Speech/Music Discrimination Using Discrete Wavelet Transform and Neural Network

摘要：

With respect to the importance of audio signals,particularly in multimedia applications and differences in speech and music processing,it requires initially to classifying speech/music,before conducting any more advanced processing.Thus,many studies have conducted on this topic during the recent decade.In this paper,a system is presented which is based on six simple statistical features that are extracted in Discrete Wavelet Transform (DWT) domain.Also,Multilayer Perceptron (MLP) neural network is used as the classifier.Experimental results show that the accuracy of proposed system is about 99％.

关键词： speech/music discrimination discrete wavelet transform neural network

作者: Mehdi Purkhiabani Ali Farrokhi Mansour Sheikhan

作者单位: Dept.of Electrical Engineering Islamic Azad University,South Tehran Branch,Tehran,Iran

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 147-150

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）