A simple but efficient voice activity detection algorithm through Hilbert transform and dynamic threshold for speech pathologies

摘要

AsimplebutefficientvoiceactivitydetectorbasedontheHilberttransformandadynamicthresholdispresentedtobeusedonthepre-processingofaudiosignals.Thealgorithmtodefinethedynamicthresholdisamodificationofaconvexcombinationfoundinliterature.Thisschemeallowsthedetectionofprosodicandsilencesegmentsonaspeechinpresenceofnon-idealconditionslikeaspectraloverlappednoise.Thepresentworkshowspreliminaryresultsoveradatabasebuiltwithsomepoliticalspeech.Thetestswereperformedaddingartificialnoisetonaturalnoisesovertheaudiosignals,andsomealgorithmsarecompared.Resultswillbeextrapolatedtothefieldofadaptivefilteringonmonophonicsignalsandtheanalysisofspeechpathologiesonfuturesworks...