TY - CONF
T1 - SHARABLE SOFTWARE REPOSITORY FOR JAPANESE LARGE VOCABULARY CONTINUOUS SPEECH RECOGNITION
AU - Kawahara, Tatsuya
AU - Kobayashi, Tetsunori
AU - Takeda, Kazuya
AU - Minematsu, Nobuaki
AU - Itou, Katsunobu
AU - Yamamoto, Mikio
AU - Yamada, Atsushi
AU - Utsuro, Takehito
AU - Shikano, Kiyohiro
N1 - Funding Information:
The authors are grateful to advisory members of the project for their comments and cooperation. We are also debt to Mr. Akinobu Lee, who has been engaged in development and assessment.
Publisher Copyright:
© 1998. 5th International Conference on Spoken Language Processing, ICSLP 1998. All rights reserved.
PY - 1998
Y1 - 1998
N2 - The project of Japanese LVCSR (Large Vocabulary Continuous Speech Recognition) platform is introduced. 1 It is a collaboration of researchers of different academic institutes and intended to develop a sharable software repository of not only databases but also models and programs. The platform consists of a standard recognition engine, Japanese phone models and Japanese statistical language models. A set of Japanese phone HMMs are trained with ASJ (Acoustic Society of Japan) databases of 20K sentence utterances per each gender. Japanese word N-gram (2-gram and 3-gram) models are constructed with a corpus of Mainichi newspaper of four years. The recognition engine JULIUS is developed for assessment of both acoustic and language models. The modules are integrated as a Japanese LVCSR system and evaluated on 5000-word dictation task. The software repository is available to the public.
AB - The project of Japanese LVCSR (Large Vocabulary Continuous Speech Recognition) platform is introduced. 1 It is a collaboration of researchers of different academic institutes and intended to develop a sharable software repository of not only databases but also models and programs. The platform consists of a standard recognition engine, Japanese phone models and Japanese statistical language models. A set of Japanese phone HMMs are trained with ASJ (Acoustic Society of Japan) databases of 20K sentence utterances per each gender. Japanese word N-gram (2-gram and 3-gram) models are constructed with a corpus of Mainichi newspaper of four years. The recognition engine JULIUS is developed for assessment of both acoustic and language models. The modules are integrated as a Japanese LVCSR system and evaluated on 5000-word dictation task. The software repository is available to the public.
UR - http://www.scopus.com/inward/record.url?scp=85128402603&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85128402603&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:85128402603
T2 - 5th International Conference on Spoken Language Processing, ICSLP 1998
Y2 - 30 November 1998 through 4 December 1998
ER -