Tsuyoki Nishikawa
Papers
2
Total Citations
12
H-Index
2
About
Tsuyoki Nishikawa is a leading researcher in robot audition and noise-robust speech processing, with foundational contributions that have shaped how machines hear in real-world environments. His work addresses the critical challenge of enabling human-robot interaction through hands-free speech recognition, even in noisy, dynamic settings. Nishikawa pioneered the **spatial subtraction array (SSA)** and **known noise superimposition** technique, a method that dramatically improves speech recognition accuracy by subtracting estimated noise power spectra from target speech signals. This innovation, detailed in his highly cited 2005 paper (8 citations), provides a robust framework for robots to understand human commands amidst background chatter and machinery. He further advanced the field with **blind sound scene decomposition**, introducing a novel **SIMO-model-based ICA** algorithm that allows humanoid robots to separate mixed binaural signals into distinct sound sources without prior knowledge. This work, also from 2005 (4 citations), is a cornerstone for auditory scene analysis in robotics. Nishikawa’s research is not only technically rigorous but also deeply practical, directly enabling more intuitive and resilient human-robot communication. His achievements underscore a career dedicated to bridging the gap between acoustic theory and real-world robotic perception.
Research Focus
Key Achievements
Top Papers
- 1
- 2