Welcome to Multimedia Information Processing Laboratory at the University of Yamanashi.
We are challenging the Acoustics, Linguistics, Picture and Signal (ALPS) Processing
Our laboratory welcomes graduate students from other universities (laboratories) both within Japan and abroad. If you are interested in deep learning or audio and multimedia information processing, please feel free to contact us. The application guidelines for the graduate school entrance examination for the academic year 2024 (Reiwa 6) are scheduled to be released around April. For more information, please refer to the university’s Entrance Examination Information and Application Guidelines. The entrance examination for the Master’s program in the academic year 2024 will be held in early July (first term) and early December (second term). In some cases, the recommendation-based entrance examination may be applicable to applicants from other universities (departments). Please contact us for more details. Entrance examination information can be found here.
Speech research at the University of Yamanashi dates back to 1955 when Dr. Minoru Shigenaga was appointed and began research on speech synthesis. The year 2024 marks the 69th anniversary of this research, which was continued by Dr. Yoshihiro Sekiguchi. Since 2013, Dr. Nishizaki has been leading the laboratory and Dr. Leow joined as Assistant Professor at 2024.
In the Nishizaki-Leow Laboratory, we primarily study intelligent information processing of four multimedia domains: Acoustic, Linguistics, Picture, and Signal (ALPS), using state-of-the-art deep learning technology. Our research encompasses basic research on acoustic and linguistic analysis and processing of human speech, research and development of applications based on these technologies, recognition of environmental sounds, and image recognition, such as text detection/recognition and handwritten character image generation. We also conduct applied research in the fields of agriculture and medicine.
Through our research and development efforts, we are committed to creating and utilizing technologies that benefit the world and to nurturing human resources who will lead the future through research.
山梨大学における音声研究は古く,昭和30年に重永実先生が赴任されて音声合成の研究を始められたことを発端とします。その後、関口芳廣先生が研究を引き継がれ、2013年からは西崎が研究室を主宰するようになりました。2024年で69年目となりました。西崎-レオ研究室では,最新の深層学習(ディープラーニング)技術を利用して,主として,音響・言語・画像・信号の4つのマルチメディア,ALPS(Acoustic, Linguistics, Picture, and Signal) の知的情報処理の研究を行っています。例えば,人間の音声を音響的・言語的に分析・処理する基礎研究やこれらを応用したアプリケーションの研究開発,環境音の認識,文字検出/認識や手書き文字画像の生成等の画像認識などを行っています。また応用研究として、農業や医療分野での応用も行っています。これらの研究・開発を通じて,世の中の役に立つ技術を生み出し活用してもらうこと, 研究を通じて未来を担う人材の育成を行うことをモットーに日々精進しています。
Positioning of our laboratory in the Department of Mechatronics / メカトロニクス工学科における本研究室の位置付け
Our laboratory belongs to the Department of Mechatronics (Mechatronics Course in the Department of Engineering from 2024) in the Faculty of Engineering for undergraduate education. This department offers cross-disciplinary study of the core subjects of mechanics, electricity, and information systems, and aims to foster human resources involved in the research and development of embedded systems for industrial and autonomous robots, home appliances, automobiles, etc. Our laboratory is a part of the information science division within the department.
In our laboratory, we conduct research on artificial intelligence (understanding human spoken language, handwritten texts) and human-machine interfaces (voice user interfaces) in robots. A user interface is a method for a computer or machine to display information to the user or for the user to input information. Our goal is to improve the usability of information systems and machines through multimedia processing.
As deep learning is beginning to be used in embedded systems, those who wish to develop intelligent robots and machines or learn related technologies are welcome to visit our laboratory!