Research Of Spatial Audio Coding Algorithm

With the continuous development of 3D video technology,people's requirements for spatial audio technology matching 3D video is increasingly strong.spatial audio could enhance audience's spot feeling,and bring audience a visual feast together with 3D video technology.The research on spatial audio technology has become one of the focus in the field of multimedia technology.Firstly,the introduction of the research status on spatial audio techniques and the main research contents and significance of this paper are given.Then,the theory knowledge of spatial audio,research contents,requirements and applications,main audio technology,output layout and so on are described in this paper.Next we propose a new encoding method based on spatial sparse representation by using a multi-channel array model for the problem of mass data storage and compression.The signal acquired in this way has a stronger correlation.Then the signal is processed using the method of compressed sensing,which can simplify the multi-channel model and greatly reduce the data quantity of storage and transmission.Through evaluate the multi-channel model,we should verify that with the increase of the number of channels,the sound quality will become better and better and find the objective evaluation parameters.On this basis we will find the redundant channels and the important channels,and achieve the purpose of simplifying multi-channel model.To solve this problem,this paper established the models of 5.1,7.1 and 22.2 channels through object experiment,and have an objective evaluation to the reconstructed sound.By comparing the quality score,covariance power spectrum density and correlation degree of the signal from reconstructed multi-channel models to original signal,we come to the conclusion that the sound quality will become better with the increase of channel amount.In this paper,the PEAQ(Perceptual Evaluation of Audio Quality)is used for the evaluation of the signal quality,and the experimental verifications are carried out respectively on the musical instrument and the speech signal.Because subjective experiment is time consuming,have very high requirements for the test conditions and test personnel and not easy to operate,the method of objective evaluation and simulation analysis is used to evaluate the test results in this paper.At the last,the work of this paper is summarized,and the future work direction is discussed.