User authentication by combining speaker verification and reverse turing test

• US 20060136219A1
• Filed: 12/03/2004
• Published: 06/22/2006
• Est. Priority Date: 12/03/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Kuansan
• Granted Patent: US 8,255,223 B2
• Time in Patent Office: Days
• US Class Current: 704/273
• CPC Class Codes: G06F 21/31 User authentication
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces

Abstract

Methods and system for authenticating a user are disclosed. The present invention includes accessing a collection of personal information related to the user. The present invention also includes performing an authentication operation that is based on the collection of personal information. The authentication operation incorporates at least one dynamic component and prompts the user to give an audible utterance. The audible utterance is compared to a stored voiceprint.

Claims

1. A method of authenticating a user, the method comprising:

(a) accessing a collection of personal information related to the user;

(b) performing an authentication operation that is based on the collection of personal information and incorporates at least one dynamic component, the authentication operation configured to prompt the user to give an audible utterance;

(c) receiving the audible utterance; and

(d) comparing the audible utterance to a stored voiceprint.

2. The method of claim 1, wherein step (a) comprises accessing the collection of personal information associated with a service program.

3. The method of claim 1, wherein step (a) comprises accessing the collection of personal information from a remote user device.

4. The method of claim 3, wherein the collection of personal information stored in the user device is annotated with semantic types.

5. The method of claim 1, wherein performing an authentication operation that incorporates at least one dynamic component comprises generating a dynamic question that asks about static information from the collection of personal information.

6. The method of claim 1, wherein performing an authentication operation that incorporates at least one dynamic component comprises generating a question that requires manipulation of static information.

7. The method of claim 1, wherein performing an authentication operation that incorporates at least one dynamic component comprises generating a static question that asks about dynamic information from the collection of personal information.

8. The method of claim 1, wherein performing an authentication operation that incorporates at least one dynamic component comprises generating a question that requires identifying an item of information that changes over time.

9. The method of claim 1, wherein receiving an audible utterance comprises receiving an audible answer to a question.

10. The method of claim 9, further comprising denying access to the user if the audible answer is incorrect before comparing to the stored voiceprint.

11. The method of claim 9, further comprising granting access to the user if the audible answer is correct and the audible answer substantially corresponds to the stored voiceprint.

12. The method of claim 1, wherein step (c) comprises receiving over a telephone network an audible utterance.

13. The method of claim 1, wherein step (c) comprises receiving over a computer network an audible utterance.

14. A machine-based system for authenticating a user, the system comprising:

an information retriever configured to access a collection of personal information related to a user;

an authentication module configured to perform an authentication operation by generating an inquiry based on the collection of personal information and incorporating at least one dynamic component; and

a voice verification module configured to process an audible utterance received in response to the inquiry.

15. The machine-based system of claim 14, wherein the information retriever is further configured to access the collection of personal information associated with a service program.

16. The machine-based system of claim 14, wherein the information retriever is further configured to access the collection of personal information associated with a remote user device.

17. The machine-based system of claim 14, wherein the collection of personal information comprises a plurality of data entities.

18. The machine-based system of claim 17, wherein each data entity is annotated with a plurality of semantic types.

19. The machine-based system of claim 18, wherein the authentication module performs an authentication operation based on at least one of the plurality of semantic types.

20. The machine-based system of claim 14, wherein the authentication module comprises a question generator configured to generate a dynamic question that asks about static information from the collection of personal information.

21. The machine-based system of claim 14, wherein the authentication module comprises a question generator configured to generate a question that requires manipulation of static information.

22. The machine based system of claim 14, wherein the authentication module comprises a question generator configured to generate a static question that asks about dynamic information from the collection of personal information.

23. The machine based system of claim 14, wherein the authentication module comprises a question generator configured to generate a question that requires identifying an item of information that changes over time.

24. The machine-based system of claim 14, wherein the authentication module further comprises an answer evaluator configured to receive an audible answer from a user.

25. The machine-based system of claim 24, wherein the answer evaluator is further configured to evaluate the correctness of the audible answer.

26. The machine-based system of claim 14, wherein the voice verification module comprises a speech verifier configured to compare the audible utterance to a stored voiceprint.

27. The machine-based system of claim 26, wherein the stored voiceprint is the user'"'"'s voiceprint.

28. The machine-based system of claim 26, wherein the voice verification module further comprises a voiceprint database, the speech verifier configured to access the voiceprint from the voiceprint database.

29. The machine-based system of claim 14, further comprising a speech recognition system for recognizing the audible utterance.

30. A computer implemented method for authenticating a user, the method comprising:

(a) accessing a collection of personal information related to a user;

(b) generating a question that incorporates at least one dynamic component;

(c) capturing an audible answer to the question; and

(d) comparing the audible answer to a stored voiceprint.

31. The computer-implemented method of claim 30, wherein step (a) comprises accessing the collection of personal information by way of an computer network.

32. The computer-implemented method of claim 30, wherein the collection of personal information is annotated with a plurality of semantic types.

33. The computer-implemented method of claim 30, wherein step (b) comprises generating a question that requires manipulation of static information from the collection of personal information.

34. The computer-implemented method of claim 30, wherein step (b) comprises generating a question that requires identifying an item of information on that changes over time.

35. The computer-implemented method of claim 30, further comprising recognizing the audible answer with a speech recognizing system after the audible answer is captured.

36. The computer-implemented method of claim 30, further comprising evaluating the audible answer for correctness after the audible answer is recognized.

37. The computer-implemented method of claim 36, further comprising denying access to the user before comparing the audible answer to the stored voiceprint if the audible answer is incorrect.

38. The computer-implemented method of claim 36, further comprising granting access to the user if the audible answer is correct and the audible answer at least substantially corresponds to the stored voiceprint.

39. A method of authenticating a user, the method comprising:

transmitting a request for authentication information to the user, the request prompts the user to functionally manipulate an element of personal information;

receiving the authentication information from the user; and

evaluating the authentication information against the personal information.

40. The method of claim 39, wherein functionally manipulating an element of personal information comprises arithmetically manipulating an element of personal information, the personal information corresponding to numerical information.

41. The method of claim 39, wherein functionally manipulating an element of personal information comprises manipulating an element of personal information that corresponds to alphabetical information.

42. The method of claim 39, further comprising:

receiving an audible utterance; and

comparing the audible utterance to a stored voiceprint.

44. The method of claim 41, wherein the of personal information is stored in the user device and annotated with semantic types.

45. The method of claim 44, further comprising:

receiving an audible utterance; and

comparing the audible utterance to a stored voiceprint.

43. A method of authenticating a user, the method comprising:

transmitting a request for authentication information to the user, the request prompts the user to relate personal information corresponding to the user'"'"'s interaction with a user device;

receiving the authentication information from the user; and

evaluating the authentication information against the personal information.