The universal development of computer and communication technologies and the spread of network on a global scale, has led to the appearance of numerous techniques for the transmission of confidential information such as cryptography. Because of the hacking and breakthrough of these technologies, the need for more sophisticated and difficult techniques and methods to transfer confidential information in a secure manner. Steganography is where secret information is hidden in a way that does not allow the third party to know hidden information. It is added with stream cipher to increase the security of information transmission and uses media such as, image, audio, text or video as a transmission medium for that information. The aim of this paper is to use a method to hide confidential information (image or audio) into audio file as a cover in such a way as to certify that hacking does not reach that information by making the size of the cover equal to the size of the original file. The audio file resulting from the masking process is similar to the original file and the human ear can not sense the change between them. By splitting the original audio cover into frames and applying the random function to the clutter of the hiding locations after separating the layers of the color image in the case of the confidential information image or fragmentation of the audio to be hidden into frames and then applying this function to select hiding locations. In the case of retrieval, the process is reversed where the embedded audio file is fragmented stego-audio and apply the random function to select hiding locations. In the audio segmentation process, we use a key that represents the size of the image to be hidden, as well as frames, and serves as a cryptographic key for the original voice to increase security. The success rate of image recovery was obtianed. The sound retrieval process was also good.

Steganography Stego-Audio PSNR SNR Stream Cipher.