| Project/Area Number |
25330130
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Multimedia database
|
|---|
| Research Institution | Toyohashi University of Technology |
|---|
Principal Investigator |
Tomoyosi Akiba 豊橋技術科学大学, 工学(系)研究科(研究院), 准教授 (00356346)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
中川 聖一 豊橋技術科学大学, リーディング大学院教育推進機構, 特任教授 (20115893)
|
|---|
| Project Period (FY) |
2013-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥4,940,000 (Direct Cost: ¥3,800,000、Indirect Cost: ¥1,140,000)
Fiscal Year 2015: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2014: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2013: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Keywords | ボイスサーチ / 自由発話音声 / 音声ドキュメント検索 / 情報検索 / テストコレクション / 音声中の検索後検出 / 音声内容検索 / 音声中の検索語検出 |
|---|
| Outline of Final Research Achievements |
Since current speech driven information retrieval systems simply replace textual input to speech, it fail to exploit the advantage comes from nature of speech, which is most natural, easy, and efficient way for humans to express their own ideas. This research work aims to develop the information retrieval systems that make use of such advantages by taking spontaneously spoken queries as their input. We collected spontaneously spoken queries through our subjective experiment, which result in compiling a reusable test collection. By using the collection, we took place the shared task of spoken document retrieval from spontaneously spoken queries within an international evaluation workshop, to which many research groups attended from home and abroad. Furthermore, by using the collected data., we also developed several novel methods for speech-driven spoken document retrieval, including that exploits acoustic features extracted from spontaneous spoken queries.
|
