2018 Fiscal Year Research-status Report
A cross-institutional study into speech recognition for language learners, analyzing usage patterns
| Project/Area Number |
18K00825
|
|---|
| Research Institution | Kanazawa University |
|---|
Principal Investigator |
Gary Ross 金沢大学, 薬学系, 准教授 (10708142)
|
|---|
| Project Period (FY) |
2018-04-01 – 2022-03-31
|
| Keywords | online learning / speech recognition / speech synthesis |
|---|
| Outline of Annual Research Achievements |
The purpose of this project is to investigate how learners of English can use online speech recognition and synthesis to improve spoken language, and to improve spoken assessment. The first year plan was to focus on refining the online speech recognition and synthesis system itself, testing its robustness, and performing a pilot study on its effectiveness.
In the first semester, the system was extensively tested and used by students to assess usability, to integrate the practices into a pedagogically sound learning environment, and to develop effective speaking practices. The results of this student feedback as well as an overview of the system was presented as a poster at JALTCALL 2018.
In the second semester, the pilot study was performed on the effectiveness of the system. In this study, a group of 70 students were randomized into 2 groups, and used the system for a range of speaking tasks both with a grammatical focus and a communicative one. Having completed these tasks, their ability on the target phrases as well as their recall and understanding of the patterns were compared. The results of this study has been accepted as a paper presentation for JALTCALL 2019, and the data will be pattern analyzed.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
Research has progressed as expected. The online system was developed beyond what was expected with major improvements and usability. A small issue that has arisen is that iOS (iPhones and iPads) do not support speech recognition in any of their browsers. Since Android support is also variable, some users can only use their PC. Over the course of the 4 year study, there are likely to be dramatic changes in support, so student attitudes to devices will also be integrated into the qualitative assessment of speaking to a device.
Aside from the mobile issue, a pilot study was successfully carried out to measure the effectiveness of the system.
|
| Strategy for Future Research Activity |
The plan for this year is to publish the results of the pilot study as a research paper, extend the pilot study to other institutions and make a more robust assessment of the effectiveness of the system.
The initial pilot study was limited to a single class with a narrow range of abilities; by testing the effectiveness in other institutions a more robust assessment of the online speech as a learning tool will be possible. The preliminary results of these studies will be presented at EUROCALL 2019.
There will also be additional development of more varied activities to maintain student interest. Finally, the lead researcher will develop further tools to analyze the patterns of words learners produce.
|
| Causes of Carryover |
This year most of the budget will go towards overseas travel costs. Specifically, for a presentation at EUROCALL 2019, and for meetings with the Shimane and Tokyo co-investigators.
Additional costs will be incurred for server rental, backup, and publications.
|
