Article(id=1242132993901404443, tenantId=1146029695717560320, journalId=1146031591421210625, issueId=1242132974196564162, articleNumber=null, orderNo=17, doi=10.3981/j.issn.1000-7857.2015.01.016, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1403193600000, receivedDateStr=2014-06-20, revisedDate=1415289600000, revisedDateStr=2014-11-07, acceptedDate=null, acceptedDateStr=null, onlineDate=1422861959993, onlineDateStr=2015-02-02, pubDate=1421078400000, pubDateStr=2015-01-13, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1422861959993, onlineIssueDateStr=2015-02-02, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1774077183458, creator=sys-migrate, updateTime=1774077183458, updator=sys-migrate, issue=Issue{id=1242132974196564162, tenantId=1146029695717560320, journalId=1146031591421210625, year='2015', volume='33', issue='1', pageStart='1', pageEnd='128', issueExtLink='null', onlineDate='null', pubDate='1421078400000', pubDateStr='2015-01-13', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=3, issueType=-1, specialIssue=null, createTime=1774077178760, creator='sys-migrate', updateTime=1774077178760, updator='sys-migrate', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=90, endPage=94, ext={EN=ArticleExt(id=1242132996724175298, articleId=1242132993901404443, tenantId=1146029695717560320, journalId=1146031591421210625, language=EN, title=Speaker recognition method based on combination of kernel functions of SVM, columnId=null, journalTitle=Science & Technology Review, columnName=null, runingTitle=null, highlight=null, articleAbstract=In speaker recognition systems, if the original data distribution is unknown, the choice of inappropriate kernel functions will result in poor support vector machine (SVM) learning performance. Thus a speaker recognition method based on a multi-grid search of parameters and a combination of kernel functions is proposed in this paper. First, the method constructs a hybrid kernel function by linearly weighted polynomial and RBF kernels. Then it proposes a multi-grid search method to adjust the weights, and thus the hybrid kernel function can adapt to the current data distribution. Finally, a SVM classifier is trained to obtain the classification results. Simulation experiments on TIMIT datasets and noisy datasets show that the recognition performance of SVM classifiers using a combination of kernel functions is better than that using linear kernels, polynomial kernels, and RBF kernels. Therefore, the proposed method can effectively improve the performance of speaker recognition systems., authors=FAN Chijie1 , SI Qiaomei1 , XU Yan2 , ZHANG Dan1 , CAI Chunhua1 , YU Xu3 , authorsList=FAN Chijie, SI Qiaomei, XU Yan, ZHANG Dan, CAI Chunhua, YU Xu, authorCompany=1. School of Engineering, Mudanjiang Normal University, Mudanjiang 157012, China;1 , 司巧梅1 , 徐岩2 , 张丹1 , 蔡春华1 , 于旭3 , authorsList=樊持杰, 司巧梅, 徐岩, 张丹, 蔡春华, 于旭, authorCompany=1. 牡丹江师范学院工学院, 牡丹江157012;
科技导报
| 研究论文 2015, 33(1): 90-94
基于组合核函数SVM的说话人识别方法
全屏
樊持杰1 , 司巧梅1 , 徐岩2 , 张丹1 , 蔡春华1 , 于旭3
作者信息
1. 牡丹江师范学院工学院, 牡丹江157012;
Speaker recognition method based on combination of kernel functions of SVM
Affiliations
出版时间: 2015-01-13
doi: 10.3981/j.issn.1000-7857.2015.01.016
文章导航
鉴于应用支持向量机进行说话人识别过度依赖于选择核函数的问题,提出一种基于组合核函数支持向量机(SVM)的说话人识别方法.对多项式核函数、径向基核函数进行线性加权,构建既具有全局核函数优点又具有局部核函数优点的组合核函数,并通过多重网格搜索调节权重系数使组合核函数适用于当前数据分布,确定组合核函数SVM 的最优参数,实现对说话人的有效识别.对TIMIT 数据集和含噪声数据集的仿真实验显示,基于组合核函数SVM 的说话人识别性能明显优于单一的多项式核函数、径向基核函数和线性核函数.
说话人识别
/
支持向量机
/
组合核函数
/
多重网格搜索
In speaker recognition systems, if the original data distribution is unknown, the choice of inappropriate kernel functions will result in poor support vector machine (SVM) learning performance. Thus a speaker recognition method based on a multi-grid search of parameters and a combination of kernel functions is proposed in this paper. First, the method constructs a hybrid kernel function by linearly weighted polynomial and RBF kernels. Then it proposes a multi-grid search method to adjust the weights, and thus the hybrid kernel function can adapt to the current data distribution. Finally, a SVM classifier is trained to obtain the classification results. Simulation experiments on TIMIT datasets and noisy datasets show that the recognition performance of SVM classifiers using a combination of kernel functions is better than that using linear kernels, polynomial kernels, and RBF kernels. Therefore, the proposed method can effectively improve the performance of speaker recognition systems.
speaker recognition
/
support vector machine
/
combination of kernel functions
/
multi-grid search
樊持杰, 司巧梅, 徐岩, 张丹, 蔡春华, 于旭.
基于组合核函数SVM的说话人识别方法.
科技导报,
2015
, 33
(1)
: 90
-94
.
DOI: 10.3981/j.issn.1000-7857.2015.01.016
FAN Chijie, SI Qiaomei, XU Yan, ZHANG Dan, CAI Chunhua, YU Xu.
Speaker recognition method based on combination of kernel functions of SVM[J].
Science & Technology Review ,
2015
, 33
(1)
: 90
-94
.
DOI: 10.3981/j.issn.1000-7857.2015.01.016
2015年第33卷第1期
PDF下载
256
52
引用本文
BibTeX
文章信息
doi: 10.3981/j.issn.1000-7857.2015.01.016
接收时间：2014-06-20
首发时间：2015-02-02
出版时间：2015-01-13
收稿日期：2014-06-20
修回日期：2014-11-07
https://castjournals.cast.org.cn/joweb/kjdb/CN/10.3981/j.issn.1000-7857.2015.01.016
引用本文
BibTeX
2种不同金属材料的力学参数
科 属数 种数 占总种数比例 属 种数 占总种数比例 鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78 小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39 多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39 红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87 小菇属 Mycena 11 5.26 光柄菇属 Pluteus 5 2.39 红菇属 Russula 17 8.13 栓菌属 Trametes 5 2.39
关闭全屏
BibTeX
EndNote
RefWorks
TxT
PDF下载
全屏
256
52
关闭全屏
北京市海淀区学院南路86号
010-62199257（仅加盟咨询）
qkjq@cast.org.cn
