Bio-phonetic multi-phrase speaker identity verification

• US 9,524,719 B2
• Filed: 12/09/2015
• Issued: 12/20/2016
• Est. Priority Date: 02/13/2003
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Chang, Hisao M.
• Primary Examiner(s): Opsasnick, Michael N
• Application Number:
  US14/963,491
• Publication Number:
  US 20160093305A1
• Time in Patent Office: 377 Days
• Field of Search: 704/246, 704/247, 704/250, 704/273, 379/88.01, 379/88.02
• US Class Current: 1/1
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/14 Use of phonemic categorisation or speech recognition prior to speaker recognition or verification
• 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 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

Systems and methods for bio-phonetic multi-phrase speaker identity verification are disclosed. Generally, a speaker identity verification engine generates a dynamic phrase including at least one dynamically-generated word. The speaker identity verification engine prompts a user to speak the dynamic phrase and receives a dynamic phrase utterance. The speaker identity verification engine extracts at least one voice characteristic from the dynamic phrase utterance and compares the at least one voice characteristic with a voice profile the generate a score. The speaker identity verification engine then determines whether to accept a speaker identity claim based on the score.

Claims

1. A method comprising:

receiving private, time-sensitive data associated with activities of an asserted speaker identity, wherein a speaker claims to have the asserted speaker identity;

prompting the speaker to speak an alpha-numeric word, wherein the alpha-numeric word is selected based on the private time-sensitive data;

extracting a voice feature from a response to the prompting;

comparing, via a processor, the voice feature with a voice profile of the asserted speaker identity, to yield a comparison; and

verifying the speaker as the asserted speaker identity when the comparison exceeds a threshold similarity between the response and the voice profile.

2. The method of claim 1, further comprising receiving the response.

3. The method of claim 1, wherein the private time-sensitive data further comprises biographical data of the asserted speaker identity, wherein the biographical data comprises a place of birth of the asserted speaker identity.

4. The method of claim 1, wherein the private time-sensitive data further comprises geographical data of the asserted speaker identity, wherein the geographical data comprises a home address of the asserted speaker identity.

5. The method of claim 1, wherein the private time-sensitive data further comprises email data of the asserted speaker identity.

6. The method of claim 1, wherein prompting the speaker to speak the alpha-numeric word further comprises:

displaying the alpha-numeric word for the speaker to articulate.

7. The method of claim 1, wherein verifying the speaker as the asserted speaker identity further comprises:

rejecting the speaker as the asserted speaker identity when a score of the comparison indicates a low confidence level.

8. The method of claim 7, further comprising tagging the alpha-numeric word as being attempted by an imposter.

9. A system comprising:

a processor; and

a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising;

receiving private, time-sensitive data associated with activities of an asserted speaker identity, wherein a speaker claims to have the asserted speaker identity;

prompting the speaker to speak an alpha-numeric word, wherein the alpha-numeric word is selected based on the private time-sensitive data;

extracting a voice feature from a response to the prompting;

comparing the voice feature with a voice profile of the asserted speaker identity, to yield a comparison; and

verifying the speaker as the asserted speaker identity when the comparison exceeds a threshold similarity between the response and the voice profile.

10. The system of claim 9, the computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising receiving the response.

11. The system of claim 9, wherein the private time-sensitive data further comprises biographical data of the asserted speaker identity, wherein the biographical data comprises a place of birth of the asserted speaker identity.

12. The system of claim 9, wherein the private time-sensitive data further comprises geographical data of the asserted speaker identity, wherein the geographical data comprises a home address of the asserted speaker identity.

13. The system of claim 9, wherein the private time-sensitive data further comprises email data of the asserted speaker identity.

14. The system of claim 9, wherein prompting the speaker to speak the alpha-numeric word further comprises:

displaying the alpha-numeric word for the speaker to articulate.

15. The system of claim 9, wherein verifying the speaker as the asserted speaker identity further comprises:

rejecting the speaker as the asserted speaker identity when a score of the comparison indicates a low confidence level.

16. The system of claim 15, the computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising tagging the alpha-numeric word as being attempted by an imposter.

17. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:

receiving private, time-sensitive data associated with activities of asserted speaker identity, wherein a speaker claims to have the asserted speaker identity;

prompting the speaker to speak an alpha-numeric word, wherein the alpha-numeric word is selected based on the private time-sensitive data;

extracting a voice feature from a response to the prompting;

comparing the voice feature with a voice profile of the asserted speaker identity, to yield a comparison; and

verifying the speaker as the asserted speaker identity when the comparison exceeds a threshold similarity between the response and the voice profile.

18. The computer-readable storage device of claim 17, having additional instructions stored which, when executed by the computing device, cause the computing device to perform operations comprising:

receiving the response.

19. The computer-readable storage device of claim 17, wherein the private time-sensitive data further comprises biographical data of the asserted speaker identity, wherein the biographical data comprises a place of birth of the asserted speaker identity.

20. The computer-readable storage device of claim 17, wherein the private time-sensitive data further comprises geographical data of the asserted speaker identity, wherein the geographical data comprises a home address of the asserted speaker identity.