计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (13): 71-73.

• 研发、设计、测试 • 上一篇    下一篇

基于支持向量机的民族语口音识别

徐翔俊1,2,毕福昆2,杨 鉴2   

  1. 1.云南警官学院北院,昆明 650223
    2.云南大学 信息学院 信息与电子科学系,昆明 650091
  • 收稿日期:2007-08-30 修回日期:2007-10-29 出版日期:2008-05-01 发布日期:2008-05-01
  • 通讯作者: 徐翔俊

Linguistic minority accents recognition using Support Vector Machine

XU Xiang-jun1,2,BI Fu-kun2,YANG Jian2   

  1. 1.Yunnan Police Officer Academy,Kunming 650223,China
    2.Department of Information & Electronic Science,Yunnan University,Kunming 650091,China
  • Received:2007-08-30 Revised:2007-10-29 Online:2008-05-01 Published:2008-05-01
  • Contact: XU Xiang-jun

摘要: 基于母语分别为傣语、傈僳语、纳西语和汉语普通话的发音人所发汉语普通话语句,利用支持向量机进行民族口音识别研究。实验结果表明,采用对每个语句提取12维MFCC参数、9维基频派生参数和9维短时平均能量派生参数等作为特征参数集及使用高斯径向基函数支持向量机的方法,男、女声的口音识别率均超过93%。

关键词: 口音识别, 支持向量机(SVM), 高斯径向基函数

Abstract:

In this paper,we use Support Vector Machine with radial basis function to classify the accents of Mandarin Speech which includes the native speakers and non-native speakers from Dai,Lisu and Naxi in Yunnan.In the feature-extraction strategy,we estimate Mel frequency cepstral coefficients,pitch and energy,and calculate their correlative statistical features.Experimental results show that this scheme can achieve an average accuracy of 93%.

Key words: accents recognition, Support Vector Machine(SVM), Gaussian radial basis function