Papers
15
Total Citations
155
H-Index
7
About
Takuma Otsuka is a leading researcher in robot audition and interactive music robotics, whose work bridges the gap between artificial listening systems and natural human-robot collaboration. His primary research areas include sound source localization, real-time audio processing, and musical synchronization for robots. Otsuka’s most influential contribution is the design and implementation of selectable sound separation on the Texai telepresence system using the HARK robot audition software (38 citations), which enables remote operators to selectively focus on specific speakers in a crowded environment. He also pioneered real-time audio-to-score alignment using particle filters for coplayer music robots (32 citations), allowing robots to synchronize with human performers during live musical ensembles. His work on incremental polyphonic audio-to-score alignment for singer robots (13 citations) and audio-visual beat-tracking for robot-human guitar duets (11 citations) further demonstrates his commitment to seamless musical interaction. Otsuka’s Bayesian extension of MUSIC for sound source localization (10 citations) advanced the field of auditory scene analysis, and his unified auditory functions based on Bayesian topic models (7 citations) proposed a more robust, integrated approach to robot hearing. Through these innovations, Otsuka has established himself as a key figure in making robots not just hear, but actively participate in complex auditory environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5Bayesian extension of MUSIC for sound source localization and tracking10 citations · 2011
- 6
- 7Unified auditory functions based on Bayesian topic model7 citations · 2012
- 8
- 9Posture estimation of hose-shaped robot by using active microphone array6 citations · 2015
- 10