DC-TseNet: A dual-channel time-domain speech enhancement network

Yihui Fu, Sining Sun, Lei Xie

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

In this paper, we propose an end-to-end dual-channel time domain speech enhancement approach, named DC-TseNet, for devices with multiple microphones such as mobile phones used in far-filed scenario like teleconferencing. DC-TseNet incorporates a computationally efficient CNN to form a unified encoder-enhancement-decoder structure that learns clean speech directly using multichannel signals. In addition, DC-TseNet is trained from both intra-channel an inter-channel features to express the relevance and difference between the collected signals from the two microphones, which makes sufficient use of spatial information and reduce the influence of recording direction on the signals. The experimental results show that the proposed dual-channel time-domain approach, with more compact model size, significantly outperforms the LSTM-based frequency-domain method. Furthermore, we find that the inter-channel information, especially the difference between two channels, is more important for a better performance gain.

源语言英语
主期刊名2020 8th International Conference on Orange Technology, ICOT 2020
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9781665418522
DOI
出版状态已出版 - 18 12月 2020
活动8th International Conference on Orange Technology, ICOT 2020 - Daegu, 韩国
期限: 18 12月 202021 12月 2020

出版系列

姓名2020 8th International Conference on Orange Technology, ICOT 2020

会议

会议8th International Conference on Orange Technology, ICOT 2020
国家/地区韩国
Daegu
时期18/12/2021/12/20

指纹

探究 'DC-TseNet: A dual-channel time-domain speech enhancement network' 的科研主题。它们共同构成独一无二的指纹。

引用此