摘要
为了改善当前音色识别系统识别准确率低、时间开销较大等问题,结合人工智能技术提出并设计基于人工智能的钢琴编曲音色识别系统。首先给出系统整体架构和功能模块设计方案,使用数字化识别方法提取钢琴乐曲音色特征矩阵,得知钢琴特征矩阵不同于其他乐器;接着分析音色倍频点,得到音强包络曲线;最后结合钢琴音色特征矩阵和音强包络曲线有效区分不同乐曲音色的差异,实现钢琴编曲音色识别。实验结果表明,利用所提系统进行钢琴编曲音色识别具有较高的识别准确率,值得进一步推广和应用。
A piano arrangement timbre recognition system based on artificial intelligence technology is proposed and designed to improve the problems of low recognition accuracy and high expenditure of time in the current timbre recognition system.The design scheme of the overall architecture and functional module of the system are given.The feature matrix of the piano music timbre is extracted by means of the digital recognition method,and it is known that the piano feature matrix is different from other instruments.The timbre frequency doubling point is analyzed and the envelope curve of timbre intensity is obtained.The differences of different music timbres are distinguished in combination with the piano timbre feature matrix and the envelope curve of sound intensity,so as to realize the timbre recognition of piano arrangement.The experimental results show that the proposed system has high recognition accuracy for the piano arrangement timbre recognition,which is worth further promotion and application.
作者
仝智倍
TONG Zhibei(College of Applied Engineering,Henan University of Science and Technology,Sanmenxia 472000,China)
出处
《现代电子技术》
北大核心
2020年第4期183-186,共4页
Modern Electronics Technique
基金
河南省自然科学基金项目(182300410265)
河南省社科联、河南省经团联调研项目(SKL-2016-986)
关键词
钢琴编曲
音色识别系统
人工智能
音色特征提取
音强包络曲线
实验分析
piano arrangement
timbre recognition system
artificial intelligence
timbre feature extraction
sound intensity envelope curve
experiment analysis
