Systems and methods for authentication program enrollment

• US 10,013,984 B1
• Filed: 01/12/2017
• Issued: 07/03/2018
• Est. Priority Date: 09/19/2014
• Status: Active Grant

• Current Assignee: United Services Automobile Association
• Original Assignee: United Services Automobile Association
• Inventors: Johnson, Zakery Layne, Mortensen, Maland Keith, Fernandez, Gabriel Carlos, Casillas, Debra Randall, Rangarajan, Sudarshan, Buckingham, Thomas Bret
• Primary Examiner(s): McFadden, Susan
• Application Number:
  US15/404,468
• Time in Patent Office: 537 Days
• Field of Search: 704235, 704273
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• 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
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 25/60 for measuring the quality of voice signals
• CPC Class Codes: H04L 63/0861 using biometrical features, e.g. fingerprint, retina-scan cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231

Abstract

Various embodiments of the technology described herein alleviate the need to specifically request enrollment information from a user to enroll the user in a voice biometric authentication program. For example, after receiving a call from a user, the system can identify the user and analyze the user'"'"'s biometric information when the user speaks a command or request. The system can use the user'"'"'s spoken command or request as enrollment information for the particular command or request or for all spoken requests. After enrollment into the voice biometric authentication program, the system can authenticate the user using biometric information before fulfilling requests or commands.

Claims

1. A non-transitory, computer-readable medium containing instructions that, when executed by one or more processors, cause a machine to:

receive, via a network, a voice interaction from a user, wherein the voice interaction includes a request or a command identifiable from a voice of the user, wherein the request or the command requests information or commands an action;

request non-voice biometric authentication information from the user to fulfill the request or the command made during the voice interaction, wherein the authentication information is used to authenticate the user;

after the user is authenticated to or above an authentication threshold, use the non-voice biometric authentication information to;

fulfill the request or the command;

enroll the user into an authentication program that uses a voice biometric authentication method by deriving characteristics of the voice of the user received during the request or the command and associating the voice characteristics with an account of the user; and

after enrollment into the authentication program, authenticate the user via the voice biometric authentication method prior to fulfilling requests or commands made during a voice navigation.

2. The non-transitory, computer-readable medium of claim 1, wherein the voice biometric authentication method comprises:

receiving a voice interaction after the enrollment into the authentication program;

deriving characteristics of a voice from the voice interaction after the enrollment into the authentication program;

comparing the characteristics of the voice in the voice interaction after the enrollment into the authentication program with characteristics previously derived from one or more previously received voice interactions from the user; and

authenticating the user when a match of the characteristics of the voice in the voice interaction after the enrollment into the authentication program and the characteristics previously derived from the one or more previously received voice interactions meet or exceed a threshold.

3. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to:

convert speech from the voice interaction into text, and wherein the voice biometric authentication method uses text-dependent speech recognition to authenticate the user.

4. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to:

receive a series of voice interactions; and

create a voice profile by analyzing the series of the voice interactions, wherein the user is enrolled after the voice profile is created.

5. The non-transitory, computer-readable medium of claim 4, wherein the series of the voice interactions includes at least two similar voice commands.

6. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to collect samples of the user'"'"'s voice during a series of voice interactions, wherein voice characteristics of the user'"'"'s voice are captured during the series of voice interactions.

7. The non-transitory, computer-readable medium of claim 6, wherein voice interactions in the series of voice interactions include a conversation with a representative.

8. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to determine a number and a quality of voice interactions, wherein the user is enrolled into the voice biometric authentication method when the number and the quality of the voice interactions exceed an enrollment threshold.

9. The non-transitory, computer-readable medium of claim 1, wherein the user is enrolled when the user speaks a spoken command during at least two different voice navigations.

10. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to notify the user that the user is enrolled into the authentication program.

11. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, further cause the machine to:

receive a second request or command;

determine an activity authentication level to fulfill the second request or command;

analyze the characteristics of the second request or command, the characteristics including at least one of;

pitch, intonation, speed, attenuation, accent, cadence, and volume;

authenticate the user via the voice biometric authentication method to an authentication level based on the characteristics of the second request or command compared to characteristics of the voice interaction; and

fulfill the second request or command when the authentication level meets or exceeds the activity authentication level.

12. The non-transitory, computer-readable medium of claim 1, wherein the instructions, when executed by the one or more processors, that cause the machine to enroll the user further cause the machine to enroll the user for specific requests using the voice navigation.

13. A non-transitory, computer-readable medium containing instructions that, when executed by one or more processors, cause a machine to:

receive a spoken command from a user, wherein the spoken command commands an action;

request authentication information from the user to fulfill the spoken command;

after receiving the authentication information, authenticate the user via a first authentication method using the authentication information, wherein the first authentication method uses non-voice biometric credentials to authenticate the user; and

after authenticating the user via the first authentication method;

fulfill the spoken command, and

enroll, without requesting enrollment information, the user into an authentication program for the spoken command, wherein the authentication program uses a second authentication method, wherein enrolling the user into the authentication program comprises determining voice characteristics from the spoken command,

wherein the second authentication method includes authenticating the user by comparing characteristics of a voice of the user speaking the spoken command during subsequent requests with the determined characteristics of the voice of the user speaking the spoken command.

14. The non-transitory, computer-readable medium of claim 13, further comprising after enrollment in the authentication program, fulfilling the subsequent requests made by voice navigation using the second authentication method.

15. The non-transitory, computer-readable medium of claim 14, wherein the user is enrolled in the authentication program for specific spoken commands, wherein the method further comprises determining whether the user is enrolled in the authentication program for the spoken command.

16. The non-transitory, computer-readable medium of claim 13, wherein the instructions, when executed by the one or more processors, further cause the machine to:

receive a second spoken command;

determine an activity authentication level to fulfill the second spoken command;

analyze the characteristics of the second spoken command, the characteristics including at least one of;

pitch, intonation, speed, attenuation, accent, cadence, and volume;

authenticate the user via the second authentication method to an authentication level based on the characteristics of the second spoken command compared to characteristics of the voice of the user speaking the spoken command; and

fulfill the second spoken command when the authentication level meets or exceeds the activity authentication level.

17. A computerized method comprising:

receiving a spoken request and authentication information from a user, wherein the spoken request includes biometric information about the user, wherein the biometric information includes voice, wherein the spoken request requests information or commands an action;

authenticating the user via a first authentication method using the authentication information, wherein the first authentication method uses non-voice biometric credentials to authenticate the user;

after authenticating the user via the first authentication method;

fulfilling the spoken request, and

determining a number and a quality of voice samples; and

enrolling, without requesting enrollment information, the request into a biometric authentication program that uses a second authentication method when the number and the quality of the voice samples exceed a threshold, and wherein enrolling the user into the biometric authentication program comprises creating a biometric profile of the user from the biometric information included in the spoken request.

18. The computerized method of claim 17, further comprising:

receiving a second spoken request;

determining an activity authentication level to fulfill the second spoken command;

analyzing characteristics of the second spoken request, the characteristics including at least one of;

pitch, intonation, speed, attenuation, accent, cadence, and volume;

authenticating the user via the second authentication method to an authentication level based on the characteristics of the second spoken request compared to characteristics of the voice of the user speaking the spoken command; and

fulfilling the second spoken command when the authentication level meets or exceeds the activity authentication level.

19. The computerized method of claim 17, further comprising:

determining whether a spoken request is enrolled in the biometric authentication program for the user.

20. The computerized method of claim 19, further comprising:

determining that the spoken request is not enrolled in the biometric authentication program;

determining that portions of the spoken request are enrolled in the biometric authentication program in separate spoken requests;

combining voice profiles of the separate spoken requests to create a voice profile of the spoken request; and

enrolling the spoken request in the biometric authentication program for the user.