Abstract: In this paper, a new robust digital audio watermarking algorithm against desynchronization attacks is proposed, in which the audio statistic characteristics and synchronization code are utilized. Firstly, the origin digital audio data are segmented and then each segment is cut into two sections. Secondly, with the spatial watermarking technique, synchronization code is embedded into the statistic average value of audio samples in the first section. Finally, the DWT is performed on the second section, and the digital watermark is embedded into the statistic average value of low frequency components.