Aiming at the problem of the picture authentication,a method of the picture interpretation with intelligence speech based on the digital watermark is put forward.The method firstly saves the characteristic information of the target speech,and then interprets the speech to note signal and embeds it into the picture with digital watermark.After extracting and recovering,the speech can be synthesized with appropriative speech synthesis system,and then it can be compared with the original characteristics.The experiment verifies the validity of the method.