基于RNN和K-means的音频智能分类方法  

Audio Intelligent Classification Method Based on RNN and K-means

作　　者：胡彦红 范凯燕 HU Yanhong;FAN Kaiyan(Zhengzhou University of Science and Technology,Zhengzhou 450000,China)

机构地区：[1]郑州科技学院,河南郑州450000

出　　处：《电声技术》2024年第11期24-26,共3页Audio Engineering

摘　　要：针对音频信号分类问题,提出一种结合循环神经网络(Recurrent Neural Networks,RNN)和K-means聚类算法的音频智能分类方法。该方法通过RNN模型提取音频信号的时间序列特征,利用K-means聚类算法聚类分析音频特征,以增强音频分类的健壮性和全面性。使用Urban Sound8K数据集评估方法。结果显示,该方法在准确率、召回率、F_(1)值等指标上均优于标准RNN模型。To solve the problem of audio signal classification,an intelligent audio classification method combining Recurrent Neural Networks(RNN)and K-means clustering algorithm is proposed.In this method,RNN model is used to extract the time series features of audio signals,and K-means clustering algorithm is used to cluster the features,so as to enhance the robustness and comprehensiveness of audio classification.The UrbanSound8K dataset evaluation method was used.The results show that the method is superior to the standard RNN model in accuracy,recall rate and F_(1) value.

关 键 词：循环神经网络(RNN) K-MEANS聚类算法 音频分类 机器学习 

分 类 号：TN912.34[电子电信—通信与信息系统]