The fundamental research on speech recognition for the visually impaired and its application to voice interaction
| Project/Area Number |
16091210
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research on Priority Areas
|
| Allocation Type | Single-year Grants |
|---|
| Review Section |
Science and Engineering
|
|---|
| Research Institution | Tokyo Woman's Christian University |
|---|
Principal Investigator |
WATANABE Takayuki Tokyo Woman's Christian University, Colledge of Culture and Communication, Professor (80202414)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YASUMURA Michiaki Keio University, Faculty of Environmental, Professor (10230244)
ODA Kochi Tokyo Woman's Christian University, College of Culture and Communication, Professor (60169307)
NISHIMOTO Takuya The University of Tokyo, Graduate School of Information Science and Technology, Research Associate (80283696)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥15,700,000 (Direct Cost: ¥15,700,000)
Fiscal Year 2006: ¥4,000,000 (Direct Cost: ¥4,000,000)
Fiscal Year 2005: ¥3,900,000 (Direct Cost: ¥3,900,000)
Fiscal Year 2004: ¥7,800,000 (Direct Cost: ¥7,800,000)
|
|---|
| Keywords | visually impaired / Home Appliances / Web / spoken dialog / fast speech / human-to-human dialog / speech synthesys / accessibility / W3C / ロービジョン / 聴覚情報処理 / ナビゲーション |
|---|
| Research Abstract |
1) Based on the user studies for visually impaired, we developed two new prototypes of remote control for air-conditioner; one was the voice recognition type and the other was the ten-key keyboard. We conducted the evaluation tests for these RCs. After these studies, we developed yet another new RCs as the final year activities; one was the haptic feedback RC and the other was the integrated cellular-phone type RC. We should extend our RC to other home appliances in the future.
2) Task completion times of sighted and blind users were measured with two kinds of Web sites: sites marked up appropriately with heading elements and sites with the same visual appearance but with no heading elements marked up. The experiment was carried out with user agents that could navigate through heading elements. The results showed that i) task completion time was reduced by as much as one half with marked up heading elements, ii) the benefits of markup on task completion time were greater for blind users, and iii) the overall difference in response time between sighted and blind users diminished with sites that were appropriately marked up.
3) We investigated the intelligibility of synthesized voices at fast speaking rates. Four-digit random numbers are used as the vocabulary of the recall test. As the result, elderly persons can recall fast speech at some levels. However, their average recall rates are lower than the young university students and the individual differences are significant. The results of this tasks we consider are affected by the difficulty of auditory perception itself and the difficulty of recall the numbers in the correct order. Leaving out the effect of latter factor, it turned out that the task performances of elderly persons and young students are almost same. The learning effects of the elderly persons are not significant in either case, though those of the young students are significant for several weeks.
|
Report
(4 results)
Research Products
(42 results)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Journal Article] Effect of Learning on Listening to Ultra-Fast Synthesized Speech2006
Author(s)
Nishimoto, T., Sako, S., Sagayama, S., Ohshima, K., Oda, K., Watanabe T.
-
Journal Title
Proceedings of the 28th IEEE Engineering in Medicine and Biology Society Annual International Conference (EMBC2006)
Pages: 5691-5694
Description
「研究成果報告書概要(欧文)」より
Related Report
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
