Some works are reported in the frequency domain watermarking using Single Value Decomposition (SVD). The two most commonly used methods are based on DCT-SVD and DWT-SVD. The commonly present disadvantages in traditional watermarking techniques such as inability to withstand attacks are absent in SVD based algorithms. They offer a robust method of watermarking with minimum or no distortion. DCT based watermarking techniques offer compression while DWT based compression offer scalability. Thus all the three desirable properties can be utilized to create a new robust watermarking technique. In this paper, we propose a method of non-blind transform domain watermarking based on DWT-DCT-SVD. The DCT coefficients of the DWT coefficients are used to embed the watermarking information. This method of watermarking is found to be robust and the visual watermark is recoverable without only reasonable amount of distortion even in the case of attacks. Thus the method can be used to embed copyright information in the form of a visual watermark or simple text.

Speech undergoes various acoustic interferences in natural environment, while many of the applications require an effective way to separate the dominant signal from the interference. In this paper, a Short-time Fourier Transform (STFT) based unsupervised method for single channel speech separation is proposed. It uses the pitch information of the dominant and interfering speakers and then generating a time frequency mask based on the pitch frequencies. Through rigorous objective and subjective evaluations, it is shown that the proposed system is capable of providing better Signal to Noise Ratio (SNR) and Perceptual Evaluation of Speech Quality (PESQ) compared to other related methods available in the literature.

No abstract

No abstract

No abstract