An End-to-End Binaural Sound Localization Model Based on the Equalization and Cancellation Theory

Tao Song, Wenwen Zhang, Jing Chen*

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

The end-to-end framework has been introduced into the binaural localization modeling and achieved higher localization accuracy than the other models, however, the reasonability and interpretability for applying the related neural networks remain unclear. It has been well documented that the auditory system relies on binaural cues for sound localization, and the equalization and cancellation (EC) theory describes how the binaural cues are extracted. In this paper, an end-to-end binaural localization model is proposed based on the EC theory. In the proposed model, a convolution neural network(CNN) with a specifically designed activation function is used to implement the EC theory. The proposed model was trained in synthesized rooms and evaluated in real rooms. Experiment results show that CNN kernels learned by the proposed model are corresponding to binaural cues, and the proposed model outperforms the current end-to-end model by a 10.73% improvement in localization accuracy and a 12.91% improvement in root mean square error(RMSE).

源语言英语
主期刊名AES Europe Spring 2022 - 152nd Audio Engineering Society Convention 2022
出版商Audio Engineering Society
275-283
页数9
ISBN(电子版)9781713855415
出版状态已出版 - 2022
已对外发布
活动AES Europe Spring 2022 - 152nd Audio Engineering Society Convention 2022 - The Hague, Virtual, 荷兰
期限: 16 5月 202219 5月 2022

出版系列

姓名AES Europe Spring 2022 - 152nd Audio Engineering Society Convention 2022

会议

会议AES Europe Spring 2022 - 152nd Audio Engineering Society Convention 2022
国家/地区荷兰
The Hague, Virtual
时期16/05/2219/05/22

指纹

探究 'An End-to-End Binaural Sound Localization Model Based on the Equalization and Cancellation Theory' 的科研主题。它们共同构成独一无二的指纹。

引用此