Audio Watermarking Based on Hybrid Low and High Wavelet Frequencies

Abstract

Transform-domain digital audio watermarking has a performance advantage over time-domain watermarking by virtue of the fact that frequency  transforms offer better exploitation of the human auditory system (HAS). In this research paper an adaptive audio watermarking is proposed based on the low and high wavelet frequencies band (LF, HF). The embedded watermark can be of any types of signal (text, audio and image). The insertion of the watermark data is performing in a frequency domain after applying discrete wavelet transformation on the cover audio segments. The normalize correlation and the signal to noise ratio metrics are used to test the performance of the proposed method in terms of the robustness and imperceptibility. Test results show that an improvement of the robustness against some type of attacks when the watermark is adaptively embedded in a different wavelet bands.