基于内容的音频检索算法  

Summarization of content-based audio retrieval algorithm

作　　者：尚永强[1] 张琳梅[1] 徐大伟[1] 

机构地区：[1]信阳农业高等专科学校计算机科学系,河南信阳464000

出　　处：《河南科技学院学报》2009年第3期69-72,共4页Journal of Henan Institute of Science and Technology(Natural Science Edition)

摘　　要：介绍了基于内容的音频检索系统的基本结构,并在研究国内外相关文献的基础上,分析了音频检索算法的主要特点,划分为以下几种类型:近似字符串匹配算法、基于树结构的音频检索算法、基于动态时间规整的音频检索算法、基于隐马尔克夫模型的音频检索算法以及其它音频检索算法.The basic structure of Content-based audio retrieval system is introduced, and the audio retrieval algorithms is summarized based on an in-depth analysis of domestic and foreign relevant literature. According to the main features of the algorithms, they are divided into the following several types： approximate string matching algorithm, tree-based algorithm, dynamic time warping algorithm, HMM-based algorithm, and other kind of the audio retrieval algorithms.

关 键 词：音频检索 近似字符串匹配 动态时间规整 隐马尔可夫模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]