Abstract: The neural networks using a combination of multiple Gated Recurrent Units (GRUs) layers in the Neural Network enhance ASC's accuracy. Due to the advantage of time sequence modelling, the GRUs can perform better class-based discrimination and mapping features. The results on the TUT Acoustic Scenes evaluation dataset demonstrates that the proposed model performs better in ASC than Long short-term memory layer. The average class-wise accuracy has shown improvement with the proposed method.

Biography: Dr. Cheng Siong Chin has a Ph.D. from Nanyang Technological University, Singapore. He is currently a Reader (Associate Professor) with Newcastle University and Adjunct Full Professor at Chongqing University, School of Automotive Engineering. He has published over 100 publications, 5 authored books, and 3 US Patents. His research interests include the design and simulation of complex systems for an uncertain environment. He is a Fellow of IMarEST, Senior Member of IEEE and the IET, and a Chartered Engineer. In 2018, he was invited as Plenary Speaker and Chief Guest for the 2018 IEEE International Conference on Power, Energy control, and Transmission Systems. He was the Panel Member for Senior Member of IEEE Application in Region 10 in 2014 and 2017. He received DCASE2019 Judges' Award (most innovative and original) for Sound Event Detection in Domestic Environments in the IEEE AASP Challenge on DCASE2019.