History of Robotics Research and Development of Japan2008Integration, Intelligence, etc.Development and Deployment of Open Source Robot Audition Software HARK (Honda Research Institute Japan Audition for Robots with Kyoto University)

Kazuhiro NakadaiHonda Research Institute Japan Co., Ltd.
Shun'ichi YamamotoKyoto University
Hiroshi G. OkunoKyoto University
Jean-Marc ValinUniversité de Sherbrooke
Hirofumi NakajimaHonda Research Institute Japan Co., Ltd.
Ryu TakedaKyoto University
Keisuke NakamuraHonda Research Institute Japan Co., Ltd.
Takami YoshidaTokyo Institute of Techinology
Takeshi MizumotoKyoto University
Takuma OtsukaKyoto University
HARK (Honda Research Institute Japan Audition for Robots with Kyoto University) was released as open source software for robot audition in April, 2008, which is a grand sum of achievements in robot audition since 2000. Uinsg a microphone array consisting of multiple microphones, three main functions such as sound source localization, sound source separation, and automatic speech recognition of separated speech can be performed with HARK. From the viewpoint of software engineering, HARK was designed with two concepts; user-friendliness that people who do not have enough knowledge on signal and speech processing can operate, and real-time processing that a system can be performed in an offline manner as well as in an online and real-time manner. For the former, GUI programming, easy installation, rich documentation including over 300 page manual and recipes, sufficient number of modules supporting from audio input to automatic speech recognition, supporting multiple platforms like Linux and Windows. For the latter, supporting both off-the-shelf and professional multi-channel audio A/D devices, advanced real-time signal processing modules, seamless connection with ROS. Its effectiveness is shown by various demos, for example, a meal order taking robot that correctly understands eleven simultaneous orders, which exceeds the legendaly capability of Prince Shotoku. Since 2008, The version-up of HARK has been annually cotinued with domestic and/or international free tutorials (as of June 2015, 8 vesion-ups and 11 tutorials have been made.). Email-based technical support is also available. Over 50,000 HARK packages have been downloaded. Recently, HARK users are increasing and HARK is applied to various research fields such as ethology and acoustics in extreme envrionments. 19th RSJ Technical Innovations Award in 2014. IEEE/RSJ IROS 2001 Best Paper Award Finalist in 2007. IEA/AIE The Best Paper Award in 2010.


Correspondence papers

Kazuhiro Nakadai, Hiroshi G. Okuno, Hirofumi Nakajima, Yuji Hasegawa, Hiroshi Tsujino:An Open Source Software System For Robot Audition HARK and Its Evaluation

IEEE-RAS Interanational Conference on Humanoid Robots (Humanoids 2008), pp.561-566, 2008

Hiroshi G. Okuno, Kazuhiro Nakadai:Robot Audition Open-Sourced Software HARK

Journal of the Robotics Society of Japan, Vol.28, No.1, pp.6-9, 2010 (in Japanese)

Kazuhiro Nakadai, Toru Takahashi, Hiroshi G. Okuno, Hirofumi Nakajima, Yuji Hasegawa, Hiroshi Tsujino:Design and Implementation of Robot Audition System "HARK" - Open Source Software for Listening to Three Simulteaneous Speakers

Advanced Robotics, Vol.24, Issue 5-6 (Jan. 2010) pp. 739-761

Kazuhiro Nakadai, Toru Takahashi, Hiroshi G. Okuno, Nakamura Keisuke, Yoshida Takami, Mizumoto Takeshi Otsuka Takuma Ince Gohkan:Introduction to Open Source Robot Audition Software HARK

The 29th annual conference of the robotics society of Japan, 2Q2-6, 2011

Related papers

[1] HARK web page:
[2] Osamu Sugiyama, Katsutoshi Itoyama, Kazuhiro Nakadai, Hiroshi G. Okuno, Sound Annotation Tool for Multidirectional Sounds based on spatial information extracted by HARK robot audition software, Proceeding of IEEE International Conference on Systems, Man, and Cybernetics (SMC 2014), 2335-2340, San Diego, 5-8 Oct. 2014.
[3] 中臺 一博, 奥乃 博: ロボット聴覚オープンソフトウェアHARKの紹介, 第15回計測自動制御学会システムインテグレーション部会講演会, キーノート講演, 2L3-1, pp.1712-1716, Dec. 15, 2014
[4] 西牟田 勇哉, 平山 直樹, 大塚 琢馬, 杉山 治, 糸山 克寿, 奥乃 博: ロボット聴覚ソフトウエアHARKを用いたクイズの同時回答を識別するロボット司会者 の設計と実装, 第37回 AI チャレンジ研究会, SIG-Challenge-B3020-09, pp.45-50, 人工知能学会, 早稲田大学, Dec. 6. 2013
[5] Takeshi Mizumoto, Takami Yoshida, Kazuhiro Nakadai, Ryu Takeda, Takuma Otsuka, Toru Takahashi, Hiroshi G. Okuno: Design and Implementation of Selectable Sound Separation on a Texai Telepresence System using HARK, Proceedings of IEEE-RAS International Conference on Robotics and Automation (ICRA-2011), pp.2130-2137, May 9-13 (10), 2011, Shanghai, China. pdf doi:10.1109/ICRA.2011.5979849

Related Article