Abstract
In Taiwan, “Computers-Assisted Language Learning” (CALL) has become a popular method for language learning. Due to the frequent tests and scheduled syllabus, a few courses focus on the listening and speaking skills. Most “English-as-a-Foreign-Language” (EFL) students have to individually practice their listening and speaking skills after schools. However, when having learning problems, EFL students difficultly gained assistance. During the periods of self-practice, one-way learning interaction and the fixed content of a CALL system seem to be two main problems. Therefore, this study integrates speech mechanisms into a web-based CALL system to form a student-centered CALL system, namely “I Love Listening and Speaking English (I.L.L.S.E.).” When EFL students actively perform self-practice, the provided “text-to-speech” (TTS) mechanism is a pattern for the practice of the listening skill and the provided “automatic-speech-recognition” (ASR) mechanism is a partner for the practice of the speaking skill. This system assists EFL students to perform self-practice through the learning materials which are created by them and two-way human-machine interactions. During the period of this study, this study adopted both qualitative and quantitative approaches to analyze the participants’ learning statuses. 35 EFL vocational senior high school students participated in the experiment. Based on the research results, the low-level participants other than the medium- and high-level participants used the CALL system more frequently. Moreover, this study found: (1) All of the participants applied the CALL system for self-practice continuously and frequently. (2) The participants created 388 sentences (205 sentences did not contain any taught vocabulary). Most participants commonly revised various sentences that existed in the online course contents and additional learning materials for self-practice. (3) The participants entered 524 words (288 words did not exist in the online course contents). The participants actively performed self-practice when meeting strange words, vocabulary tenses, long-term vocabulary, and specified words. (4) Most participants obtained better performance when they had more and more experiences regarding the online tests. (5) In the online discussions, the participants clearly represented and shared their ideas. (6) Accompanying with the increased learning time, the participants frequently used the system functions. This study also applied the Mean-test method to comprehend the top five system functions where the participating students benefited, including: (1) this system clearly offered pronunciation; (2) the online test was useful; (3) this system was helpful for the practice of speaking skill; (4) this system was helpful for the practice of listening skill; and (5) this system assisted the participating students for learning English. In sum, the main contributions for the integration of a CALL system and English learning, including: (1) enhancing the seamless learning; (2) assisting EFL students to perform self-practice after classes; (3) enabling teachers to understand the students’ learning status; (4) implementing two-way learning interaction in real-time.