PhD students Mr HAO Xiang and Mr MA Chenxiang from the Department of Computing at PolyU won the Best Paper Award at the 2024 IEEE Conference on Artificial Intelligence (IEEE CAI 2024) for their collaborative paper titled “When Audio Denoising Meets Spiking Neural Network”. The research was conducted in collaboration with Prof. TAN Kay Chen, Interim Head and Chair Professor of Computational Intelligence of the Department of Data Science and Artificial Intelligence of PolyU, Dr WU Jibin, Assistant Professor of the Department of Data Science and Artificial Intelligence and the Department of Computing, and Miss YANG Qu from the National University of Singapore.
 
The award-winning paper presents an innovative solution for audio denoising by integrating a novel gated spiking neural model (GSN) with an optimised FullSubNet based on spiking neural networks (SNNs). This approach enables efficient processing of both full-band and sub-band frequencies while significantly reducing computational overhead. Additionally, the paper introduces a metric discriminator-based loss function that selectively enhances desired performance metrics without compromising others. Empirical evaluations demonstrate the superior performance of Spiking-FullSubNet in computational efficiency and denoising performance, which was also recognised as the winner of Track 1 (Algorithmic) at the 2023 Intel Neuromorphic Deep Noise Suppression Challenge.

IEEE CAI 2024, held in Singapore from 25 to 27 June, is a prominent industry-organised event dedicated to artificial intelligence and its applications in various industries. The conference serves as a platform for connecting AI enterprise leaders, innovators, and stakeholders from industry, government, and start-ups and showcases the latest advancements in AI research and applications.
 

Hao, X., Ma, C., Yang, Q., Tan, K. C., & Wu, J. (2024). When Audio Denoising Meets Spiking Neural Network. 2024 IEEE Conference on Artificial Intelligence (CAI), 1524–1527. https://doi.org/10.1109/CAI59869.2024.00275