摘要
为了实现音乐情感识别的舞台灯光自动控制,需对音乐文件进行情感标记;针对人工情感标记效率低、速度慢的问题,开展了基于音乐情感识别的舞台灯光控制方法研究,提出了一种基于支持向量机和粒子群优化的音乐情感特征提取、分类和识别算法;首先以231首MIDI音乐文件为例,对平均音高、平均音强、旋律的方向等7种音乐基本特征进行提取并进行标准化处理;之后组成音乐情感特征向量输入支持向量机(SVM)多分类器,并利用改进的粒子群算法(PSO)优化分类器参数,建立标准音乐分类模型;最后设计灯光动作模型,将新的音乐文件通过离散情感模型与灯光动作相匹配,生成舞台灯光控制方法;实验结果表明了情感识别模型的有效性,与传统SVM多分类模型相比,明显提高了音乐情感的识别率,减少了测试时间,从而为舞台灯光设计人员提供合理参考。
In order to realize the automatic control of stage lighting of music emotion recognition,it is necessary to mark the emotion of music file.Aiming at the problem of low efficiency and slow speed of artificial emotion marking,the stage lighting control method based on music emotion recognition is studied,and a music emotion feature extraction,classification and recognition algorithm based on support vector machine and particle swarm optimization is proposed.Taking 231 MIDI music files as an example,the basic features of music,such as average pitch,average intensity and melody direction,are extracted and standardized.Then and then the multi-classifier of support vector machine(SVM)is formed,and the parameters of classifier are optimized by using improved particle swarm optimization(PSO)algorithm to establish standard music classification model.Finally,the lighting is designed action model,the new music file is matched with the lighting action through the discrete emotion model,and the stage lighting control method is generated.Experimental results show the effectiveness of the emotion recognition model,compared with SVM traditional multi-classification model,obviously improve the recognition rate of music emotion,reduce the test time,so as to provide a reasonable reference for stage lighting designers.
作者
段中兴
严洁杰
Duan Zhongxing;Yan Jiejie(School of Information&Control Engineering,Xi'an University of Architecture and Technology,Xi'an 710055,China;State Key Laboratory of Green Building in Western China,Xi'an University of Architecture and Technology,Xi'an 710055,China)
出处
《计算机测量与控制》
2020年第11期95-100,共6页
Computer Measurement &Control
基金
国家自然科学基金(51678470)。
关键词
音乐情感分类
支持向量机
粒子群优化
music emotion classification
support vector machine
particle swarm optimization
