Securerobust and hybrid watermarking for speech signal using discrete wavelettransform discrete cosine transform andsingular value decomposition

A digital watermark is defined as inaudible data,permanently embedded in a speech file for authenticating the secret data. the main goal of this paper is to embed a watermark in the speech signal without any degradation. here the hybrid watermarking is performed based on the three techniques such as discrete cosine transform (dct) with singular value decomposition (svd) and discrete wavelet transform (dwt) and it is optimized by performing the separation of speech and silent regions using a voice activity detection algorithm. the performances were evaluated based on peak signal to noise ratio (psnr) and normalized cross correlation (ncc). the result shows that the optimization method performs better than the existing algorithm and it is robust against different kinds of attacks. it also shows that the algorithm is efficient in terms of robustness,security,and imperceptibility and also the watermarked signal is perceptually similar to the original audio signal. © school of engineering,taylor’s university.