计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (1): 12-14.DOI: 10.3778/j.issn.1002-8331.2011.01.004

• 博士论坛 • 上一篇    下一篇

基于修正Mel子带系数的文本无关的说话人识别

张庆芳1,2,赵鹤鸣2   

  1. 1.苏州经贸职业技术学院,江苏 苏州 215009
    2.苏州大学 电子信息学院,江苏 苏州 215006

  • 收稿日期:2010-09-08 修回日期:2010-11-19 出版日期:2011-01-01 发布日期:2011-01-01
  • 通讯作者: 张庆芳

Mel frequency subband coefficients based text independent speaker recognition

ZHANG Qingfang1,2,ZHAO Heming2   

  1. 1.Suzhou Institute of Trade & Commerce,Suzhou,Jiangsu 215009,China
    2.School of Electronics and Information Engineering,Suzhou University,Suzhou,Jiangsu 215006,China
  • Received:2010-09-08 Revised:2010-11-19 Online:2011-01-01 Published:2011-01-01
  • Contact: ZHANG Qingfang

摘要: 与文本无关的说话人识别具有用户使用方便、可应用范围较宽等优点,是当前说话人识别技术的研究重点。对文本无关说话人识别系统中的特征参数提取进行了研究,通过对Mel子带系数进行修正,增强了说话人识别系统中说话人之间的频带差异,提高了特征空间中类别的可分性,得到了更能体现说话人个性特征的Mel子带系数,从而提高了说话人识别系统的平均正确识别率。

关键词: 说话人识别, 与文本无关, 矢量量化, Mel子带

Abstract: Text-independent speaker recognition is an important branch research field of speaker recognition because of its ease-to-use and potential?applications in the information technology.This paper focuses on the problems of the feature extraction.Through modifying the Mel frequency subband coefficients,it?enhances?the?differences of the frequency between the people in the system,improves the ability of separating the classes in the feature space.And the parameters can emphasize people’s individuality and increase the average accuracy of recognition.

Key words: speaker recognition, text independent, vector quantization, Mel frequency subband

中图分类号: