Deep Noise Suppression Challenge – ICASSP 2021

Region: Global

The accepted papers for Track 1 are below. Paper acceptance was done through the normal ICASSP peer-review process.

Place

Performance Rank

Team

Authors

Title

1

1

Chinese Academy of Sciences

Andong Li, Wenzhe Liu, Xiaoxue Luo, Chengshi Zheng, Xiaodong Li

ICASSP 2021 deep noise suppression challenge: Decoupling magnitude and phase optimization with a two-stage deep network

2

2

Sogou

Jingdong Li, Dawei Luo, Yun Liu, Yuanyuan Zhu, Zhaoxia Li, Guohui Cui, Wenqi Tang, Wei Chen

Densely Connected Multi-Stage Model with Channel Wise Subband Feature for Real-Time Speech Enhancement

3

5

Seoul National University, Supertone

Hyeong-Seok Choi , Sungjin Park , Jie Hwan Lee, Hoon Heo, Dongsuk Jeon, Kyogu Lee

Real-time denoising and dereverberation with tiny recurrent u-net

4

11

Westlake University, Inner Mongolia University, Inria Grenoble Rhone-Alpes

Xiang Hao , Xiangdong Su , Radu Horaud, and Xiaofei Li

FullSubNet: a full-band and sub-band fusion model for real-time single-channel speech enhancement

Challenge_Results