Speech Signal Processing on Graphs: Graph Topology, Graph Frequency Analysis and Denoising  

作　　者：WANG Tingting GUO Haiyan LYU Bin YANG Zhen 

机构地区：[1]Broadband Wireless Communication and Sensor Network Technology Key Lab,Nanjing University of Posts and Telecommunications,Nanjing 210003,China

出　　处：《Chinese Journal of Electronics》2020年第5期926-936,共11页电子学报（英文版）

基　　金：supported by the National Natural Science Foundations of China(No.61671252,No.61271335,No.61901229);the Natural Science Research of Higher Education Institutions of Jiangsu Province(No.19KJB510008);a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

摘　　要：The paper investiga tes the hidden relationships among speech samples by applying graph tools.Specifically,we first estimate an applicable graph topology for unstructured speech signals,which can map speech signals into the ver tex domain successfully and cons true t as Speech graph signals(SGSs).On the basis,we define a new graph Fourier transfbrm for SGSs,which can investigate its related graph Fourier analysis.Moreover,we propose a new Graph st ruc ture spec tral sub traction(GSSS)method for speech enhancement under different noisy environments.Simulation results show that the performance of the GSSS method can be significantly improved than the classical Basic spectral subtraction(BSS)method in terms of the average Segmentai signal-tonoise ratio(SSNR),Perceptual evaluation of speech quality(PESQ)and the computational complexity.

关 键 词：Graph signal processing SPEECH Graph Fourier transfbrm Graph topology Denoising. 

分 类 号：TN912.3[电子电信—通信与信息系统]