发音
纠正性反馈
计算机科学
介绍(产科)
考试(生物学)
控制(管理)
多媒体
语音识别
人机交互
自然语言处理
心理学
数学教育
人工智能
语言学
医学
古生物学
哲学
生物
放射科
作者
Yi Hsuan Wang,Shelley Shwu‐Ching Young
摘要
The purpose of the study is to explore and describe how to implement a pedagogical ASR-based intelligent computer-assisted speaking learning (iCASL) system to support adult learners with a private, flexible and individual learning environment to practice English pronunciation. The iCASL system integrates multiple levels of corrective feedback and allows learners to practice English speaking with immediate diagnosis of their utterances. The information of the multiple levels of feedback includes a list of words that are pronounced accurately and inaccurately, while audio recasting and demonstration of model utterances are adopted to facilitate speaking. To evaluate the effectiveness of the system, a total of 38 adults from Taiwan participated in this experiment, divided into an experimental and a control group. The control group practiced English speaking using the single-level-feedback system in which only a speaking score and waveform diagram was presented as feedback, while the experimental group was given the three-level-feedback iCASL system integrating implicit and explicit elements in the feedback presentation. The empirical evaluation reveals that there were significant differences between the pre-test and post-test speaking scores only for the learners with three-level-feedback iCASL system. The study is in accordance with the previous research that indicates specific feedback is needed and helpful for further enhancing their English pronunciation. Within the three-level feedback, the second level, which provides an explicit textual description with immediate recasting of the learners' utterances, is the most useful feedback for self-correcting learning. The learners received simultaneous dual information from the audio and visual presentation and hence had better potential to improve their English speaking than the learners who were provided with single modality feedback. Moreover, the learners were satisfied with the iCASL system for self-paced learning and they mentioned that the path suggestion function was especially beneficial because it helped them to manage their learning. The adult learners used the accumulated fragmented time of the informal learning opportunities to achieve improvement in their spoken language. Some research issues and suggestions are also presented for future reference.
科研通智能强力驱动
Strongly Powered by AbleSci AI