改进的BP算法在音乐分类中的应用研究  

Study on the application of improved BP algorithm in music classification

作　　者：黄欢欢 卢亚玲 HUANG Huan-huan;LU Ya-ling(School of Electrical and Electronic Engineering,Wuhan Polytechnic University,Wuhan 430023,China)

机构地区：[1]武汉轻工大学电气与电子工程学院,湖北武汉430023

出　　处：《武汉轻工大学学报》2018年第4期70-75,共6页Journal of Wuhan Polytechnic University

摘　　要：音乐分类研究已经持续多年,但目前的检索效率并不理想。利用倒谱系数法提取音乐特征,通过基于弹性梯度、动量项、学习因子改进的BP算法,对民歌、古筝、摇滚音乐、流行音乐四类音乐进行音乐分类,经过对比BP算法的分类结果,改进的BP算法在识别率及鲁棒性等方面均有显著提升。Music classification research has been going on for many years, but the current retrieval efficiency is not satisfactory. Using the cepstrum coefficient method to extract the musical features, the four types of music, folk songs, guzheng, rock music and pop music, are classified by the BP algorithm based on the elastic gradient, momentum and learning factors. After comparing the classification results of the BP algorithm, the improved BP algorithm is significantly improved in the recognition rate and robustness.

关 键 词：音乐 分类方法 动量 BP算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TN912[自动化与计算机技术—控制科学与工程]