SpeechRecognitionVictorZue,RonCole,&puterScience,Cambridge,Massachusetts,USAOregonGraduateInstituteofScience&Technology,Portland,Oregon,USACarnegieMellonUniversity,Pittsburgh,Pennsylvania,USA1DefiningtheProblemSpeechrecognitionistheprocessofconvertinganacousticsignal,capturedbyamicrophoneoratelephone,,mands&control,dataentry,,,-wordspeechrecognitionsystemrequiresthatthespeakerpausebrieflybetweenwords,,orextemporaneouslygenerated,speechcontainsdisfluencies,---ausermustprovidesamplesofhisorherspeechbeforeusingthem,whereasothersystemsaresaidtobespeaker-independent,-,-work,-,isperplexity,looselydefinedasthegeometricmeanofthenumberofwordsthatcanfollowawordafterthelanguagemodelhasbeenapplied(seesectionforadiscussionoflanguagemodelingingeneralandperplexityinparticular).Finally,therearesomeexternalparametersthatcanaffectspeechrecognitionsystemperformance,,,theacousticr
语音识别 外文翻译 外文文献 英文文献 来自淘豆网m.daumloan.com转载请标明出处.