摘要
针对当前乐曲节拍识别系统存在的一些缺陷,如精度低,误识率高等问题,为了获得更优的乐曲节拍识别结果,提出了基于音频指纹技术的乐曲节拍识别系统。首先分析当前乐曲节拍识别的研究进展,设计了乐曲节拍识别硬件系统的结构,然后对乐曲节拍信号进行采集和放大处理,建立乐曲节拍音频的指纹数据库,最后将待识别乐曲节拍指纹与数据库中的指纹进行匹配,根据匹配结果实现乐曲节拍识别。仿真实验结果表明,该系统提高了乐曲节拍识别精度,减少了乐曲节拍的漏识率,具有十分广泛的应用前景。
In view of some defects of the current music beat recognition system,such as low accuracy and high error rate,in order to obtain better music beat recognition results,a music beat recognition system based on audio fingerprint technology is proposed.Firstly,the research progress of music beat recognition is analyzed,and the structure of music beat recognition hardware system is designed.Then,the music beat signal is collected and amplified,and the fingerprint database of music beat audio is established.Finally,the fingerprint of music beat to be identified is matched with the fingerprint in the database,and the music beat recognition is realized according to the matching results,the accuracy of music beat recognition is high,and this system reduces the missing rate of music beat,and has a very wide application prospect.
作者
刘红梅
LIU Hongmei(College of Humanities and Arts,AKsu Vocational and Technical College,AKsu 843000,China)
出处
《微型电脑应用》
2021年第7期137-139,143,共4页
Microcomputer Applications
关键词
乐曲节拍
硬件系统
识别精度
漏识率
指纹数据库
tune beat
hardware system
recognition accuracy
leakage rate
fingerprint database
