An Adaptive Method Based on Multiscale Dilated Convolutional Network for Binaural Speech Source Localization. (30th December 2020)

• Record Type: Journal Article
• Authors: Wu, Lulu; Liu, Hong; Yang, Bing; Ding, Runwei
• Other Names: Yang Zhile Academic Editor.
• Is Part Of: Complexity. Volume 2020(2020)
• Publication Date: 2020-12-30
• Subjects: Chaotic behavior in systems -- Periodicals; Complexity (Philosophy) -- Periodicals; 003
• DOI: 10.1155/2020/5819624
• Languages: English
• ISSNs: 1076-2787

Abstract

Abstract : Most binaural speech source localization models perform poorly in unprecedentedly noisy and reverberant situations. Here, this issue is approached by modelling a multiscale dilated convolutional neural network (CNN). The time-related crosscorrelation function (CCF) and energy-related interaural level differences (ILD) are preprocessed in separate branches of dilated convolutional network. The multiscale dilated CNN can encode discriminative representations for CCF and ILD, respectively. After encoding, the individual interaural representations are fused to map source direction. Furthermore, in order to improve the parameter adaptation, a novel semiadaptive entropy is proposed to train the network under directional constraints. Experimental results show the proposed method can adaptively locate speech sources in simulated noisy and reverberant environments.