基于相空间重构和支持向量机的和弦识别

Chord Recognition Based on Reconstructed Phase Space and SVM

利用相空间重构方法提取和弦音频中非线性特征参量,将部分参量作为训练集来构造支持向量机(SVM)分类器,另一部分作为测试集进行识别效果的检验,仿真实验表明,基于相空间重构和支持向量机的方法能够有效地进行和弦的识别,可以进一步应用到连续的音乐识别当中。

This paper presents a method for chord recognition by utilizing nonlinear parameter based on reconstructed phase space technique.After extracting parameters from chord signals,SVM classifier are developed using part of the nonlinear parameters and the rest are used as test data to check out the recognition efficiency.The results demonstrate that the method based on Reconstructed phase Space and SVM could be used in chord recognition effectively.