An interactive dialog system for learning Japanese

Subarashii is a system that uses automatic speech recognition (ASR) to offe r first-level, computer-based exercises in the Japanese language for beginn ing high school students. Building the Subarashii system has identified str engths and limitations of ASR technology. The system was tested with 34 stu dents at Silver Creek High School in San Jose, California and with 13 stude nts at Stanford University in Stanford, California. Recognition accuracy wa s measured and user errors were analyzed. The functional accuracy defined a s the percentage of time when the system performs the correct functional be havior turned out to be generally higher than the per-utterance speech reco gnition accuracy. (C) 2000 Elsevier Science B.V. All rights reserved.