ABSTRACT
A novel algorithm for embedding perceptually transparent watermark into uncompressed, raw audio sequences is presented. The scheme takes advantage of temporal masking phenomenon in HAS in order to embed watermark data below the masking threshold of audio signal. Detection of the watermark is done by blind detection, without using the original audio. No complex transforms to and from frequency domain are performed neither on embedding or extraction part of the proposed scheme, resulting in the calculation simplicity of embedding and detection process. The scheme proved to be robust against mp3 and AAC compression and the embedded watermark is perceptually transparent, introducing no audible distortion and resulting in a BER of about 2%.
