粒子群优化算法和支持向量机的电子音乐信号分类研究

Research on electronic music signal classification based on particle swarm optimization algorithm and support vector machine

电子音乐信号具有非平稳性变化特点,当前难以准确描述电子音乐信号的变化特点,使得电子音乐信号分类准确性不够,为了提高电子音乐信号分类准确性,提出粒子群优化算法和支持向量机的电子音乐信号分类方法。首先,分析当前国内对电子音乐信号的分类研究现状,并采集电子音乐信号;然后,对电子音乐信号分类进行噪声过滤操作,并提取电子音乐信号变化特征;最后,结合粒子群优化算法和支持向量机的优点,建立电子音乐信号分类模型,并采用多种类型的电子音乐信号进行分类性能测试实验。结果表明,粒子群优化算法和支持向量机可以有效区分各种电子音乐信号,电子音乐信号分类准确性高,使得电子音乐信号分类误差控制在实际应用区间内,同时,电子音乐信号分类准确性和效率要显著好于对比电子音乐信号分类方法。

The electronic music signals have the characteristic of non⁃stationary change,so it is difficult to accurately describe their change characteristics at present,which makes the electronic music signal classification accuracy unsatisfied.In order to improve the classification accuracy,an electronic music signal classification method combining particle swarm optimization algorithm and support vector machine is proposed.The current research status of electronic music signal classification in China is analyzed in this paper.The electronic music signals are collected first,and then subjected into noise filtering operation for the extraction of their change characteristics.The electronic music signal classification model is built in combination with the advantages of particle swarm optimization and support vector machine.Various types of electronic music signals are adopted to test the classification performance of the proposed model.The results show that the model based on particle swarm optimization algorithm and support vector machine can effectively distinguish all kinds of electronic music signals,so its classification accuracy is high,which keeps the electronic music signal classification error within a reasonable range for the practical application.The classification accuracy and efficiency obtained with the proposed method are better than those obtained with other methods.