| 基于PCP特征的和弦识别研究与探讨 |
| |
| 引用本文： | 王峰,张雪英,李炳男.基于PCP特征的和弦识别研究与探讨[J].数学的实践与认识,2010,40(5). |
| |
| 作者姓名： | 王峰 张雪英 李炳男 |
| |
| 作者单位： | 1. 太原理工大学信息工程学院,山西,太原,030024
2. 飞力浦(中国)投资有限公司太原办事处,山西,太原,030024 |
| |
| 基金项目： | 山西太原理工大学校科技发展基金资助 |
| |
| 摘 要： | 讨论了音乐识别领域中和弦的四种不同识别方法,给出了基于PCP特征的和弦识别算法.使用PCP作为和弦的特征作为输入送至隐马尔可夫模型中训练,利用Baum-Welch算法估计模型参数,通过Viterbi算法得到正确和弦.通过实验获得了76%的识别率,验证了该算法的可行性.
|
| 关 键 词： | 计算机音乐 和弦 音级轮廓 隐马尔可夫模型 |
Research And Discuss of Chord Recognition Based on PCP |
| |
| WANG Feng,ZHANG Xue-ying,LI Bing-nan.Research And Discuss of Chord Recognition Based on PCP[J].Mathematics in Practice and Theory,2010,40(5). |
| |
| Authors: | WANG Feng ZHANG Xue-ying LI Bing-nan |
| |
| Affiliation: | WANG Feng,ZHANG Xue-ying,LI Bing-nan (1.College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,China) (2.Philips(China) Investment Company limited,Taiyuan Office,China) |
| |
| Abstract: | In this paper,four different methods in music recognition are discussed,a algorithm of chord recognition based On PCP is presented.For features we use PCP vectors,it is trained by HMM,estimated parameters by Baum-Welch algorithm,finally we acquired accurate chords by Viterbi.The experiment indicates that its chord recognition average accuracy reaches 76 percent,proves feasibility of this algorithm. |
| |
| Keywords: | computer music chord pitch class profile(PCP) hidden markov model(HMM) |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
