基于声门特征的说话人识别研究-计算机应用技术专业论文.docxVIP

ABSTRAC ABSTRAC l ABSTRACT Speaker recognition(SR)，which identifies or verifies people by their voice，is regarded “the most ilatui al and convenient one among the methods of biometl its Most tare—ofthe— art fcatm—es which are used in SR systeln only reflect the flequency response of speake【’s vocal tract．However、they do not take into account their potential influences froln glottis’ riblation．1：1 this thesis，we first made a study of main points SR system，the current advancements of it，and the recent methods of utilizing glottal features in particular Then based on it，we proposed our models and algorithms to better utilizing glottal features in SR system．The main contribution of the work are as the followings： l The：nodel and the compensation algorithm，which could alleviate the cepstral features’influences from glottis’changes，is proposed for the SR system we have pointed out that the unaffected cepstral features would behavior[nero discriminative than the traditionsl ones．And we implement it by using both long-term and short—term glottal features to get the compensated ones 0ur algorithm improved the performance in ideal environment What is more，due to the robustness of glottal features，we also SUCCESSfully extended the algorithm to the complex multi．channel environment It showed good performance in the SRMC database with 303 person 2 We proposed our Parallel Gaussian Mixture Modds in model level to utilize the COl relation information between short—term glottal features and cepstraI features．0n the assumption that cepstral features Is independent from glottal ones。traditional speaker mod— els ignored the potential influences ofenlarging the intra-speaker distance Apart from them， Parallel Gaussian Mixture Modds jointly utilized the short—term glottal features and eepstral features．And with the priori knowledge of glottal features，we use one probability modeI to characterize such relationship．which could improve the performanee of the system 3．Glottal Information B