Show simple item record

dc.contributor.advisor史晓东
dc.contributor.author许永林
dc.date.accessioned2016-02-14T08:18:23Z
dc.date.available2016-02-14T08:18:23Z
dc.date.issued2005-08-11 23:27:01.0
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/52092
dc.description.abstract语音识别是模式识别领域中重要的一个研究方向,其中关键词识别技术是最近一二十年人机接口领域研究的热点,其在监听、自动电话接听、语音检索以及语音录入等方面有广泛的应用前景。传统的关键词识别系统是建立在隐马尔可夫模型(HMM)的基础之上的,语音信号经过特征提取后得到特征向量,通过隐马尔可夫模型对观察值序列(即特征向量)进行训练和分类,从而获得在统计概率意义上的最佳状态序列,最终识别出未知的语音。支持向量机(SVM)作为一种强大的机器学习理论,在多维非线性模式分类的应用中己经取得非常好的性能。本文将SVM分类技术嵌入到基于HMM的语音识别架构中,从而构造出一种混合的SVM/HMM语音识别系统。同时还解...
dc.description.abstractSpeech recognition is one of the important branches of pattern recognition, while keyword spotting becomes an active hotspot in the field of human-computer interface researches within past two decades. It has very wide application in related fields, such as automatic monitor, automatic telephone answering, speech search and speech recording. Traditional speech recognition systems are based on the ...
dc.language.isozh_CN
dc.relation.urihttps://catalog.xmu.edu.cn/opac/openlink.php?strText=10353&doctype=ALL&strSearchType=callno
dc.source.urihttps://etd.xmu.edu.cn/detail.asp?serial=10667
dc.subjectHMM/SVM
dc.subject关键词识别
dc.subjectViterbi-Beam搜索
dc.subjectHMM/SVM
dc.subjectKeyword Spotting
dc.subjectViterbi-Beam search
dc.title基于SVM/HMM的关键词识别系统研究
dc.title.alternativeResearch on Keyword Spotting System Based on SVM/HMM
dc.typethesis
dc.date.replied2005-08-11
dc.description.note学位:工学硕士
dc.description.note院系专业:计算机与信息工程学院计算机科学系_计算机应用技术
dc.description.note学号:200228040


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record