| Project/Area Number |
15300035
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Nara Institute of Science and Technology |
|---|
Principal Investigator |
SARUWATARI Hiroshi Nara Institute of Science and Technology, Graduate School of Information Science, Associate Professor, 情報科学研究科, 助教授 (30324974)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SHIKANO Kiyohiro Nara Institute of Science and Technology, Graduate School of Information Science, Professor, 情報科学研究科, 教授 (00263426)
KAWANAMI Hiromichi Nara Institute of Science and Technology, Graduate School of Information Science, Assistant Professor, 情報科学研究科, 助手 (80335489)
TODA Tomoki Nara Institute of Science and Technology, Graduate School of Information Science, Assistant Professor, 情報科学研究科, 助手 (90403328)
李 晃伸 奈良先端科学技術大学院大学, 情報科学研究科, 助手 (80332766)
|
|---|
| Project Period (FY) |
2003 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥13,900,000 (Direct Cost: ¥13,900,000)
Fiscal Year 2006: ¥3,100,000 (Direct Cost: ¥3,100,000)
Fiscal Year 2005: ¥2,800,000 (Direct Cost: ¥2,800,000)
Fiscal Year 2004: ¥3,600,000 (Direct Cost: ¥3,600,000)
Fiscal Year 2003: ¥4,400,000 (Direct Cost: ¥4,400,000)
|
|---|
| Keywords | Sound scene / Blind signal processing / Source separation / Binaural / Audio augmented reality / Speech processing / 音響拡張現実 |
|---|
| Research Abstract |
In this research, we resolve a source separation problem, which was considered within acoustics problem, by combining information theory and acoustic signal processing theory. Also we mainly aim to construct a general system consisting of "sound scene decomposition, synthesis, and modification," unlike a simple elemental technology of source separation. As a result, in particular, we successfully realize the various sound-scene-processing system which is based on SIMO-ICA algorithm, binaural-type independent component analysis, as follows :
(1) We can realize the blind source separation system which combines information theory and spatial acoustic signal processing.
(2) We conducted a theoretical analysis of SIMO-ICA-based blind source separation algorithm, and developed the real-time implementation. Also the evaluations on the system were carried out under realistic acoustic environments.
(3) We prepared a specific database for binaural sound source separation experiment.
(4) We can show an SIMO-ICA's applicability to hands-free speech recognition via evaluation of efficacy of sound source separation and speech recognition performances.
(5) Sound scene decomposition-synthesis system (binaural hearing-aid system) can be developed by using the real-time ICA unit.
The related publications to this research project are as follows : 17 journal papers, 53 reviewed international conference papers, 8 invited talks, and 53 domestic conferences, including 11 scientific awards.
|
