Project/Area Number |
08680399
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Intelligent informatics
|
Research Institution | Nara Institute of Science and Technology |
Principal Investigator |
SHIKANO Kiyohiro Graduate School of Information Science, Nara Institute of Science and Technology, Professor, 情報科学研究科, 教授 (00263426)
|
Co-Investigator(Kenkyū-buntansha) |
NAKAMURA Satoshi Graduate School of Information Science, Nara Institute of Science and Technology, 情報科学研究科, 助教授 (30263429)
|
Project Period (FY) |
1996 – 1997
|
Project Status |
Completed (Fiscal Year 1997)
|
Budget Amount *help |
¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 1997: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1996: ¥1,300,000 (Direct Cost: ¥1,300,000)
|
Keywords | Dictation / HMM pheneme model / language model / Continuous Speech Recognition / 音声認識 / ディクテーションシステム / 音声データベース |
Research Abstract |
Dictation is one of the final goal of speech recognition. In USA and EU,there have been researched in the speech recognition projects supported by the governments, such like ARPA and ESPRIT.To catch up with these projects, we had organized "Large-Vocabulary Continuous Speech Recongnition Database WG". The WG made it possible to collect a large ammount of speech database for the newspaper articles by the collaboration with many industrial research laboratories and universities. The WG activity has been extended to the IPA project for Japanese dictation software development. This grant-aid for scientific research not only promoted the research activities in the investigater's laboratory, but also helped to organize the IPA nation-wide dictation project. The IPA dictation project shows 92% word accuracy for the 5,000-word newspaper article task. The developed acoustic models, language models and recognition program becomes the Japanese performance targets for speech recognition research and development.
|