基于自適應(yīng)子帶功率譜熵靜音檢測的g.729改進(jìn)算法

1、哈爾濱工程大學(xué)碩士學(xué)位論文基于自適應(yīng)子帶功率譜熵靜音檢測的G.729改進(jìn)算法姓名：章東升申請學(xué)位級別：碩士專業(yè)：信號與信息處理指導(dǎo)教師：陳立偉20081201哈爾濱工程大學(xué)碩士學(xué)位論文ABSTRACTItisnecessarytomakelowratespeechcoderresolvethelimitofbroadbandresource、析nltheveryfastdevelopmentofdigitalmobileandmulti

2、mediacommunicationAstheharvests，therewerealotofmidelowratecodebasedonbothparametricandmixedcodepresentedbyscholarsThosearethebasicofthe丌UnewrecommendationforspeechcodeItisveryimportantforthoseharvestsapplyingwidelyandqui

3、cklyITUTRecommendationG729wasapprovedin1996containingthedescriptionofanalgorithmforthecodingofspeechsignalsat8kb／susingConjugateStructureAlgebraicCode—ExcitedLinearPrediction(csACELP)，anditiswidelyusedinmultimediacommuni

4、cations，cellmobilecommunicationsandIPnetphoneTheCSACELPcoderisbasedonheCode—ExcitedLinearPrediction(CELP)codingmodelconsistingofseveralkeypointssuchasLinearPredictionCodinganalysis，VectorQuantizationConjugate—StructureAl

5、gebraic—CodebookPerceptualWeighing，andSOonG729istheresearchobjectofthisthesisbecauseofitsefficiencyofcompressionandgoodqualityofsynthesisspeechBut(2729isthemostcomplexityalgorithmthatITU—Teverproposed，itsbitrateisfixedan

6、dithaven’tcompressedthesilencetimebetweeneffectivespeechesThisthesisadoptsadaptivemultibandspectralentropyVoiceActivityDetectionwhichwon’taddextracomputingcomplexitytoimproveG729algorithmandenhanceitscompressionratioAdap