Kwabena Sarpong

University of Electronic Science and Technology of China

Papers

Total Citations

H-Index

About

Kwabena Sarpong is making impactful strides at the intersection of affective computing and assistive robotics, with a primary focus on multimodal speech emotion recognition. His most-cited work, "Multimodal Speech Emotion Recognition Using Modality-Specific Self-Supervised Frameworks" (2023, 10 citations), addresses a critical challenge in human-robot interaction: enabling machines to reliably interpret human emotional states. Sarpong’s key contribution lies in developing self-supervised learning frameworks that leverage modality-specific features—such as vocal tone, facial expressions, and physiological signals—to improve the accuracy and robustness of emotion recognition systems. This work is foundational for creating socially aware robots that can respond appropriately to human affect, a prerequisite for seamless collaboration in healthcare, education, and domestic settings. By tackling the inherent variability and noise in real-world emotional data, Sarpong’s research advances the reliability of these systems, moving them closer to practical deployment. His approach not only enhances robot empathy but also opens new avenues for human-centered AI design. As a rising voice in the field, Sarpong’s contributions are shaping the next generation of emotionally intelligent machines, promising safer and more intuitive human-robot partnerships.