基于波原子变换的语音信号去噪毕业论文

《基于波原子变换的语音信号去噪毕业论文》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、Contents1.INTRODUCTION12.WAVEATOMTRANSFORM42.1DefinitionofWaveAtomTransform42.21-DCharacteristicsinWaveAtomTransform62.3TheStructureof1-Dfilter-bankinWaveAtom……………………………….…103.THRESHOLDINGMETHODFORNOISEREDUCTION173.1HardandSoftThresholdings173.2Determi

2、nationofThreshold204.EXPERIMENTSANDDISCUSSIONS……………………………………………..274.1ExperimentalConditions…………………………………………………………..274.2ResultsandDiscussions285.CONCLUSION37REFERENCE38ABSTRACT(INKOREAN)40i401.INTRODUCTIONSpeechisoneofthemostnaturalandconvenientwaysof

3、intercommunication.Thus,speechsignalprocessingtechniquesforaman-machineinterfacelikespeechrecognition,speechsynthesisandspeechcodinghavedrawnmuchattention.Asspeechprocessingmovedfromthelaboratorytothefield,ithasbecomeincreasinglyimportanttodealwiththea

4、mbientnoise.Sincethequalityandintelligibilityaredegradedbytheambientnoise,theproblemofreducingnoisecomponentsofthenoisyspeechisstillregardedasanimportantissueinthefieldofspeechresearch[1-3].Mostclassicalmethodsforhandlingambientwidebandnoisewerevariant

5、sofanapproachbasedonsubtractinganestimateofthespectrumofthenoisefromthatofthenoisyspeech[4-7].AndthenanovelapproachfornoisereductionusingthewavelettransformhasbeenproposedbyD.L.Donoho[8].Itemploysthethresholdingtechniqueinthewaveletdomainandhasshowntoh

6、avegoodpropertiesforawideclassofsignalscorruptedbyadditivewhiteGaussiannoise.Overthepasttwodecades,numeroussignalprocessingtechniqueswiththewavelettransformhavebeendevelopedtomakeuseofitssuperioritytotheconventionalFouriertransform.Recently,waveatomtra

7、nsformhasbeenproposed,andhasshownits40potentialindenoisingexperimentsonbothseismicandfingerprintimages[1].Waveatomwasintroducedasavariantof2Dwaveletpacketsobeyingtheparabolicscaling.Inanutshell,waveatomsinterpolateexactlybetweendirectionalwaveletsandGa

8、bor,inthesensethattheperiodofoscillationsofeachwavepacket(wavelength)islinkedtothesizeoftheessentialsupport(diameter)bytheparabolicscaling,i.e.,wavelength~(diameter)2[1,2].Inthisthesis,weapplythe1Dwaveatomtransformtonoisyspeechsignalsfo