STREAM ATTENTION BASED U-NET FOR L3DAS23 CHALLENGE
Honglong Wang ( Tianjin University); Yanjie Fu (Tianjin University); Junjie Li (Tianjin University); Meng Ge (Tianjin University); Longbiao Wang (Tianjin University); xinyuan qian (National University of Singapore)
-
Members: FreeSPS
IEEE Members: $11.00
Non-members: $15.00
10 Jun 2023
Machine learning applications of 3D audio are gaining increasing interest in recent years. In this paper, we proposea stream attention based U-Net to remove background noise and reverberation based on ICASSP Signal Processing Grand
Challenge 2023: L3DAS23 Challenge1 Audio-only track task1 3D Speech Enhancement. Results show that proposed method achieves superior performance than the official baseline model.