Papers
4
Total Citations
138
H-Index
3
About
S. Yamamoto is a pioneering researcher in robot audition, whose work has fundamentally advanced how humanoid robots perceive and understand speech in noisy, real-world environments. His key research areas include sound source localization, separation, and automatic speech recognition for robots—collectively known as robot audition. Yamamoto’s major contribution is the development of a system that integrates the active direction-pass filter (ADPF) with Missing Feature Theory (MFT), enabling a humanoid robot to recognize three simultaneous speech streams in real time. This breakthrough, demonstrated with the robot SIG, was the first of its kind and is documented in his most-cited paper (69 citations). His 2004 and 2005 follow-up works (each with 33 citations) further solidified this achievement, showing that robots could not only separate but also understand overlapping voices—a critical step toward natural human-robot interaction. Yamamoto’s research has been instrumental in moving robot audition from isolated laboratory conditions to practical, multi-source environments, making him a key figure in the field. His work continues to inspire advances in assistive robotics, teleoperation, and intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Making a robot recognize three simultaneous sentences in real-time33 citations · 2005
- 4