Papers
5
Total Citations
78
H-Index
5
About
Shigeki Matsuda is a leading researcher in human-robot interaction and robust automatic speech recognition (ASR), whose work has been instrumental in enabling communication robots to operate effectively in noisy, real-world environments. His major contributions center on developing ASR systems that maintain high recognition accuracy despite background noise and across speakers of different ages, including both adults and children—a critical challenge for practical robotics. His most influential work, "A Robust Speech Recognition System for Communication Robots in Noisy Environments" (2008, 34 citations), and its predecessor (2006, 21 citations), established foundational techniques for noise-robust ASR in interactive robots. Matsuda further advanced the field by pioneering methods for detecting robot-directed speech, introducing the innovative Multimodal Semantic Confidence (MSC) measure. This approach, detailed in papers from 2010 (totaling over 20 citations), allows robots to distinguish speech intended for them from speech directed at others by integrating audio, visual, and motion cues during physical tasks. His work bridges speech processing and situated understanding, making human-robot interaction more natural and intuitive. Matsuda’s research remains highly influential for engineers and researchers developing socially aware robots capable of seamless communication in dynamic, unstructured settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5