摘要
研究一种基于卷积神经网络(Convolutional Neural Network,CNN)和门控循环单元(Gated Recurrent Unit,GRU)相结合的环境声音分类方法。首先,分析CNN-GRU模型的基本结构;其次,探讨模型进行环境声音分类的数学原理;最后,采用ESC-50数据集在MATLAB平台上对所提方法进行测试。实验结果表明,CNN-GRU模型的准确率、精确率、召回率及F1值分别达到了0.92、0.91、0.89及0.90,验证了该模型在处理环境声音分类任务中的有效性和健壮性。
A methodology of environmental sound systematization based on the Convolutional Neural Network(CNN)and Gated Recurrent Unit(GRU)is studied.Firstly,the basic structure of the CNN-GRU model is analyzed;secondly,the mathematical principle of the model for environmental sound classification is discussed;finally,the proposed method is tested on the MATLAB platform using the ESC-50 dataset.The experimental results show that the accuracy rate,precision rate,recallrate and F,value of CNN-GRU model reach 0.92,0.91,0.89 and 0.90,respectively,which verifies the effectiveness and robustness of the model in processing environmental soundclassificationtasks.
作者
徐圣林
XU Shengin(China Pharmaceutical University,Nanjing 211198,China)
出处
《电声技术》
2024年第10期54-56,共3页
Audio Engineering
作者简介
徐圣林(1972-),男,本科,工程师,研究方向为计算机网络、电子信息、物联网应用。
