Tamper-resistant element for use in speaker recognition

Patent metadata

Field	Value
Publication number	US-10056084-B2
Application number	US-201615216196-A
Country	US
Kind code	B2
Filing date	Jul 21, 2016
Priority date	Jan 17, 2014
Publication date	Aug 21, 2018
Grant date	Aug 21, 2018

Abstract

A tamper-resistant element for use in speaker recognition, the tamper-resistant element being adapted for storing data representing speaker information based on speaker recognition enrollment data and for checking whether information based on a speaker recognition testing signal matches the speaker information, characterized in that the tamper-resistant element is also adapted for carrying out a data integrity check, a system comprising such a tamper-resistant element and method for speaker recognition.

First claim

The invention claimed is:

1. A tamper-resistant element, adapted for speaker recognition, the tamper-resistant element being adapted for storing data representing speaker information based on speaker recognition enrollment data and for checking whether first information based on a speaker recognition testing signal matches the speaker information, wherein the tamper-resistant element is adapted for storing information for use in a spoof check, and for carrying out a spoof check using second information based on the speaker recognition testing signal, and in that the tamper-resistant element is also adapted for carrying out a data integrity check on the first information and the second information, wherein the data integrity check comprises an injection integrity check on whether the speaker recognition testing signal has been substituted by a signal that had been previously produced and accepted matching.

2. The tamper-resistant element according to claim 1 , wherein the tamper-resistant element is adapted for storing data representing information for use in the data integrity check.

3. The tamper-resistant element according to claim 1 , wherein the tamper-resistant element is adapted for text-dependent speaker recognition, wherein the tamper-resistant element is optionally adapted for use with a text-dependent Hidden Markov Model approach to speaker recognition.

4. The tamper-resistant element according to claim 1 , wherein the tamper-resistant element is adapted for text-independent speaker recognition, wherein the tamper-resistant element is adapted for use with a text-dependent i-vector-based speaker recognition approach.

5. The tamper-resistant element according to claim 1 , wherein the tamper-resistant element comprises one of: a smart card, a secure server, a trusted platform module, and an ARM trust zone.

6. A tamper-resistant element, adapted for speaker recognition, the tamper-resistant element being adapted for storing data representing speaker information based on speaker recognition enrollment data and for checking whether first information based on a speaker recognition testing signal matches the speaker information, wherein the tamper-resistant element is adapted for storing information for use in a spoof check, and for carrying out a spoof check using second information based on the speaker recognition testing signal, and in that the tamper-resistant element is also adapted for carrying out a data integrity check on the first information and the second information, wherein the data integrity check comprises a speaker recognition knowledge based integrity check to check that no data is modified to take advantage of knowledge of the speaker recognition system, and wherein the tamper-resistant element is optionally also adapted for storing information for use in the speaker recognition knowledge based integrity check.

7. The tamper-resistant element according to claim 6 , wherein the tamper-resistant element is adapted for storing data representing information for use in the data integrity check.

8. The tamper-resistant element according to claim 6 , wherein the tamper-resistant element is adapted for text-dependent speaker recognition, wherein the tamper-resistant element is optionally adapted for use with a text-dependent Hidden Markov Model approach to speaker recognition.

9. The tamper-resistant element according to claim 6 , wherein the tamper-resistant element is adapted for text-independent speaker recognition, wherein the tamper-resistant element is adapted for use with a text-dependent i-vector-based speaker recognition approach.

10. The tamper-resistant element according to claim 6 , wherein the tamper-resistant element comprises one of: a smart card, a secure server, a trusted platform module, and an ARM trust zone.

11. A system comprising a tamper-resistant element for speaker recognition, the tamper-resistant element being adapted for storing data representing speaker information based on speaker recognition enrollment data and for checking whether first information based on a speaker recognition testing signal matches the speaker information, wherein the tamper-resistant element is adapted for storing information for use in a spoof check, and for carrying out a spoof check using second information based on the speaker recognition testing signal, and in that the tamper-resistant element is also adapted for carrying out a data integrity check on the first information and the second information,
wherein the system is adapted for deriving information based on the speaker recognition testing signal from the speaker recognition testing signal outside of the tamper-resistant element, and
wherein the system is adapted for deriving said first information based on the speaker recognition testing signal outside of the tamper-resistant element in parallel to deriving said second information based on the speaker recognition testing signal outside of the tamper-resistant element.

12. The system according to claim 11 , wherein one, two, three or more hyperparameters are stored outside of the tamper-resistant element.

13. The system according to claim 11 , wherein the system is adapted for deriving said first information based on the speaker recognition testing signal outside of the tamper-resistant element.

14. The system according to claim 11 , wherein the system is adapted for deriving said second information based on the speaker recognition testing signal outside of the tamper-resistant element.

15. The system according to claim 11 , wherein the system is adapted for maintaining data integrity of data representing information based on the speaker recognition testing signal or any data derived from the speaker recognition testing signal outside of the tamper-resistant element.

16. The system according to claim 11 , wherein the system is adapted such that the tamper-resistant element is adapted to carry out a spoof check of the speaker recognition testing signal based on features derived from the speaker recognition testing signal outside of the tamper-resistant element.

17. The system according to claim 11 , wherein the tamper-resistant element comprises one of: a smart card, a secure server, a trusted platform module, and an ARM trust zone.

18. A system comprising a tamper-resistant element for speaker recognition, the tamper-resistant element being adapted for storing data representing speaker information based on speaker recognition enrollment data and for checking whether first information based on a speaker recognition testing signal matches the speaker information, wherein the tamper-resistant element is adapted for storing information for use in a spoof check, and for carrying out a spoof check using second information based on the speaker recognition testing signal, and in that the tamper-resistant element is also adapted for carrying out a data integrity check on the first information and the second information,
wherein the system is adapted for deriving information based on the speaker recognition testing signal from the speaker recognition testing signal outside of the tamper-resistant element, and
wherein the system is adapted such that the tamper-resistant element is adapted to carry out a check of a speaker recognition knowledge based integrity of the data representing information based on the speaker recognition testing signal received from outside of the tamper-resistant element.

19. The system according to claim 18 , wherein one, two, three or more hyperparameters are stored outside of the tamper-resistant element.

Assignees

• Cirrus Logic Int Semiconductor Ltd

Inventors

• Aviles-Casco Carlos Vaquero
• Rodriguez Luis Buera
• Gomar Marta Garcia

Classifications

• G06F21/72

  in cryptographic circuits · CPC title

• G10L19/018

  Audio watermarking, i.e. embedding inaudible data in the audio signal · CPC title

• G10L17/16

  Hidden Markov models [HMM] · CPC title

• G06F21/34

  involving the use of external additional devices, e.g. dongles or smart cards · CPC title

• G10L17/00

  Speaker identification or verification techniques · CPC title