SYSTEM AND METHOD FOR AUTHENTICATION USING SPEAKER VERIFICATION TECHNIQUES AND FRAUD MODEL

• US 20140222678A1
• Filed: 01/29/2014
• Published: 08/07/2014
• Est. Priority Date: 02/05/2013
• Status: Active Grant

• Current Assignee: Visa International Service Association (Visa, Inc.)
• Original Assignee: Visa International Service Association (Visa, Inc.)
• Inventors: Sheets, John F., Wagner, Kim R.
• Granted Patent: US 9,117,212 B2
• Time in Patent Office: Days
• US Class Current: 705/44
• CPC Class Codes: G06Q 20/3821 Electronic credentials
• CPC Class Codes: G06Q 20/4014 Identity check for transactions
• CPC Class Codes: G06Q 20/40145 Biometric identity checks
• CPC Class Codes: G06Q 20/4016 involving fraud or risk level assessment in transaction processing
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase

Abstract

Embodiments of the invention provide for speaker verification on a communication device without requiring a user to go through a formal registration process with the issuer or network. Certain embodiments allow the use of a captured voice sample attempting to reproduce a word string having a random element to authenticate the user. Authentication of the user is based on both a match score indicating how closely the captured voice samples match to previously stored voice samples of the user and a pass or fail response indicating whether the voice sample is an accurate reproduction of the word string. The processing network maintains a history of the authenticated transactions and voice samples.

Claims

1. A method for authenticating a user for a transaction, comprising:

providing, by a device, a word string that comprises a random element;

transmitting an audio segment, to a server computer, wherein the audio segment originated from the user;

authenticating the user for the transaction based at least in part on a response, from the server computer, to the transmitted audio segment; and

receiving, from the server computer, an indication that the user is authenticated for the transaction, wherein the indication is received after a delay in updating the fraud model for the user is updated with the audio segment upon the audio segment being held in a queue for a predetermined period of time, and when no fraud has been reported for the predetermined period of time.

2. The method of claim 1 further comprising receiving a match score indicative of a comparison between the audio segment and a plurality of previously received audio segments created by the user or a model generated based on the previously received audio segments.

3. The method of claim 1 wherein the audio segment is a reproduction, by the user, of the provided word string.

4. The method of claim 1 wherein the response is indicative of whether the received audio segment is an accurate reproduction of the provided word string.

5. The method of claim 1 wherein the word string is seven words or less.

6. The method of claim 1 wherein the server computer is a voice biometric matching server.

7. The method of claim 1 wherein the fraud model is based on a plurality of audio segments received from the user.

8. The method of claim 1 further comprising displaying, by the device, the word string to the user.

9. The method of claim 1 wherein the random element is preceded by a fixed element of the word string.

10. The method of claim 9 wherein the fixed element is greater than the random element.

11. A device, comprising:

a processor; and

a non-transitory computer-readable storage medium, comprising code executable by the processor for implementing a method for authenticating a user for a transaction, the method comprising;

providing, by the device, a word string that comprises a random element;

transmitting an audio segment, to a server computer, wherein the audio segment originated from the user;

authenticating the user for the transaction based at least in part on a response, from the server computer, to the transmitted audio segment; and

receiving, from the server computer, an indication that the user is authenticated for the transaction, wherein the indication is received after a delay in updating the fraud model for the user is updated with the audio segment upon the audio segment being held in a queue for a predetermined period of time, and when no fraud has been reported for the predetermined period of time.

12. The device of claim 11 wherein the method further comprises receiving a match score indicative of a comparison between the audio segment and a plurality of previously received audio segments created by the user or a model generated based on the previously received audio segments.

13. The device of claim 12 wherein the audio segment is a reproduction, by the user, of the provided word string.

14. The device of claim 11 wherein the response is indicative of whether the received audio segment is an accurate reproduction of the provided word string.

15. The device of claim 11 wherein the word string is seven words or less.

16. The device of claim 11 wherein the server computer is a voice biometric matching server.

17. The device of claim 11 wherein the fraud model is based on a plurality of audio segments received from the user.

18. The device of claim 11 wherein the device further comprises a display configured to display the word string to the user.

19. The device of claim 11 wherein the random element is preceded by a fixed element of the word string.

20. The device of claim 19 wherein the fixed element is greater than the random element.