| Co-Investigator(Kenkyū-buntansha) |
飛松 省三 九州大学, 医学研究院, 教授 (40164008)
松尾 行雄 東北学院大学, 教養学部, 教授 (40323117)
Remijn GerardB. (REMIJN Gerard B. / REMIJN GerardB.) 九州大学, 芸術工学研究院, 准教授 (40467098)
上田 和夫 九州大学, 芸術工学研究院, 准教授 (80254316)
|
|---|
| Budget Amount *help |
¥40,300,000 (Direct Cost: ¥31,000,000、Indirect Cost: ¥9,300,000)
Fiscal Year 2017: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Fiscal Year 2016: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
Fiscal Year 2015: ¥6,110,000 (Direct Cost: ¥4,700,000、Indirect Cost: ¥1,410,000)
Fiscal Year 2014: ¥13,910,000 (Direct Cost: ¥10,700,000、Indirect Cost: ¥3,210,000)
Fiscal Year 2013: ¥8,450,000 (Direct Cost: ¥6,500,000、Indirect Cost: ¥1,950,000)
|
|---|
| Outline of Final Research Achievements |
Conventional technologies related to speech transmission are often based on crude impressions of speech announcements or environments. However, the most important thing about speech announcements is whether the contents are conveyed correctly, and it is desired to develop speech signals resilient against noise and reverberation. We performed basic and applied studies from the viewpoints of auditory psychology, speech science, and neurophysiology. Our multivariate analyses revealed that four factors are sufficient to express essential spectral fluctuations of speech: Noise-vocoded speech resynthesized from the four factors was highly intelligible. One of the factors related to a frequency range around 600-1700 Hz corresponded to what is called ‘sonority’ in phonology.
|
