Audio Classification Algorithm Based on Nonlinear Characteristics Analysis

摘要：

In this paper, an audio classification algorithm based on nonlinear characteristics analysis is proposed. According to the nonlinearity of audio signals, the phase space reconstruction technology of nonlinear dynamics, the recurrence plot and the recurrence quantification analysis are introduced to audio characteristics analysis, and the audio signals are classified into four types: noise-like, transient, harmonic-like and mixed signal. Test results indicate that the proposed algorithm has good classification accuracy.

作者: Liyan Zhang Changchun Bao Xin Liu Xingtao Zhang Feng Bao Bing Bu

作者单位: Speech and Audio Signal Processing Laboratory, School of Electronic Information and Control Engineer Speech and Audio Signal Processing Laboratory, School of Electronic Information and Control Engineer

会议类型: 国际会议

会议名称: 2011亚太信号与信息处理协会年度峰会(APSIPAASC 2011)

会议地点: 西安

会议语种:英文

页码: 1-4

在线出版日期: 2011-10-18（万方平台首次上网日期，不代表论文的发表时间）