Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment

Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment

ID:40848683

大?。?97.34 KB

頁數(shù):7頁

時間:2019-08-08

Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment_第1頁
Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment_第2頁
Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment_第3頁
Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment_第4頁
Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment_第5頁
資源描述:

《Two Methods of Automatic Evaluation of Speech Signal Enhancement Recorded in the Open-Air MRI Environment》由會員上傳分享,免費在線閱讀,更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫。

1、MEASUREMENTSCIENCEREVIEW,17,(2017),No.6,257-263Journalhomepage:http://www.degruyter.com/view/j/msrTwoMethodsofAutomaticEvaluationofSpeechSignalEnhancementRecordedintheOpen-AirMRIEnvironmentJi?íP?ibil1,AnnaP?ibilová2,IvanFrollo11InstituteofMeasurementScience,SAS,Dúbravsk

2、ácesta9,SK-84104Bratislava,Slovakia,Jiri.Pribil@savba.sk2InstituteofElectronicsandPhotonics,FEE&IT,SUT,Ilkovi?ova3,SK-81219Bratislava,SlovakiaThepaperfocusesontwomethodsofevaluationofsuccessfulnessofspeechsignalenhancementrecordedintheopen-airmagneticresonanceimagerduri

3、ngphonationforthe3Dhumanvocaltractmodeling.ThefirstapproachenablestoobtainacomparisonbasedonstatisticalanalysisbyANOVAandhypothesistests.ThesecondmethodisbasedonclassificationbyGaussianmixturemodels(GMM).TheperformedexperimentshaveconfirmedthattheproposedANOVAandGMMclas

4、sifiersforautomaticevaluationofthespeechqualityarefunctionalandproducefullycomparableresultswiththestandardevaluationbasedonthelisteningtestmethod.Keywords:Acousticnoisesuppression,magneticresonanceimaging,speechprocessing,statisticalanalysis.1.INTRODUCTIONannotationoft

5、hespeechcorpus,etc.[13]).Theobjectiveapproachesformeasuringthespeechsignalquality[14]Constructionof3Darticulatorymodelsisnecessaryforcomprise,forexample,evaluationofdifferencesbetweenthebetterrepresentationofthehumanvocaltractfunctionandspeechspectralenvelopes[11]orspec

6、traldistances[12],etc.thesubsequentarticulatoryspeechsynthesis.Forthisreason,Thesefeaturesmaybecomparedandmatchedusingthetheaudiosignalmustberecordedsimultaneouslywiththestatisticalapproaches,liketheanalysisofvariancesimagescanning[1].Themagneticresonanceimaging(MRI)(AN

7、OVA)[15],[16]orhypothesistests[17],[18].ThefinaldeviceisusedtoobtainthevocaltractimagesoftheevaluationintheseapproachesbearstheformofaarticulatingpersonthatliesinthescanningareawhiletherecognitionscorethatcanbeobtainedbythemethodsbasedMRsequenceisrunning[2],[3].TheMRIeq

8、uipmentonartificialneuralnetworks,thenearestneighbor[19],consistsofagradientcoilsystemproducingthreeorthogonal

當(dāng)前文檔最多預(yù)覽五頁,下載文檔查看全文

此文檔下載收益歸作者所有

當(dāng)前文檔最多預(yù)覽五頁,下載文檔查看全文
溫馨提示:
1. 部分包含數(shù)學(xué)公式或PPT動畫的文件,查看預(yù)覽時可能會顯示錯亂或異常,文件下載后無此問題,請放心下載。
2. 本文檔由用戶上傳,版權(quán)歸屬用戶,天天文庫負(fù)責(zé)整理代發(fā)布。如果您對本文檔版權(quán)有爭議請及時聯(lián)系客服。
3. 下載前請仔細閱讀文檔內(nèi)容,確認(rèn)文檔內(nèi)容符合您的需求后進行下載,若出現(xiàn)內(nèi)容與標(biāo)題不符可向本站投訴處理。
4. 下載文檔時可能由于網(wǎng)絡(luò)波動等原因無法下載或下載錯誤,付費完成后未能成功下載的用戶請聯(lián)系客服處理。