5th International Conference on Spoken Language Processing Sydney, Australia November 30 - December 4, 1998

Sharable Software Repository for Japanese Large Vocabulary Continuous Speech Recognition

Tatsuya Kawahara (1), Tetsunori Kobayashi (2), Kazuya Takeda (3), Nobuaki Minematsu (4), Katsunobu Itou (5), Mikio Yamamoto (6), Atsushi Yamada (7), Takehito Utsuro (8), Kiyohiro Shikano (8)

(1) Kyoto Univ., Japan
(2) Waseda Univ., Japan
(3) Nagoya Univ., Japan
(4) Toyohashi Univ. of Tech., Japan
(5) ETL, Japan
(6) Tsukuba Univ., Japan
(7) ASTEM, Japan
(8) Nara Institute of Science and Technology, Japan

The project of Japanese LVCSR (Large Vocabulary Continuous Speech Recognition) platform is introduced. 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.

Full Paper

Bibliographic reference.  Kawahara, Tatsuya / Kobayashi, Tetsunori / Takeda, Kazuya / Minematsu, Nobuaki / Itou, Katsunobu / Yamamoto, Mikio / Yamada, Atsushi / Utsuro, Takehito / Shikano, Kiyohiro (1998): "Sharable software repository for Japanese large vocabulary continuous speech recognition", In ICSLP-1998, paper 0763.