【摘要】 基于传统的LPC倒谱特征和KC复杂性特征建立了一个说话人确认系统,采用了YOHOspeakerverification数据库训练模板和测试该说话人确认系统,取得了较好的说话人确认效果。实验证明:传统的基于线性理论基础上的说话人识别特征提取方法,与基于非线性理论的KC复杂性特征基本无相关性,因此这两类特征如能相互结合,有着良好的互补特性,能够大幅度的提高系统性能。这说明新提出的说话人的KC复杂性特征是一个非常有用的传统线性特征的有效辅助特征。更多还原

【Abstract】 The automatic speaker verification in this paper introduces one novel feature selection and extraction method, computation complexity features. Computation complexity feature can extract that nonlinear character of speech signal, which overcomes the disadvantage of the traditional linear feature extraction method. The corpus used in the tests is YOHO speaker verification corpus. Through the experiments, one conclusion comes that the combination of the nonlinear features and traditional linear features can reduce the verification errors markedly and gain more accurate results.更多还原