Research Of Speaker Recognition Arithmetic Based On Pan Boolean Algebra

With the development of information technology, individual identification and information security become more and more important. Speaker recognition (SR) technique is one of biological technologies with which computer can automatically identify the individual based on human characteristics. Compared with traditional methods, the new method proposed in forgotten or this paper is more convenient, safer, and not easy to be replaced. Speaker recognition can be used under a lot of circumstances, such as the telephone electronic commerce, military wiretapping, information retrieval, and so on.This dissertation bases on <> of national defence researching fund has done deeply research on the method and application of Speaker Recognition. As both Speaker Recognition and Pan-Boolean algebra are effective logic abstract searching particular laws of things and according with people's comprehend. So it can be proved that applying Pan-Boolean algebra into Speaker Recognition is available.This dissertation begins with the analysis of speech signal pretreatment and does some points test, draws the characteristic parameter of the introduction pronunciation through MATLAB speech processing toolbox, gets rid of the aphonia segment and yawp of speech signal, offers effective speech segment for the distill of speech characteristic parameter. The study adopts MFCC and the difference as the characteristic parameter of Speaker Recognition to describe the nonlinear characteristic of the ear's hearing frequency and improves systematic discernment performance. Through analyzing the basic principle and system configuration of Speaker Recognition and reviewing the existing Speaker Recognition technique, we can apply the method of combining VQ, HMM, GMM and Pan-Boolean algebra in Speaker Recognition and constitute the Speaker Recognition System. VQ is easy and tend to realize, HMM can comminute the speech dynamically, it can embody not only the dynamic characteristic but also the stationary characteristic of the speech. GMM is a special HMM, it uses combination of many weighted Gaussian density function to approach the distributing density of speaker characteristic vector in characteristic space, so it is also a good way. For the systems based respectively on VQ, HMM, GMM have their own advantages and disadvantages, however, there exists some disadvantage if only uses single method Speaker Recognition. It can reflects its characteristics if uses them together instead of only using one of them. So this dissertation uses three different methods of Speaker Recognition before doing recognition on Pan-Boolean algebra and gets high rate of recognition.At last, the thesis gives a summary of my work and points out some problems and ideas for further research.