음성인식 프로그램에서 학습자 주도형 말하기 촉진 방안에 관한 연구

The purpose of this paper is to suggest how to facilitate student-centered learning in speech recognition English-speaking programs. Previous programs have commonly shown a serious problem in that learners quickly feel bored or discouraged in continuing speaking activities with the software. One of the main reasons is that learners are not allowed to actively manipulate their expressions in conversations. In order to facilitate student-centered speaking practice, the paper suggests adopting three approaches in establishing the database: 1) accommodating various expressions which carry the same illocution; 2) conversational branching by using Yes/No questions and adjacent-pair expressions; 3) utilizing ‘a helper’ to control the flow of the conversation and to practice structural patterns. After reviewing the key features of the recent speech recognition product Talklish Bible (TB), the paper attempts to demonstrate a sample lesson by applying those approaches to TB. It is also hoped that a future speech recognition program would give more consideration to providing learners with more opportunities for their voluntary participation in speaking with the computer software. Any addition of such approaches to a new program would make it possible to encourage Korean learners who want to improve their English speaking abilities without going abroad.