Embedding Limitations with Audio-watermarking Method Based on Cochlear-delay Characteristics

Abstract

We investigated embedding limitations with our proposed method of audio watermarking, which is based on the concept of embedding inaudible watermarks into an original sound by controlling its phase characteristics in relation to cochlear delay. We also improved the method by designing a parallel architecture for cochlear delay filters. We then evaluated our previous and improved methods to investigate embedding limitations by carrying out four objective experiments: LSD, PEAQ, bit-detection, and robustness tests. The results revealed that the methods could be used to inaudibly embed the watermarks into original signals and to accurately and robustly detect the embedded data from the watermarked signals. We also found that embedding limitations with the improved method amounted to 384 bps while that with our previous method amounted to 128 bps.