechnologyhavealsoindirectlyinfluencedourprogress.Theavailabilityoffastcomp