Research Of Speaker Recognition Base On VQ And GMM Models

Speaker recognition is a kind of biological certification technology and it makes use of the speech coefficients which represent the speaker's physiological and physical feature to identify speaker. In the biological certification area, speaker recognition widely draws the attention because of its convenience, efficiency and accuracy. It can be applied to a number of fields, such as security, justice, military affairs, finance and services. Because of that, lots of scientific researchers are involved in the research, marking great development. However it is not ripe very much.This paper is mainly about a text-dependent speaker recognition system based on vector quantification (VQ) methods, a text-independent speaker recognition system based on Gaussian mixture models. We use MFCC coefficient as the feature parameter set.Mainly works in this paper: (1), The problem of the feature parameter pick-up in speaker recognition, expound the vocal mathematical model, LPC analysis, LPC, Mel-cepstral coefficient in detail.(2), It mainly introduce some different methods of the speaker recognition. The usability of the vector quantization (VQ) technique in speaker recognition is dissertation in detail. The essence of VQ is to use several special features to represent the whole feature in solution so that to achieve the aim of compressing and recognition .At the same time, this paper introduce the theory and carrying out of the GMM,(3),Research into the system. In VQ model, research into influence of the system capability on the code-size and the enactment about of threshold value. In GMM model, research into the order and enframe influence of the system capability.In the end, a conclusion of this thesis and the prospect of the future work are drawn.