作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2010, Vol. 36 ›› Issue (8): 194-196. doi: 10.3969/j.issn.1000-3428.2010.08.068

• 人工智能及识别技术 • 上一篇    下一篇

NAP序列核函数在话者识别中的应用

邢玉娟1,李 明2   

  1. (1. 甘肃联合大学理工学院,兰州 730000;2. 兰州理工大学计算机与通信学院,兰州 730000)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-04-20 发布日期:2010-04-20

Application of NAP Sequence Kernel Function in Speaker Verification

XING Yu-juan1, LI Ming2   

  1. (1. School of Science and Engineering, Gansu Lianhe University, Lanzhou 730000; 2. School of Computer and Communication, Lanzhou University of Technology, Lanzhou 730000)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-04-20 Published:2010-04-20

摘要: 针对话者识别系统中特征向量不定长和交叉信道干扰等问题,提出一种基于超向量的扰动属性投影(NAP)核函数。该函数是一种新型的序列核函数,使支持向量机能在整体语音序列上分类,移除核函数空间中与话者识别无关的信道子空间信息。仿真实验结果表明,该函数可有效提高支持向量机的分类性能和话者识别系统的识别准确率。

关键词: 扰动属性投影, 高斯混合模型超向量, 话者识别, 支持向量机

Abstract: For the sake of solving the problem of variable-length feature vectors and channel impact which existed in speaker verification, a novel kernel function based on Gaussian Mixture Model(GMM) supervector, called Nuisance Attribute Projection(NAP) mapping KL divergence linear kernel function, is proposed in this paper. This function can not only be in the interest of enabling Support Vector Machine(SVM) to classify on whole audio sequences, but also has the benefit that channel subspace, which causes variability, is removed in kernel space. By doing so, the classification performances of SVM and verification accuracy of system are improved excellently. Simulation experimental results demonstrate the effectiveness of this kernel function.

Key words: Nuisance Attribute Projection(NAP), Gaussian Mixture Model(GMM) supervector, speaker verification, Support Vector Machine(SVM)

中图分类号: