Audio Non-watermarking Based On The Audio Signal Feature

In recent years the fast development of digital audio processing technology and the popularization of intemet make it dead easy for people to accesse,edit and distribute digital audio works. The copyright protection of digital audio works becomes a burning question. As a potential solution,digital audio watermarking technology is paid extensive attention by researchers at home and abroad and becomes an international academic focus. But at present, the proposed watermarking algorithm in the time domain, spatial domain or frequency domain is to embed the watermark information embedded in these modified original digital works,so the method will lead to a certain degree of distortion and the conflicts of imperceptible and robust of the digital watermarking. As a solution to this contradiction digital watermarking has been an effective way to attention. But the focus areas of the research is in the image, and the "embedded" watermark are mostly meaningless. The meaningless watermark can only be statistical to determine whether there is a watermark and can not directly give information of the copyright symbol to us. To solve the above problem, some scholars use the audio signal characteristics to "embedded" meaningful watermarking in the original digital works and has get robust of the method. In this paper we also use these audio features to "embedded" watermarking, the main work is the two aspects in the following:(1) The human auditory system is very sensitive,so to achieve the balance between robust of audio watermarking system and transparency is very difficult. In order to solve this problem in principle, we studied "audio watermarking" technology. First, we proposed an improved multi-stage vector quantization algorithm by studying vector quantization algorithm in the past,and the improved multi-stage vector quantization reduces the complexity of the computational and reduces the running time of the program. Then we proposed algorithm that a linear prediction coefficient based on multistage vector quantization of speech watermarking because of the stationarity of short-term using of the natural speech signal. And the simulation results verify that the algorithm has good robustness. (2) Dealing with low frequency components is the most difficult part of dealing with the music. We will destroy the music whether recording or mix the music if we are in a little carelessness. So most of the pirates would not consider to attack the low-frequency components of the audio signal.In this paper, we use zero-watermarking algorithm based on the low frequency components in the audio signal without modifying any data of the audio signal to "embedded" a meaningful number of zero-watermark in the audio signal. Simulation results verify that the algorithm has good robustness and can effectively resist the audio signal attack such as the zoom range, weight, and low pass filtering and so on.