Based on the requirement of watermark recovery, watermarking techniques may be broadly classified under one of two schemes: blind watermarking scheme and non-blind watermarking scheme. For the blind watermarking schemes, the embedded watermark can be extracted even if the original signal is not available. Thus, the owner does not have to keep a copy of the original signal. For the non-blind watermarking scheme, the original signal is required for extracting the watermark and hence only the owner of the original signal will be able to perform the task.

 

In this lecture, two audio watermarking techniques:  the bit replacement approach and the time-frequency technique, are presented to exemplify the two approaches. The bit replacement approach belongs to the blind category and the time-frequency technique belongs to the non-blind watermarking scheme. The performance of the two schemes under attacks will be discussed.  The generic approach of using both channels of stereo audio signals will be illustrated.