Abstract
In this paper, a novel imperceptible, fragile and blind watermark scheme is proposed for speech tampering detection and self-recovery. The embedded watermark data for content recovery is calculated from the original discrete cosine transform (DCT) coefficients of host speech. The watermark information is shared in a frames-group instead of stored in one frame. The scheme trades off between the data waste problem and the tampering coincidence problem. When a part of a watermarked speech signal is tampered with, one can accurately localize the tampered area, the watermark data in the area without any modification still can be extracted. Then, a compressive sensing technique is employed to retrieve the coefficients by exploiting the sparseness in the DCT domain. The smaller the tampered the area, the better quality of the recovered signal is. Experimental results show that the watermarked signal is imperceptible, and the recovered signal is intelligible for high tampering rates of up to 47.6%. A deep learning-based enhancement method is also proposed and implemented to increase the SNR of recovered speech signal.
Highlights
With the development of Internet and communication technology, the utilization of multimedia data is becoming common in our daily life, but multimedia data can be tampered with it may be distorted during spreading through the Internet, and may be manipulated by an adversary.It is important to develop algorithms to protect and recover data
One advantage of watermark algorithms is that the algorithm can localize the tampering area, and on that basis, a number of watermarking schemes aiming at recovering multimedia data have been developed [1,2]
Many watermark algorithms used for the self-recovery in the image literature have been proposed in [3,4]
Summary
With the development of Internet and communication technology, the utilization of multimedia data is becoming common in our daily life, but multimedia data can be tampered with it may be distorted during spreading through the Internet, and may be manipulated by an adversary. When the data or the coefficients are embedded in the host, one common method is to embed data from one region in another region This kind of algorithm has a problem, in that when both regions of the host and the data represents it are tampered, the recovery work is impossible. To avoid the impact of inverse Fourier transformation on the fragile watermark, we apply the method not to the spectrum, which is more similar to a digital image, but to the time domain of the signal. The scheme avoids both the watermark data waste problem and tampering coincidence problem. We conduct several experiments to compare the enhancement effects of different neural network architectures
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
