Automatic Identification of Chinese Musical Instruments

摘要：

This paper addresses the problem of identifying the Chinese musical instruments in raw audio data. Given the rapidly increasing importance of digital music distribution, as well as the fact that large web-based music collections are continuing to increase in size exponentially, it is obvious that the ability to effectively navigate with these large collections is very important and desirable. Automatic musical instrument identification (MII) could be very helpful for such large music database management. In this paper, algorithms for automatic MII are explored. Five kinds of Chinese musical instruments were selected for this work. Two sets of features, STFT-based feature and MFCC-based feature, are extracted from waveform files. Then a Bayesian network classifier was trained. Experimental results showed that the average recognition rate was high and the classifier using MFCC-based feature could get better performance than the classifier using STFT-based feature.

关键词： Chinese Musical Instruments Instrument Identification

作者: Song Lin Wen Wu Lingyun Xie

作者单位: the Communication Acoustics Laboratory in Communication University of China, Beijing 100024 China

会议类型: 国际会议

会议名称: The Second International Symposium on Intelligence Computation and Applications(ISICA 2007)(第二届智能计算及其应用国际会议)

会议地点: 武汉

会议语种:英文

在线出版日期: 2007-09-21（万方平台首次上网日期，不代表论文的发表时间）