Granted Patent B2

US 8,077,836 · App. 12/182,182 · Granted Dec 13, 2011

Abstract

Claims (41)

1. A method for registering a voice of a party in order to provide voice verification for communications with a service center, the method comprising:

capturing verbal communication spoken by the party during interaction, initiated by the party, between the party and an agent of the service center;

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

storing the voice model to provide voice verification of the party during a subsequent communication with the service center,

wherein the registering is conducted without having notified the party of the registration.

2. The method according to claim 1 , further comprising:

prompting a requester for information;

capturing speech of the requester;

processing the captured speech of the requester;

comparing the processed speech of the requester with the voice model; and

determining whether processed speech of the requester matches the voice model, the determination comprising whether the requester is the party.

3. The method according to claim 2 , further comprising:

creating a voice model associated with the requester, the voice model associated with the requester being obtained by processing the captured speech of the requester.

4. The method according to claim 2 , further comprising comparing the processed speech of the requester with a second voice model associated with the party.

5. The method according to claim 2 , further comprising forwarding the requester for further screening when the processed speech of the requester does not match the voice model and servicing the party when the processed speech of the requester matches the voice model.

6. The method according to claim 2 , further comprising updating the voice model associated with the party if it is determined that the requester is the party associated with the voice model, the updating comprising one of modifying the voice model and replacing the voice model.

7. The method according to claim 1 , wherein the capturing the verbal communication comprises isolating spoken word of the party from spoken word of the agent.

8. The method according to claim 6 , wherein at least one subsequent communication from the party is originated from a location distinct from the origin of the interaction initiated by the party.

9. The method according to claim 1 , wherein creating the voice model comprises using one of a Gaussian mixture model, hidden Markov model, machine learning algorithm, maximum entropy model, and support vector machine model.

10. The method according to claim 1 , wherein voice models from a plurality of callers are collected to create a cohort model.

11. The method according to claim 1 , wherein the interaction comprises a voice over internet protocol communication.

12. The method according to claim 1 , wherein the interaction comprises a wireless network communication.

13. The method according to claim 1 , wherein the interaction comprises a landline communication.

14. A system for providing voice registration of a voice of a party to provide voice verification for communications with a service center, the system comprising:

at least one processor configured to capture verbal communication spoken by the party during interaction between the party and an agent of the service center, create a voice model associated with the party from the captured verbal communication, and store the voice model to provide voice verification of the party during a subsequent communication with the service center, wherein the registration is conducted without having notified the party of the registration.

15. The system according to claim 14 , further comprising at least one processor configured to compare captured and processed speech of a requester with the stored voice model and to determine whether the requester is the party.

16. The system according to claim 15 , wherein the processor updates the voice model of the party after a subsequent communication with the party.

17. A non-transitory computer readable medium storing a computer program that registers a voice of a party in order to provide voice verification for communications with a service center, comprising:

a capturing code segment that captures verbal communication spoken by the party during interaction, initiated by the party, between the party and an agent of the service center;

a creating code segment that creates a voice model associated with the party, the voice model being obtained by processing the captured verbal communication spoken by the party; and

a storing code segment that stores the voice model to provide voice verification of the party during a subsequent telephone call,

wherein the registering is conducted without having notified the party of the registration.

18. The non-transitory computer readable medium according to claim 17 , further comprising:

a prompting code segment that prompts a requester for information;

a second capturing code segment that captures speech of the requester;

a processing code segment that processes the captured speech of the requester;

a comparing code segment that compares the processed speech of the requester with the voice model; and

a determining code segment that determines whether processed speech of the requester matches the voice model, the determination comprising whether the requester is the party.

19. The non-transitory computer readable medium according to claim 18 , further comprising:

an updating code segment that updates the voice model associated with the party if it is determined that the requester is the party associated with the voice model.

20. The non-transitory computer readable medium according to claim 19 , wherein at least one subsequent communication from the party is originated from a location distinct from the origin of the interaction initiated by the party.

Assignments (16)