Development of Material Automatic Generation System Based on the Analysis of Phonemic Errors in English Vocabulary Listening

摘要

With analyzing the learning log-data, personalized learning patterns of different students could be detected, and the personalized materials could be automatically created and offered to learners for guiding their specific learning processes and solving weak points. In this paper, we propose a new approach to develop a CALL system including an error-detecting algorithm and a material-creating module. And in our approach, we pay much of our attention on the detection of phonemic errors. The system can detect the phonemic errors of Japanese learners' in English vocabulary listening by analyzing the relative learning log data, and automatically create multiple-choice question materials to help students take practices to enhance perception on the phonemes that they distinguish difficultly.