计算机应用 ›› 2005, Vol. 25 ›› Issue (04): 968-970.DOI: 10.3724/SP.J.1087.2005.0968

• 典型应用 • 上一篇    下一篇

乐音识别方法及应用

徐国庆1,2,杨丹1,王彬洁3,文俊浩1   

  1. 1.重庆大学软件学院; 2.武汉化工学院; 3.武汉化工学院计算机科学与工程学院
  • 发布日期:2005-04-01 出版日期:2005-04-01

Method and the application of musical tone recognition

XU Guo-qing1,2, YANG Dan1, WANG Bin-jie3, WEN Jun-hao1   

  1. 1. Faculty of Software Engineering,Chongqing University,Chongqing 400030,China; 2. College of Computer Science and Engineering,Wuhan Institute of Chemical Technology, Wuhan Hubei 430074,China; 3.Wuhan Institute of Chemical Technology, Wuhan Hubei 430074,China
  • Online:2005-04-01 Published:2005-04-01

摘要: 通过研究乐音的声音和物理特性,提出一种识别乐音信号的方法,该方法实现在频域的 精确定位,在基音频率检测上优于单一小波方法,在识别效果方面优于DTW方法,并开发了乐音识 别和自动作曲系统。此方法可以为乐音识别提供参考。

关键词:  乐音识别, 端点检测, 小波变换, 快速傅立叶变换

Abstract: This thesis investigate the musical and physical characters of the musical sound, and advances a new method to recognize the musical tone firstly. This method realizes the precise frequency resolution. It is more advantageous than the pitch detection method and the single wavelet method. It is also more effective than Dynamic Time Warping method. This method is ideal for the pitch recognition of the musical sound. It is first advanced in the technology domain and it has a high application value.

Key words: musical tone recognition, edge detection, DWT, FFT, melodize automatically

中图分类号: