A Privacy-Preserving Multipurpose Watermarking Scheme for Audio Authentication and Protection

In the scenario of cloud computing, it is necessary to consider the privacy protection of the audio content as well as copyright protection and content authentication simultaneously. In this paper, we propose an encrypted audio multipurpose watermarking scheme based on cloud storage and computing. We firstly perform encrypted domain discrete wavelet transform on the encrypted audio to obtain encrypted coefficients. By using the spread spectrum method, we embed the robust and the fragile watermarks into different sub-bands. We then propose algorithms to extract the robust and the fragile watermarks from the encrypted watermarked audio. Some experiments are conducted to verify the validity and feasibility of the proposed scheme. The results show that the proposed scheme has good performances of robustness, fragility, and tamper location.