Research On Voiceprint Recognition System And Pattern Recognition Algorithm

The identity recognition technology based on biometric characteristics was animportant issue in the recent international research. Voiceprint recognition, in whichthe identity of the speaker was determined by voice recognition, had the widespreadapplication value in the fields of System Security Authentication, JudicialIdentification, Electronic Interception，etc.Voiceprint recognition which was a type of voice recognition can be classifiedinto two categories, speaker verification and speaker identification from the view ofapplication as well as recognition with relevant test and recogntion with irrelevantfrom the view of recognition condition. In voiceprint recognition individualinformation features were focused, ignoring the content of the voice signal. Therewere two key technologies in voiceprint which were feature extraction in which thespeaker’s voice characteristics were described by the feature parameters extractedfrom the voice signal in acoustic or statistical terms, as well as recognition model bywhich robot could learn and memorize the speaker’s characteristics in order to realizethe recognition of the speaker.This paper demonstrated the principles of voiceprint recognition technology andemphasizes on studying as follows:(1) Extraction of voice signal: keynote period, zero-crossing rate, brightness,Linear Prediction Coefficients, LPC, Linear Cepstral Prediction Coefficients, LPCC,Mel-Frequency Cepstrum Coefficients, MFCC, etc(2) voiceprint recognition approaches and models: Gaussian Mixture Models,Implicit Markov Model, Vectorization Model, Artificial Neural Network(ANN)Model, Support Vector Machine Model. The recognition effect of the existing algorithms is susceptible to environmentalnoise, voice variation and other factors. According to this problem, Basing on theexisting voice pattern recognition technology, this paper improved the calculationmethod and did plenty of experiments (experiments use the voice data which isgathered under different noise environment of the early, middle and late periods of aday. And during the one-month voice collection, the person got bad cold caused voicevariation.). Experiment results showed that the improved algorithm can effectivelyovercome the impact of environment noises and voice variation. This paper did workas follows:(1) MFCC extracted form voice characteristics was improved, while the impacton voice signal was reduced by applying frequency masking algorithm; The accuracyof calculation was resolved in LF, HF, MF respectively, in turn the recognition ratewas improved to a certain extent.(2) A novel method of Initial codebook selection was presented by improvingVector Quantization Model: With the Hypersphere Extreme Selection Method and theimproved LBG algorithm, the produce of empty cell during the converging processwas reduced with an effectively improved recognition rate.(3) Applying Labview graphical programming into the system of voiceprintrecognition, a graphical virtual instrument panel was established by using powerfulgraphical environment and hardware resources in order to realize the real timeselection and analysis of voice signal and the modularization, intellectualizationthrough other softwares at the advantage of low cost, convenient analysis ofstatistic,good management.