A Computer Assisted Language Learning System based on Error Trends Grouping

Abstract

Mastering several languages almost becomes a necessity in today's globalization environment. Computer Assisted language learning (CALL) systems are as a popular second language learning tool because of its inherent advantages. Through analyzing traditional language education and existing CALL tools, we have developed an experimental CALL system which can give relevance feedbacks through anatomy animation. It is based on deliberate data trained recognition to give error trend relevant feedbacks to language learners. The feedbacks are provided through different visualized speech features as well as articu-latory anatomy animation. The visual features are easier to interpret and compare to the correct pronunciations so that the learning process can be made easier. Comparative evaluations to a simple playback system without feedbacks show that the proposed system is more effective.