Bio-Phonetic Multi-Phrase Speaker Identity Verification

• US 20090259470A1
• Filed: 06/24/2009
• Published: 10/15/2009
• Est. Priority Date: 02/13/2003
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Chang, Hisao M.
• Granted Patent: US 9,236,051 B2
• Time in Patent Office: Days
• US Class Current: 704/246
• 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 for verifying an identity of a user comprising:

prompting a user to speak a first dynamic phrase (FDP) comprising at least one dynamically-generated word, the dynamically-generated word generated based on data associated with the user;

extracting at least one voice characteristic from a received FDP utterance and comparing the at least one voice characteristic with a voice profile; and

determining whether to accept a speaker identity based on the comparison of the at least one voice characteristic with the voice profile.

2. The method of claim 1, further comprising:

receiving the FDP utterance.

3. The method of claim 1, wherein the at least one dynamically-generated word is not a numeric word.

4. The method of claim 1, wherein the data associated with the user comprises biographical data associated with the user.

5. The method of claim 1, wherein the data associated with the user comprises geographical data associated with the user.

6. The method of claim 1, wherein the data associated with the user comprises time-sensitive transaction data associated with the user.

7. The method of claim 1, wherein prompting a user to speak a FDP comprises:

transmitting a query to a user, wherein an answer to the query comprises the FDP.

8. The method of claim 1, wherein determining whether to accept the speaker identity comprises:

rejecting the speaker identity when a score based on the comparison of the at least one voice characteristic with the voice profile indicates a low confidence level.

9. The method of claim 8, further comprising:

tagging the FDP as being attempted by an imposter.

10. The method of claim 1, wherein determining whether to accept the speaker identity comprises:

accepting the speaker identity when a score based on the comparison of the at least one voice characteristic with the voice profile indicates a high confidence level.

11. A computer-readable storage medium comprising a set of instructions for verifying an identity of a user, the set of instructions to direct a processor to perform acts of:

prompting a user to speak a first dynamic phrase (FDP) comprising at least one dynamically-generated word, the dynamically-generated word generated based on data associated with the user;

extracting at least one voice characteristic from a received FDP utterance and comparing the at least one voice characteristic with a voice profile; and

determining whether to accept a speaker identity based on the comparison of the at least one voice characteristic with the voice profile.

12. The computer-readable storage medium of claim 11, wherein the data associated with the user comprises biographical data associated with the user.

13. The computer-readable storage medium of claim 11, wherein the data associated with the user comprises geographical data associated with the user.

14. The computer-readable storage medium of claim 11, wherein the data associated with the user comprises time-sensitive transaction data associated with the user.

15. The computer-readable storage medium of claim 11, wherein prompting a user to speak a FDP comprises:

transmitting a query to a user, wherein an answer to the query comprises the FDP.

16. A system for identifying a user, the system comprising:

a voice profile database comprising a plurality of voice profiles; and

a speaker identity verification engine in communication with the voice profile database, the speaker identity verification engine operative to;

prompt a user to speak a first dynamic phrase (FDP) comprising at least one dynamically-generated word, the dynamically-generated word generated based on data associated with the user;

extract at least one voice characteristic from a received FDP utterance and comparing the at least one voice characteristic with a voice profile; and

determine whether to accept a speaker identity based on the comparison of the at least one voice characteristic with the voice profile.

17. The system of claim 16, wherein the data associated with the user comprises biographical data associated with the user.

18. The system of claim 16, wherein the data associated with the user comprises geographical data associated with the user.

19. The system of claim 16, wherein the data associated with the user comprises time-sensitive transaction data associated with the user.

20. The system of claim 16, wherein to prompt the user to speak the FDP, the speaker identity verification engine is further operative to transmit a query to a user, wherein an answer to the query comprises the FDP.