摘要
音乐风格反映了音乐作品的总体基本特征,是音乐欣赏、分析、研究的基础。针对音乐风格流派分析技术的核心——旋律特征描述和特征匹配,发展了非毗邻层连接的前馈神经网络结构,给出了误差反传训练算法的分类器,并进行了实验研究。结果表明,非毗邻层连接的前馈神经网络结构有优越的识别性能和极快的收敛速度。
Music style reflects the total basic characteristic of music work,which is the foundation of music listening,anal sis and research.Aiming at the core of music style genre analytical technical：cantus characteristic description and character tic match,this paper develops the feedforward neural network structure of not-adjacent layer conjunction,gives the machi that the error margin is inverted spread the classification of training the calculate way and carries on the experimental researc The result indicates that the front feedback nerve network structure of not-adjacent layer conjunction has predominant fun tion on identify and splitting velocity on convergence.
出处
《计算机工程与应用》
CSCD
北大核心
2011年第27期246-248,共3页
Computer Engineering and Applications
关键词
音乐
风格
特征
神经网络
music
style
characteristic
nerve network
