An audio information hiding scheme with high — Rui Xu2, Xue-mei Li1 capacity based on compressed sensing and psychoacoustic model

Audio information hiding is a technique, which is used to hide secret message into the audio flows. In this paper, an audio information hiding scheme with high-capacity is proposed, which uses both Compressed Sensing (CS) and psychoacoustic model algorithms for improving the capacity of hidden message and the auditory imperceptibility. The scheme first applies CS technology to compress the secret speech and reduce the embedded information bit rate, and then employs psychoacoustic model algorithm for determining embedding strength of secret information. The scheme takes advantages of DWT and synchronization signal of m-sequence to realize blind extraction. Experiments results show that this scheme not only has good imperceptibility robustness, but also has high-capacity and blind extraction.