A study on Japanese pronunciation learning system for Korean using speech recognition

摘要

This paper describes a CALL (Computer Assisted Language Learning) system for teaching the pronunciation of Japanese to Korean speaker. First, Japanese sentences uttered by adult Korean speakers were evaluated by Japanese native speakers. Then, the Japanese Learning System based on the evaluation result was developed. Our CALL system asks the learner to read a sentence including minimal pairs. Speech recognition technology is used to compare Japanese speech uttered by a learner to the utterance by a native speaker and the system automatically calculates intelligibility scores which indicate the similarity between the learner's speech and the standard Japanese native's speech. Furthermore, when the learner make pronunciation mistake, the learner confirm his/her mispronunciation. The system also gives a proper instruction of the pronunciation.