An Adaptive Method Based on Multiscale Dilated Convolutional Network for Binaural Speech Source Localization
Joint Authors
Yang, Bing
Ding, Runwei
Wu, Lulu
Liu, Hong
Source
Issue
Vol. 2020, Issue 2020 (31 Dec. 2020), pp.1-7, 7 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2020-12-30
Country of Publication
Egypt
No. of Pages
7
Main Subjects
Abstract EN
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.
American Psychological Association (APA)
Wu, Lulu& Liu, Hong& Yang, Bing& Ding, Runwei. 2020. An Adaptive Method Based on Multiscale Dilated Convolutional Network for Binaural Speech Source Localization. Complexity،Vol. 2020, no. 2020, pp.1-7.
https://search.emarefa.net/detail/BIM-1142565
Modern Language Association (MLA)
Wu, Lulu…[et al.]. An Adaptive Method Based on Multiscale Dilated Convolutional Network for Binaural Speech Source Localization. Complexity No. 2020 (2020), pp.1-7.
https://search.emarefa.net/detail/BIM-1142565
American Medical Association (AMA)
Wu, Lulu& Liu, Hong& Yang, Bing& Ding, Runwei. An Adaptive Method Based on Multiscale Dilated Convolutional Network for Binaural Speech Source Localization. Complexity. 2020. Vol. 2020, no. 2020, pp.1-7.
https://search.emarefa.net/detail/BIM-1142565
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-1142565