华南理工大学学报(自然科学版) ›› 2015, Vol. 43 ›› Issue (5): 100-106.doi: 10.3969/j.issn.1000-565X.2015.05.016

• 计算机科学与技术 • 上一篇    下一篇

Logistic 核函数及其在语音识别中的应用

刘晓峰1 张雪英2† Zizhong John Wang3   

  1. 1. 太原理工大学 数学学院,山西 太原 030024; 2. 太原理工大学 信息工程学院,山西 太原 030024;3. Departmentof Mathematics and Computer Science,Virginia Wesleyan College,Norfolk 23502,Virginia,USA
  • 收稿日期:2014-10-08 修回日期:2014-12-19 出版日期:2015-05-25 发布日期:2015-05-07
  • 通信作者: 张雪英(1964-),女,教授,博士生导师,主要从事语音信号处理研究. E-mail:tyzhangxy@163.com
  • 作者简介:刘晓峰(1979-),男,讲师,博士生,主要从事智能计算研究. E-mail: liuxinyu1206@163. com
  • 基金资助:

    国家自然科学基金资助项目(61072087)

Logistic Kernel Function and its Application to Speech Recognition

Liu Xiao-feng1 Zhang Xue-ying2 Zizhong John Wang3   

  1. 1. College of Mathematics,Taiyuan University of Technology,Taiyuan 030024,Shanxi,China;2. College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,Shanxi,China;3. Department of Mathematics and Computer Science,Virginia Wesleyan College,Norfolk 23502,Virginia,USA
  • Received:2014-10-08 Revised:2014-12-19 Online:2015-05-25 Published:2015-05-07
  • Contact: 张雪英(1964-),女,教授,博士生导师,主要从事语音信号处理研究. E-mail:tyzhangxy@163.com
  • About author:刘晓峰(1979-),男,讲师,博士生,主要从事智能计算研究. E-mail: liuxinyu1206@163. com
  • Supported by:
    Supported by the National Natural Science Foundation of China(61072087)

摘要: 核函数是支持向量机(SVM)的核心,直接决定着 SVM 的性能. 为提高 SVM 在语音识别问题中的学习能力和泛化能力,文中提出了一种 Logistic 核函数,并给出了该Logistic核函数是 Mercer 核的理论证明. 在双螺旋、语音识别问题上的实验结果表明,该Logistic 核函数是有效的,其性能优于线性、多项式、径向基、指数径向基的核函数,尤其是在语音识别中,该 Logistic 核函数具有更好的识别性能.

关键词: Logistic 核函数, 语音识别, 支持向量机, Mercer 核

Abstract: Kernel function is the core of support vector machine (SVM) and directly affects the performance of SVM. In order to improve the learning ability and generalization ability of SVM for speech recognition,a Logistic kernel function,which is proved to be a Mercer kernel function,is presented. Experimental results on bi-spiral and speech recognition problems show that the presented Logistic kernel function is effective and performs better than linear,polynomial,radial basis and exponential radial basis kernel functions,especially in the case of speech rec-ognition.

Key words: Logistic kernel function, speech recognition, support vector machines, Mercer kernel