研究課題/領域番号 |
20K00807
|
研究機関 | 高知工科大学 |
研究代表者 |
ダニエルズ ポール 高知工科大学, 共通教育教室, 教授 (50307245)
|
研究分担者 |
熊井 信弘 学習院大学, 付置研究所, 教授 (00248999)
|
研究期間 (年度) |
2020-04-01 – 2023-03-31
|
キーワード | automated / scoring / speech / speaking / feedback |
研究実績の概要 |
Several developmental updates were completed in 2020 to our open-source automated speech scoring platform. Extensive testing was conducted to investigate both the accuracy of the computer-based scores and the effectiveness of the speaking tasks. Initial results indicated an acceptable reliability level of the computer-based scores generated by the platform. In addition, improved speed and accuracy of speaking feedback was observed when accessing the platform from a mobile device. The researchers conducted in-class research on the use of the automated speaking assessment system, and found that students' overall listening comprehension scores improved between pre and post-testing. A descriptive analysis was also conducted in this study using observation and an online survey. The results indicated that the students found the listening-focused activities engaging, and felt that their listening abilities improved, especially with intensive listening and speaking tasks. Finally, initial research was conducted on using smart devices rather than a personal computer or mobile device to administer the computer-scored speaking tasks. Initial tests indicated that learners were better able to work together in groups completing speaking activities that were administered via a smart device rather than a personal device.
|
現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
The platform was successfully deployed in online language learning classrooms during the first half of the year, although some issues emerged when users were accessing the speaking platform via a mobile device. During the second half of the academic year, learners mainly accessed the platform via on-campus PCs and fewer issues were encountered.
The platform coding proceeded efficiently in both the development and improvements stages. Frequent and productive online communication between both researchers and the system programmer ensured that the project moved ahead on schedule. Overall the system appears to be stable and robust at this point, with several hundred users accessing the online speaking tasks.
|
今後の研究の推進方策 |
1. Improved support for mobile learning environments will be evaluated. 2. Speaking activities that can be auto-scored using a smart device such as a smart speaker will be developed and evaluated. 3. Scores derived from the automated speech scoring platform will be compared to human-generated scores for similar speaking tasks to determine the reliability of the computerized scoring methods. 4. Student-generated speaking data will be continually assessed to further improve the reliability of the speaking scores generated by the computer.
|
次年度使用額が生じた理由 |
Because of the COVID-19 restrictions in 2020, the researchers were not able to attend academic meetings and thus incurred an amount that will be used in 2021.
|