Environmental Acoustic Intelligence: Intelligent Sound Perception to Context-aware Active Noise Control
IEEE Queensland Section Signal Proc & Comms Chapter is pleased to invite you to a Distinguished Lecturer talk (hybrid) by Prof. Woon-Seng Gan from Nanyang Technological University (NTU), Singapore.
Title: Environmental Acoustic Intelligence: Intelligent Sound Perception to Context-aware Active Noise Control
Date and time: Aug 14, 4- 5 pm
Venue: 49-601, the University of Queensland
In this presentation, Professor Woon-Seng Gan will share insights into the links between advanced acoustic perception and Active Noise Control (ANC), highlighting recent developments, key challenges, and future opportunities in the field.
This event offers an excellent opportunity for academics, researchers, students, and industry professionals to engage with a leading expert and participate in stimulating discussions.
Light refreshments will be provided. We look forward to your participation.
Abstract:
Environmental Acoustic Intelligence (EAI) is transforming conventional noise monitoring into context-aware systems capable of intelligently perceiving, interpreting, and responding to complex acoustic environments. This talk focuses on the critical link between advanced acoustic perception, including sound event detection, localization, and scene classification and real-world interventions like adaptive Active Noise Control (ANC).
Crucially, the presentation examines the paradigm shift from traditional adaptive filtering techniques, such as the Filtered-x Least Mean Square (FxLMS) algorithm, toward advanced AI-driven approaches for ANC. We explore the deployment of Convolutional Neural Networks (CNNs), leveraging advanced representations like 1D CNN encoders and latent space projections to achieve highly responsive Selective Fixed-filter ANC (SFANC). Furthermore, the integration of Reinforcement Learning (RL) is discussed as a mechanism for developing autonomous acoustic agents capable of continuously fine-tuning their control strategies in dynamic, unpredictable environments.
By addressing practical constraints such as latency, acoustic-path variations, and edge-computing limitations, these AI-driven methodologies empower closed-loop EAI systems. Ultimately, this progression moves beyond passive monitoring to implement intelligent, context-aware noise mitigation that adapts seamlessly to both complex physical environments and human perceptual needs.
Biography:
Woon-Seng Gan is a Professor of Audio Engineering and Director of the Smart Nation TRANS Lab at Nanyang Technological University (NTU), Singapore, where he leads efforts to translate AI research into real-world deployments. His research bridges the physical world and signal processing, with a specific focus on active noise control (ANC) and spatial audio, resulting in over 400 publications, eight patents, and three co-authored books. A Fellow of the AES, and IET, he was named an IEEE Signal Processing Society Distinguished Lecturer in 2023 and received the prestigious 2017 APSIPA Sadaoki Furui Prize Paper Award for his work on ANC. In addition to serving as the President of the Asia Pacific Signal and Information Processing Association (APSIPA) and Technical Program Chair for International Conference on Acoustic, Speech and Signal Processing (ICASSP) 2022, Prof. Gan actively advises government agencies on environmental noise sensing and mitigation, and has had his work featured in global media outlets such as the New York Times, CBS, and CNN.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
Loading virtual attendance info...
- the University of Queensland
- Brisbane, Queensland
- Australia
- Building: 49
- Room Number: 601
- Starts 06 August 2026 02:00 PM UTC
- Ends 14 August 2026 06:00 AM UTC
- No Admission Charge