残差网络在婴幼儿哭声识别中的应用  

Application of Residual Network to Infant Crying Recognition

作　　者：谢湘[1] 张立强 王晶[1] XIE Xiang;ZHANG Liqiang;WANG Jing(School of Information and Electronics,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]北京理工大学信息与电子学院,北京100081

出　　处：《电子与信息学报》2019年第1期233-239,共7页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(61473041;11590772;61571044)~~

摘　　要：该文使用语谱图结合残差网络的深度学习模型进行婴幼儿哭声的识别,使用婴幼儿哭声与非哭声样本比例均衡的语料库,经过五折交叉验证,与支持向量机(SVM),卷积神经网络(CNN),基于Gammatone滤波器的听觉谱残差网络(GT-Resnet)3种模型相比,基于语谱图的残差网络取得了最优结果,F1-score达到0.9965,满足实时性要求,证明了语谱图在婴幼儿哭声识别任务中能直观地反映声学特征,基于语谱图的残差网络是解决婴幼儿哭声识别任务的优秀方法。The deep learning model based on the residual network and the spectrogram is used to recognize infant crying.The corpus has balanced proportion of infant crying and non-crying samples.Finally,through the 5-fold cross validation,compared with three models of Support Vector Machine(SVM),Convolutional Neural Network(CNN)and the cochleagram residual network based on Gammatone filters(GT-Resnet),the spectrogram based residual network gets the best F1-score of 0.9965 and satisfies requirements of real time.It is proved that the spectrogram can react acoustics features intuitively and comprehensively in the recognition of infant crying.The residual network based on spectrogram is a good solution to infant crying recognition problem.

关 键 词：婴儿哭声识别 深度学习 残差网络 语谱图 

分 类 号：TP391.42[自动化与计算机技术—计算机应用技术]