A Study on an Efficient Tampering Detection and Localization Method for Speech Signals

Due to the development of digital technologies, new social issues relating to maliciousattacks and unauthorized tampering to speech, such as content replacement and voicemorphing, have arisen. Using advanced speech analysis and synthesis tools enables thespeech to be tampered without leaving any perceptual clues. As an important informationcarrier, the originality, integrity, and authenticity of speech signals should be strictlyconfirmed.Authentication and tampering detection of digital signals is one of the mainapplications of digital watermarking. In this thesis, an efficient speech watermarkingmethod is proposed to generate self-embedding speech signals in that hash representationof a speech signal, which is assumed as watermark, is embedded into the signal itselfwithout affecting the original quality. The proposed system is intended to satisfy blindness,inaudibility, and fragility against malicious modifications.The proposed watermarking method in this thesis is a kind of fragile watermarkingand thus the hash information (watermark) in the tampered regions is destroyed whentampering occurs. This feature helps the receiver to detect and localize the tamperingregions by comparing the original hash information and the extracted hash from thereceived speech. The perfect match of the hash information confirms the integrity andoriginality of the received speech; otherwise it indicates tampering.In this thesis, performance of the proposed system is tested on 40 read speech fileswhich are International news and Burmese news read by 20 female and 20 maleannouncers. The proposed method is implemented in MATLAB and fragility againstmalicious modification is evaluated by applying various kinds of tampering such ascompression, zeroing, adding noise, time scaling, and reverberation attacks on thewatermarked speech files. Experimental results show that the proposed method is relevantwith the main requirements of a good watermarking scheme: inaudibility and blindness inaddition to fragility. Therefore, the proposed system is really useful for applications ofcriminal investigation and digital forensics where the integrity and originality of the speechevidence is extremely important