From Auditory Cognitive Neuroscience to Open-Source EEG/Audio Wearables and Unified Algorithmic Frameworks, a Cognitive Science Speaker Series Presentation
Speaker: Hwan Shim, Ph.D.
Title: From Auditory Cognitive Neuroscience to Open-Source EEG/Audio Wearables and Unified Algorithmic Frameworks
Short Bio: Hwan Shim, Ph.D., is an Assistant Professor of Electrical and Computer Engineering Technology at the Rochester Institute of Technology (RIT) and Principal Investigator of the Music and Audio Cognition Lab (MACL, rit.edu/macl). He earned his Ph.D. in Electrical Engineering and Computer Science from Seoul National University (2008). Dr. Shim’s background spans academia and industry, with prior appointments at Samsung Electronics (Principal Engineer), Stanford University, the University of Iowa, and Dong-Ah Institute of Media and Arts.
Abstract: Understanding the neural mechanisms of selective attention to a target speaker in complex acoustic environments provides a basis for the development of neuro-steered hearing devices. Auditory cognitive neuroscience studies have demonstrated that selective attention modulates cortical responses to target and competing sounds, motivating auditory attention decoding (AAD) approaches that infer a listener’s attentional focus from EEG.
This talk reviews recent progress toward translating these findings from controlled laboratory experiments into practical, real-world systems. I will first introduce neural mechanisms of auditory selective attention and the evolution of AAD from linear stimulus–response models to deep-learning and semantic decoding approaches. I will then present our ongoing development of an open-source EEG/audio wearable integrating dry EEG electrodes, microphone arrays, embedded processing, and audio output, together with efforts to reduce the number of EEG channels required for reliable decoding. Finally, I will discuss unified preprocessing and algorithmic frameworks for combining heterogeneous EEG datasets and improving cross-dataset generalization. Together, these efforts illustrate a path toward multimodal, wearable, and adaptive auditory attention decoding systems capable of operating beyond the laboratory and ultimately supporting real-world neuro-steered hearing technologies.
ASL-English interpreters have been requested. Light refreshments will be provided.
Event Snapshot
When and Where
Who
This is an RIT Only Event
Interpreter Requested?
Yes