| 基于HMM／SVM的音频自动分类 |
| |
| 引用本文： | 史东承,韩玲艳,于明会. 基于HMM／SVM的音频自动分类[J]. 吉林工学院学报, 2008, 29(2): 178-182 |
| |
| 作者姓名： | 史东承 韩玲艳 于明会 |
| |
| 作者单位： | 长春工业大学计算机科学与工程学院 吉林长春130012 |
| |
| 摘 要： | 提出一种基于隐马尔可夫模型和支持向量机混合模型的音频分类方法，用于语音、音乐、语音＋音乐、静音4类音频分类。首先利用4个HMM分类器对音频进行初步分类，确定最可能的两种音频分类结果，再用相应的SVM分类器做最终判决。实验结果表明，隐马尔可夫模型和支持向量机的两级分类器分类性能较好。
|
| 关 键 词： | 音频分类 隐马尔可夫模型 支持向量机 |
| 文章编号： | 1674-1374(2008)02-0178-05 |
| 修稿时间： | 2007-09-10 |
Automatic audio stream classification based on hidden markov model and support vector machine |
| |
| SHI Dong-cheng,HAN Ling-yan,YU Ming-hui. Automatic audio stream classification based on hidden markov model and support vector machine[J]. Journal of Jilin Institute of Technology, 2008, 29(2): 178-182 |
| |
| Authors: | SHI Dong-cheng HAN Ling-yan YU Ming-hui |
| |
| Affiliation: | (School of Computer Science ＆ Engineering, Changchun University of Technology, Changchun 130012, China) |
| |
| Abstract: | A kind of auto stream classification based on HMM and SVM is put forward to break down the speech,music,speech music and mute four kinds of auto stream.First,the audio is simply classified with four HMM classifiers to determine it belong to which area,and then it is fixed with SVM classifier.Experimental results show that the classifier has good properties. |
| |
| Keywords: | audio classification hidden markov model(HMM) support vector machine. |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
