A Novel Approach to Generate Watermarks Using Auditory Features for Audio Watermarking

The increasing number of problems related to the ownership protection of the multimedia content & its authentication due to digital revolution had lead to evolution of digital watermarking techniques. Digital watermarking provides a viable solution in minimizing this problem and boosting the confidence among multimedia content i.e. text, image, video, audio providers to make available there content on a public domain. But the watermarks used in most of the watermarking methods are not personalized or unique because of which the problem of ambiguity arises if same watermark is used by two individuals. In this paper we propose a novel method to generate a unique personalized watermark which can act as authenticating information for any watermarking algorithm specially targeting an audio. The watermarks are generated through feature vectors extracted from the speech of an individual which are unique for a particular person. The uniqueness of the watermark is shown through exhaustive experiments.