摘要： 针对支持向量机在大样本情况下训练速度慢的缺点，引入权重最优位置策略改进量子粒子群优化算法，通过改进的Michigan编码方案对语音参数进行编码，构造分类规则适应度函数实现基于加权量子粒子群的分类器设计。在说话人识别中的应用结果表明，该分类器具有较好的抗噪性能和较高的识别速度。

关键词: 说话人识别, 支持向量机, 量子粒子群优化, 分类器

Abstract: Aiming at the shortage of Support Vector Machine(SVM) slow practice speed in the case of large sample, this paper introduces weighted optimal position strategy to improve Quantum Particle Swarm Optimization(QPSO) algorithm, processes coding for voice parameters by improving Michigan coding scheme, and constructs new classified rule fitness function to realize designing of classifier based on weighted quantum particle swarm. Application results of speaker recognition show that this classifier has better performance of noise proof and recognition speed.

Key words: speaker recognition, Support Vector Machine(SVM), Quantum Particle Swarm Optimization(QPSO), classifier

中图分类号: 

• TP393