Detecting And Localization Of Digital Audio Forgeries Based On Singularity Analysis With Wavelet Packet

Digital audio watermarking and signature are widely used for authentication.However, these techniques will become powerless in many actual situations becauseof their requirement of additional information. Digital audio forensic techniques arenecessary for digital audio. In this thesis, we propose an audio forensics scheme todetect and locate speech audio forged operations in time domain (including deletion,insertion, substitution and splicing) by performing discrete wavelet packetdecomposition and analyzing singularity points of audio signals.In the thesis, we first analyze the forged operations and find that the audiosignals will often generate new singular points because of the decrease or breaking ofthe correlation property of those samples close to the tampering position. Then weutilize the singularity analysis based on wavelet packet and design five parameters(which is different for the sample rate of digital audio file) to propose an approachwhich can detect and locate audio forgeries in time domain.Finally, extensive experimental results have demonstrated that the proposedmethod can better achieve the goals that identify whether a given speech file has beentampered (e.g., part of the content deleted or replaced) previously and further locatethe forged positions.