Tomas Dekens

Papers

Total Citations

H-Index

About

Tomas Dekens is a robotics researcher specializing in human-robot interaction, auditory perception, and conversational systems for humanoid platforms. His work focuses on enabling robots to better perceive and respond to human speech in real-world, noisy environments. Dekens made significant early contributions to the field with his 2011 paper on an event-based conversational system for the Nao robot, which has garnered 16 citations and laid groundwork for more natural human-robot dialogue. His 2012 study on the effect of speech denoising algorithms on sound source localization for humanoid robots, cited 6 times, critically evaluated four major denoising techniques as weighting functions for Generalized Cross-Correlation (GCC), while also examining how microphone frequency response impacts localization accuracy. This work directly addressed the challenge of robots understanding speech in noisy settings—a key barrier to practical deployment. Dekens’ research has helped advance auditory processing for social robots, with implications for assistive technology and collaborative robotics. His findings continue to inform the development of more robust, context-aware robotic systems capable of engaging with humans in dynamic acoustic environments.