为提高清浊音判决的准确度,本文提出了一种利用计算机仿真产生训练样本,采用多指标综合评价方法,运用支持向量机产生判决准则的新方法。该方法克服了传统判据标准的缺陷,试验表明,该方法的分类性能要优于其他传统方法。 In order to improve the accuracy of voiced / unvoiced decision, this paper presents a new method of using computer simulation to generate training samples, using multi-index comprehensive evaluation method and using support vector machine to generate decision criteria. This method overcomes the shortcomings of the traditional criterion. The experimental results show that the proposed method is superior to other traditional methods in classification performance.