Title:
SPEAKER AUTHENTICATION SYSTEM, METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/075012
Kind Code:
A1
Abstract:
Provided is a speaker authentication system capable of achieving robustness against hostile samples. A data storage unit 112 stores data related to the voice of a speaker. Each of a plurality of voice processing units 11 performs speaker authentication on the basis of the input voice and the data stored in the data storage unit 112. A post-processing unit 116 specifies one speaker authentication result, on the basis of the speaker authentication results obtained respectively by the plurality of voice processing units 11. Pre-processing methods or parameters to be applied to the voice in each voice processing unit 11 are different for each voice processing unit 11.
Inventors:
MOMIYAMA SATORU (JP)
Application Number:
PCT/JP2019/040805
Publication Date:
April 22, 2021
Filing Date:
October 17, 2019
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06F21/32; G10L17/00
Domestic Patent References:
| WO1995005656A1 | 1995-02-23 |
Foreign References:
| US20080177684A1 | 2008-07-24 |
Other References:
XU WEILIN, EVANS DAVID, QI YANJUN: "Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks", NDSS 2018, 4 April 2017 (2017-04-04), pages 1 - 15, XP080760758
Attorney, Agent or Firm:
IWAKABE Fuyuki et al. (JP)
Download PDF:
Previous Patent: COMPONENT PRESENTATION DEVICE, COMPONENT PRESENTATION SYSTEM, COMPONENT PRESENTATION METHOD, AND COM...
Next Patent: AIR CONDITIONER
Next Patent: AIR CONDITIONER