Transparent voice registration and verification method and system

• US 8,406,382 B2
• Filed: 11/09/2011
• Issued: 03/26/2013
• Est. Priority Date: 07/30/2008
• Status: Active Grant

• Current Assignee: AT&T Alex Holdings LLC (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Gilbert, Mazin
• Primary Examiner(s): Gauthier, Gerald
• Application Number:
  US13/292,436
• Publication Number:
  US 20120051525A1
• Time in Patent Office: 503 Days
• Field of Search: 340/5.22, 379/88.01, 379/88.02, 379/88.03, 382/218, 704/219, 704/223, 704/236, 704/249, 704/260, 704/273, 704/275, 705/3, 705/67, 705/75, 713/181, 717/173, 726/7
• US Class Current: 379/88.01
• 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: H04M 2201/36 Memories
• CPC Class Codes: H04M 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 2203/6045 Identity confirmation
• CPC Class Codes: H04M 3/385 using speech signals
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/51 Centralised call answering arrangements requiring operator intervention , e.g. call or contact centers for telemarketing

Abstract

A method includes registering a voice of a party in order to provide voice verification for communications with an entity. A call is received from a party at a voice response system. The party is prompted for information and verbal communication spoken by the party is captured. A voice model associated with the party is created by processing the captured verbal communication spoken by the party and is stored. The identity of the party is verified and a previously stored voice model of the party, registered during a previous call from the party, is updated. The creation of the voice model is imperceptible to the party.

Claims

1. A method comprising:

receiving a call from a party at a voice response system;

capturing verbal communication spoken by the party;

creating a voice model associated with the party, the voice model being created by processing the captured verbal communication spoken by the party;

storing the created voice model in a memory;

verifying the identity of the party; and

updating a previously stored voice model of the party, the previously stored voice model of the party having been registered during a previous call received from the party,

wherein the creating of the voice model is imperceptible to the party.

2. The method according to claim 1, wherein verifying the identity comprises determining voiceprint distinctions between the previously stored voice model of the party and the created voice model.

3. The method according to claim 1, further comprising prompting the party for information when the party requests access to account information.

4. The method according to claim 3, wherein the information comprises identifying information associated with the party.

5. The method according to claim 1, wherein verifying the identity comprises comparing the created voice model with the previously stored voice model of the party and identifying a match between the created voice model and the previously stored voice model of the party.

6. The method according to claim 1, wherein the previously stored voice model of the party was created during a registration in which the party was not notified that the previously stored voice model was being created for voice registration.

7. The method according to claim 1, wherein the previously stored voice model of the party is updated by replacing the previously stored voice model of the party with the created voice model.

8. The method according to claim 1, further comprising replacing the previously stored voice model of the party when it is determined that a distinction exists between the created voice model and the previously stored voice model of the party.

9. The method according to claim 1, further comprising associating the created voice model of the party with the previously stored voice model of the party so that a future voice verification can be performed using both of the created voice model and the previously stored voice model of the party.

10. The method according to claim 1, wherein the capturing comprises isolating at least one spoken word of the party from at least one spoken word of an agent of the entity.

11. A system comprising:

a receiver that receives a call; and

at one least server that prompts a party, captures verbal communication spoken by the party during interaction between the party and an agent, creates a voice model associated with the party from the captured verbal communication, stores the created voice model, verifies the identity of the party, and updates a previously stored voice model of the party, wherein the creation of the voice model is imperceptible to the party.

12. A non-transitory computer readable storage medium that stores a computer program that is executable by a computer to:

receive a call from a party;

prompt the party for information;

capture verbal communication spoken by the party;

create a voice model associated with the party, the voice model being created by processing the captured verbal communication spoken by the party;

store the created voice model;

verify the identity of the party; and

update a previously stored voice model of the party, the previously stored voice model of the party having been registered during a previous call received from the party,

wherein the creation of the voice model is imperceptible to the party.

13. The non-transitory computer readable medium according to claim 12, wherein the computer verifies the identity by determining voiceprint distinctions between the previously stored voice model of the party and the created voice model.

14. The non-transitory computer readable medium according to claim 12, wherein the party is prompted for the information when the party requests access to account information.

15. The non-transitory computer readable medium according to claim 12, wherein the information comprises identifying information associated with the party.

16. The non-transitory computer readable medium according to claim 12, wherein the computer verifies the identify by comparing the created voice model with the previously stored voice model of the party and identifying a match between the created voice model and the previously stored voice model of the party.

17. The non-transitory computer readable medium according to claim 12, wherein the previously stored voice model of the party was created during a registration in which the party was not notified that the previously stored voice model was being created for voice registration.

18. The non-transitory computer readable medium according to claim 12, wherein the computer updates the previously stored voice model of the party by replacing the previously stored voice model of the party with the created voice model.

19. The non-transitory computer readable medium according to claim 12, wherein the computer updates the previously stored voice model of the party by replacing the previously stored voice model of the party when it is determined that a distinction exists between the created voice model and the previously stored voice model of the party.

20. The non-transitory computer readable medium according to claim 12, wherein the computer updates the previously stored voice model of the party by associating the created voice model of the party with the previously stored voice model of the party so that a future voice verification can be performed using both of the created voice model and the previously stored voice model of the party.