Design and Development of an English Listening Website
| Project/Area Number |
12680218
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational technology
|
|---|
| Research Institution | The University of Aizu |
|---|
Principal Investigator |
TRIPP Steven D. University of Aizu, Computer Engineering, Professor, コンピュータ理工学部, 教授 (90271585)
|
|---|
| Project Period (FY) |
2000 – 2001
|
| Project Status |
Completed (Fiscal Year 2001)
|
| Budget Amount *help |
¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 2001: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2000: ¥1,500,000 (Direct Cost: ¥1,500,000)
|
|---|
| Keywords | English / Vocabulary / CBI / text analysis / vocabulary / LISTENING / MARK UP / GLOSSARY / AUDIO / SEMANTIC / NAVIGATION |
|---|
| Research Abstract |
This purpose of this project was to design and develop a website for practicing English listening. There were two parts to this project. The first was to develop software for the automatic markup of computer science documents with html tags and Java Script. The second part was to develop streaming Quick Time movies of the same documents with subtitles.
The first part of the project was realized by first doing and analysis of computer language and determining a basic set of common vocabulary. This set was used as a stoplist to define words the would NOT be marked up for dictionary lookup. The final markup pages allow one-click dictionary lookup with sample pronunciation of many words.
In the second part of the project the same documents were read and recorded onto video tape. That tape was then digitized and converted to "hinted" movies that can be accessed as streaming QuickTime. Several systems were tried for attaching subtitles to the movies. Although, this appears to be simple, in reality it is not. It was finally determined that the documents should be divided into four to five word phrases and save as text files. The text files were then converted to Quick Time movies in which each phrase appears for about two seconds. The movie was then played back as teleprompter for the video recording of the document. Thus the voice was synchronized with the text and not vice versa. Finally the text movies was merged with the voice movie and the final movie was "hinted" and place on the streaming server.
Finally links to both marked-up documents and streaming movies were placed on a webpage for student access. In all eight documents were marked up and turned to streaming movies with subtitles.
|
Report
(3 results)
Research Products
(2 results)
