A logarithmic method for speech watermarking based on the optimum sub-band in the bark scale

Abstract

In this paper a novel speech watermarking method is presented. In order to embed information we use the μ-law quantization method which is widely used in speech coding. In this algorithm, the embedding process is performed in Bark domain. To keep the watermark imperceptible and increase its robustness, the best bark and the optimum parameter μ are determined and used for data hiding. For controlling the quality of the watermarked signal Perceptual Evaluation of Speech Quality (PESQ) algorithm in an iterative method is used. The results show the higher robustness of the proposed technique in comparison with the traditional method.