Hirofumi Nakajima

About

Hirofumi Nakajima is a pioneering researcher in the field of robot audition, dedicated to enabling robots to perceive, localize, and interpret human speech in complex, real-world acoustic environments. His most celebrated contribution is the development of HARK (Honda Research Institute Japan Audition for Robots with Kyoto University), an open-source robot audition software system capable of simultaneously processing speech from multiple speakers — a landmark achievement that has garnered over 200 citations since its 2010 publication. HARK integrates sound source localization, blind source separation (BSS), and automatic speech recognition into a flexible, hardware-agnostic framework applicable to diverse robotic platforms. Nakajima's work has substantially advanced the theoretical underpinnings of BSS, particularly through his development of adaptive step-size parameter control methods that make separation algorithms robust and practical in unpredictable environments. He has further extended robot audition into dynamic scenarios involving moving speakers and robots, ego-noise estimation, and multi-microphone array integration for real-time tracking. His rock-paper-scissors referee robot memorably demonstrated these capabilities in an engaging real-world context. Collectively, his body of work — accumulating over 560 citations — has meaningfully shaped the trajectory of human-robot interaction research.

Research Focus

Key Achievements

Top Papers

1. 1
   Design and Implementation of Robot Audition System 'HARK' — Open Source Software for Listening to Three Simultaneous Speakers

   203 citations · 2010
2. 2
   An open source software system for robot audition HARK and its evaluation

   70 citations · 2008
3. 3
   Blind Source Separation With Parameter-Free Adaptive Step-Size Method for Robot Audition

   70 citations · 2009
4. 4
   Adaptive step-size parameter control for real-world blind source separation

   46 citations · 2008
5. 5
   Real-Time Tracking of Multiple Sound Sources by Integration of In-Room and Robot-Embedded Microphone Arrays

   39 citations · 2006
6. 6
   Robot audition for dynamic environments

   30 citations · 2012
7. 7
   A robot referee for rock-paper-scissors sound games

   29 citations · 2008
8. 8
   Sound source separation of moving speakers for robot audition

   28 citations · 2009
9. 9
   Assessment of general applicability of ego noise estimation

   24 citations · 2011
10. 10
    An easily-configurable robot audition system using Histogram-based Recursive Level Estimation

    22 citations · 2010

Key Collaborators

Contact & Links