一个基于汉语单音节的讲话者识别系统的设计

A Speaker Recognition System Design Based on Chinese Character Syllable

结合汉语字单音节的特点，以０～９十个数字的音节设计局部码本，提出了在训练和识别过程中的ＦＳＶＱ算法，使讲话者识别系统的正识率达到９５％．

Based on property of chinese character being a single syllable, we study FSVQs algorithm of training codebook and speaker recognition in (0～9) ten chinese characters being designed partial codebook ,so that to yield speaker recognition accuracy rate 95%.