Research on robust speaker verification against spoofing attacks by voice imitation
Project/Area Number |
25330206
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Perceptual information processing
|
Research Institution | Tokyo City University |
Principal Investigator |
Iwano Koji 東京都市大学, メディア情報学部, 教授 (90323823)
|
Co-Investigator(Kenkyū-buntansha) |
篠田 浩一 東京工業大学, 情報理工学院, 教授 (10343097)
|
Project Period (FY) |
2013-04-01 – 2017-03-31
|
Project Status |
Completed (Fiscal Year 2016)
|
Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2013: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
|
Keywords | 話者照合 / 模倣音声 / 詐称攻撃 / 話者認識 / 個人認証 |
Outline of Final Research Achievements |
In this research, we experimentally analyzed the effect of spoofing attacks by voice imitation on speaker verification systems and acoustical characteristics of the imitated voice. These analyses were conducted by using our original speech data consisting of professional and non-professional impersonators' imitated voice. The analysis results show that the voice imitation by non-professional impersonators significantly increases the success rate of spoofing attacks and the higher success rate is yielded by the professional impersonator's imitation. We also proposed a method for quantitatively evaluating the quality of voice imitation. The method reveals that the professional imitator efficiently approaches his voice characteristics towards the target speaker's voice.
|
Report
(5 results)
Research Products
(6 results)