Papers
26
Total Citations
331
H-Index
11
About
Kazuyoshi Yoshii is a leading researcher in robot audition and human-robot interaction, specializing in enabling robots to perceive and interact with sound in real-world environments. His pioneering work focuses on musical beat tracking, sound source localization, and speech enhancement for mobile robots. Yoshii’s most influential contribution is developing biped robots that can autonomously synchronize their steps to musical beats while listening through their own microphones—a breakthrough that bridges robotics and music cognition. His 2007 paper on beat-synchronized biped locomotion has garnered 43 citations, while his 2016 work on sound-based localization for in-pipe snake robots (41 citations) demonstrates the practical application of audio processing in GPS-denied environments. Yoshii has also advanced multichannel speech enhancement through Bayesian low-rank and sparse decomposition (25 citations) and developed nested infinite Gaussian mixture models for acoustic event recognition (20 citations). His notable achievements include creating robots that can sing along to music, count beats aloud, and perform audio-visual beat tracking for dancing with humans. Through these innovations, Yoshii has established himself as a key figure in making robots more perceptive and interactive through sound.
Research Focus
Key Achievements
Top Papers
- 1
- 2Sound-based online localization for an in-pipe snake robot41 citations · 2016
- 3
- 4
- 5
- 6
- 7
- 8
- 9A Robot Singer With Music Recognition Based On Real-Time Beat Tracking.13 citations · 2008
- 10