Research areas

• Degign and compilation of a real-time MRI database of speech articulation"
• Design and compilation of a balanced corpus of written Japanese
• Analysis of spontaneous speech corpus
• Analysis of language variation using spontaneous speech corpus
• Analysis of the transmission mechanism of the paralinguistic information in speech
• Prosodic structure of Japanese
• etc.

Short CV

• 1956: Born in Kyoto, Japan.
• 1980: BA from Sophia University (French studies).
• 1982: MA from Sophia University (Linguistics).
• 1982-83: Ph.D. student of linguistics, Sophia University.
• 1984-88: Lecturer in Japanese linguistics, Tottori University.
• 1989: Researcher, Dept. Language Behavior, The National Language Research Institute.
• 1993: Senior researcher, Dept. Language Behavior, The National Language Research Institute.
• 1993-94: Visiting scholar, Dept. Linguistics, the Ohis State University.
• 1994: Section Head, Dept. Language Behavior, The National Language Research Institute.
• 2001-2005: Section Head, Dept. Language Research, The National Institute for Japanese Language (former National Language Research Institute).]
• 2006-2009: Group Leader, Corpus Compilation Group, Dept. Language Research, The National Institute for Japanese Language.
• 2009-2012: Professor and director, Dept. Corpus Studies, The National Institute for Japanese Language and Linguistics.
• 2013-2016: Deputy director-general, Professor and director, Dept. Corpus Studies, The National Institute for Japanese Language and Linguistics.
• 2009-2019: Director, Center for Corpus Development.
• 2016-2022: Professor, Spoken Language Division.

About my research activities

All my research works are concerned with spoken language. The earliest studies done while I was a graduate students involved quantitative analysis of vowel mergers in two Japanese dialects, Tsugaru and Izumo, statistical analysis of the factors of vowel devoicing, and experimental investigation about the syllable structure of Akita dialect.

After 1984, when I obtained a lecturership at the Tottori University, I spent most of my time doing dialect geography and sociolinguistics of Tottori and Izumo dialects. I conducted linguistic interview of more than 800 subjects with the help of my students. It was also this period that I developed a PC-based map editing program, called EGL, for dialect geographers.

In 1989, I moved to the phonetics laboratory of the National Language Research Institute (National Institute for Japanese Language, after 2001). Early studies at the NLRI were concerned with experimental studies of Japanese intonation including both Tokyo Japanese and so-called 'acdcentless' dialects like Kumamoto Japanese.

In 1993 I could have occasion to visit the phonetics laboratory of the Ohio State University led by professor Mary Beckamn as a visiting scholar, and stayed there for 10 months. It was also at this occasion that I had acquaintance of professors Osamu Fujimura and Ilse Lehiste.

In 1997, I started experimental study about the transmission of 'paralinguistic' information by speech, inspired by the pioneering work of professor Hiroya Fujisaki. During the course of this study I was helped by many researchers outside the NLRI: Professor Shigeru Kiritani of the Research Institute of Logopedics and Phoniatrics, University of Tokyo, Professor Hideki Kasuya of Utsunomiya University, Dr. Masaaki Honda of the NTT Basic Research Laboratory, and Dr. Kiyoshi Honda of ATR Human Information Processing Laboratoy.

In 1999, before I could finalize the work of paralangauge, I was suddenly involved in a new joint project among NLRI, Communications Research Laboratory, and Tokyo Institute of Technology. The project was supervised by professor Sadaoki Furui of Tokyo Institute of Technology, and its primary aim consisted in the development of new technology for the processing of spontaneous speech. The mission of the NLRI group that I managed consisted in the construction of a large-scale corpus of spontaneous Japanese, which is known today as the Corpus of Spontaneous Japanese. Construction of the world's largest spontaneous speech corpus within five years required hard working, but I could somehow achieved the goal due mostly to the efforts of young colleagues.

Since 2006, I started with my colleagues of the NIJL (former NLRI) a new project aiming at the construction of more than a hundred-million words balanced corpus of contemporary written Japanese known as the BCCWJ, which is the most important corpus of the NIJL's KOTONOHA corpus development initiative. The goal of the project is to make the BCCWJ publicly available in the year of 2011.

In the summer of 2006, shortly after its beginning, the project was celebrated by the news that our proposal for the grant-in-aid for scientific research on priority area program was accepted by the Ministry of Education. To our utter surprise, our proposal was accepted with little modification (the filling rate being more than 97%). Currently more than 50 researchers joined the project in and outside the NIJL.

From May 2011, more than 100 million words are available for full-text retrieval at https://chunagon.ninjal.ac.jp/
Also, in August 2011, Chuunagon, a web interface for the retrieval of POS-analyzed BCCWJ data becomes publicly available. Users who want to use Chuunagon are requested to apply for registration by snail mail. Please visit https://chunagon.ninjal.ac.jp/.

In 2017, I started a new project in the field of speech production: The Real-Time MRI Database of Articulatory Movement (rtMRIDB). This database visualized the changing shape of the mid-sagittal section of the vocal tract by means of the real-time MRI movie technique. Part of the data is available here.