Sound event localization and detection based on deep learning

作     者：ZHAO Dada DING Kai QI Xiaogang CHEN Yu FENG Hailin ZHAO Dada;DING Kai;QI Xiaogang;CHEN Yu;FENG Hailin

作者机构：School of Mathematics and StatisticsXidian UniversityXi’an 710071China Science and Technology on Near-Surface Detection LaboratoryWuxi 214035China 

出 版 物：《Journal of Systems Engineering and Electronics》 (系统工程与电子技术（英文版）)

年 卷 期：2024年第35卷第2期

页      面：294-301页

核心收录：

学科分类：0711[理学-系统科学] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 0808[工学-电气工程] 07[理学] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the National Natural Science Foundation of China(61877067) the Foundation of Science and Technology on Near-Surface Detection Laboratory(TCGZ2019A002,TCGZ2021C003,6142414200511) the Natural Science Basic Research Program of Shaanxi(2021JZ-19) 

主　　题：sound event localization and detection(SELD) deep learning convolutional recursive neural network(CRNN) channel attention mechanism 

摘      要：Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research *** recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research *** paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional ***-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input *** features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results *** channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless ***,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD *** experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.