This paper describes Japanese efforts to collect speech samples of various Japanese dialects and to create speech databases. A Grant-in-Aid for Scientific Research for Priority Area project, entitled Trosodic Features of Spoken Japanese', started in 1989, is an attempt to collect various dialect speech samples from all over Japan and create speech databases for the purpose of linguistic, phonetic, acoustic and physiological analyses. Plans exist for collecting common speech items from 100 districts all over Japan. These comprise about 1450 common items, including words and sentences. All recordings are to be performed using DAT. Some of these are to be edited and converted into compact discs and CD-ROMs.
Bibliographic reference. Itahashi, Shuichi (1991): "Large scale Japanese dialect speech corpora", In EUROSPEECH-1991, 513-516.