哼唱搜索技術(shù)的研究與實(shí)現(xiàn).pdf

1、南開(kāi)大學(xué)碩士學(xué)位論文哼唱搜索技術(shù)的研究與實(shí)現(xiàn)姓名：玄成根申請(qǐng)學(xué)位級(jí)別：碩士專(zhuān)業(yè)：計(jì)算機(jī)科學(xué)與技術(shù)；計(jì)算機(jī)系統(tǒng)結(jié)構(gòu)指導(dǎo)教師：白剛201105AbstractAbstractTheQuerybyHummingisacontentbasedmusicsearchingtechnology，whichisacomprehensiveresearchingprobleminvolvingsignalprocessingandpatternreco

2、gnitionTheQuerybyHummingisdifferentfromthetraditionaltextbasedsearchingmethod，providingpeoplewithanaturalsearchingmethodwhichistotakethehummedspeechsignalasallinputTheQuerybyHummingtechnologyhasawideapplicationprospectsi

3、nsearchingenginesanddigitallibrariesIntheQuerybyHummingsystem，users啪retrievetlleirexpectedsongsbyhummingapieceofmelodyTheretrievalprocessingis弱follow：firstasong’SmelodyiShummedintoinputdeviceandfeaturesaleextractedfromth

4、emelody；thenstartretrievalinthemusicdatabaseusingthefeatures；finallythesystemreturnsasong’SlistaccordingtothesimilaritybetweenthefeaturesThispaperdescribesthesongsmakinguseofacharacterstringcomposedbyU，D，S)alphabetEachch

5、aracterrepresentsrelativedifferencesinpitchbetweensuccessivenotesTheseUDSstringsarecalledpitchcontourandarealsothefeaturesofquerybyhummingsystemThepapercombinestheautocorrelationanalysisandcepstrumanalysistoexecutepitcht

6、rackingThenasegmentationbysentenceprocessisrunforeverysonginthedatabaseandeachsentenceispresentedaspitchcontourLaterKmeansclusteringisappliedinthesentencesetandallsongsalecodedbytheclasslabelThelabelstringistakenasfeatur

7、eThehummedspeechsignalcanbeconvertedtoalabelstringbythesanlewayFinallyanapproximatestringmatchingalgorithmCanbeusedtomatchthelabelstringsofinputspeechsignalandthesongsinthedatabase，returningatop10songslistAsrepresentedby