I am HAO Xiang (Shawn) CV, a Ph. D. Candidate at the Department of Data Science and Artificial Intelligence at The Hong Kong Polytechnic University, supervised by Prof. WU Jibin and Prof. TAN Kay Chen.
My research focuses on developing next-generation signal processing strategies to enhance communication for individuals with hearing loss in challenging listening environments. By integrating principles from auditory perception with digital signal processing and deep learning, I build solutions that are both effective and practical.
A key contribution of my Ph. D. work is the development of a real-time hearing aid research platform, which has been successfully deployed on a commercial chip. This platform features a low-power, low-latency (< 5 mW, < 6 ms) DNN-based speech enhancement model, coupled with personalized hearing loss compensation using standard prescriptions (e.g., NAL-NL2) and Wide Dynamic Range Compression (WDRC). This end-to-end system, supported by our professional audio testing equipment, provides a powerful tool for the realistic evaluation and rapid prototyping of advanced hearing devices.