基于支持向量机的说话人识别研究

Speaker Recognition Research Based on Support Vector Machine

解决说话人识别问题具有重要的理论价值和深远的实用意义，本文在研究支持向量机理论的基础上，采用支持向量机的分类算法实现说话人识别系统的训练和测试，并将小波去噪技术应用于说话人识别的预处理过程中，改善进入说话人识别系统的语音质量。实验表明，在说话人识别系统中，支持向量机结合小波去噪可以获得较好的识别率。

There are important theoretic value and far-reaching practical meaning to resolve the question of speaker recognition. SVM classification algorithm is used to realize the speaker recognition system＇s trainiog and testing based on the research of Support Vector Machine theory. And wavelet de-noising technology is also applied to the pre-process to improve the quality of speech signal input into the speaker recogrntion system. Experimental results have shown that the speaker recognition system＇s recognition rate was greatly increased by the combination of SVM and wavelet de-nosing