The protection of Intellectual Property Rights (IPR) arise from The leakage problem which involves the illegal duplication, unlawful tampering, and wrongful distribution of multimedia files. Digital watermarking hides data (watermarks) in a matter that is invisible or inaudible to the human observer or listener respectively. An audio watermarking system have been proposed in this thesis in order to embed digital watermarks in the audio signal while retaining its perceptual quality to the listener. The system uses different techniques: Low Bit Encoding (LBE) in the time domain, and the human auditory characteristics in the frequency domain. In the frequency domain, the Fast Fourier Transform (FFT) with a segmentation approach is used to embed the watermarks. A good result in quality is achieved by using the frequency domain technique because the chosen locations in the spectrum of the frequency domain made no major affect on the perceptual quality of the audio file. The imperceptibly of the hidden watermarks are measured by using peak signal to noise ratio (PSNR) metric. It has been proven that it has good quality (PSNR above 40 dB). To increase the system security the original audio file is not used in the extraction process and a secret key file is also generated in order to extract the embedded watermarks. An attack is used in the proposed work in the form of added noise. The result of this attack shows that the embedded watermarks can be fully reconstructed with its redundancy under channel SNRc of at least 35 dB. A listening test was used1* to prove the system transparency, the tested sample has different backgrounds and they prove that the process was transparent.