Generation of voice profiles

• US 8,380,504 B1
• Filed: 05/06/2010
• Issued: 02/19/2013
• Est. Priority Date: 05/06/2010
• Status: Active Grant

• Current Assignee: T-Mobile Innovations LLC (Deutsche Telekom AG)
• Original Assignee: Sprint Communications Company LP (Deutsche Telekom AG)
• Inventors: Peden, Mark Douglas, Youngs, Simon, Koller, Gary Duane, Jethwa, Piyush
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US12/775,270
• Time in Patent Office: 1,020 Days
• Field of Search: 704/246, 704/247, 704/249, 704/250, 704/233, 704/270.1
• US Class Current: 704/246
• CPC Class Codes: G10L 13/086 Detection of language
• CPC Class Codes: G10L 15/005 Language recognition

Abstract

Embodiments of the present invention provide systems, methods, and computer-readable media for generating a voice characteristic profile based on detected sound components. In embodiments, a call is initiated between a first caller and a second caller. Information communicated during the call is monitored to determine that sound components have been spoken by the first caller. The sound components are determined to be associated with a language dialect. Further, the sound components are stored in association with the first caller. In particular, the sound components are stored in association with the first caller in a voice characteristic profile of the first caller.

Claims

1. Non-transitory computer-readable media having computer-executable instructions embodied thereon that, when executed by one or more computing devices, cause the one or more computing devices to perform a method of generating and using voice characteristic profiles, the method comprising:

receiving information communicated by a first caller during a phone conversation between the first caller and a second caller via a first caller'"'"'s device and a second caller'"'"'s device;

analyzing the information communicated by the first caller;

identifying a set of sound components from the information communicated by the first caller;

matching a subset of the set of sound components with a threshold number of sound components associated with at least one voice characteristic;

determining a first voice characteristic based on matching the subset of the set of sound components with the threshold number of sound components associated with the first voice characteristic;

generating a voice characteristic profile of the first caller, the voice characteristic profile comprising the first voice characteristic; and

using the voice characteristic profile of the first caller to modify information communicated by the second caller to match the first voice characteristic associated with the first caller to enhance a communication between the first caller'"'"'s device and the second caller'"'"'s device.

2. The computer-readable media of claim 1, further comprising:

determining a second voice characteristic based on matching the subset of the set of sound components with a threshold number of sound components associated with the second voice characteristic.

3. The computer-readable media of claim 2, wherein the first voice characteristic is a primary dialect and the second voice characteristic is a secondary dialect.

4. The computer-readable media of claim 3, wherein an interpretation of a word consistent with both the primary and secondary dialects is used when a conflict exists with a generic dialect.

5. The computer-readable media of claim 3, wherein an interpretation of a word consistent with the primary dialect is used when a conflict exists with a secondary dialect.

6. A method of generating and using a voice characteristic profile associated with a first caller, comprising:

receiving, at a computing device, information communicated during a phone conversation between a first caller and a second caller via a first caller'"'"'s device and a second caller'"'"'s device;

analyzing information communicated by the first caller;

identifying a set of sound components that originate from the first caller;

matching a subset of the set of sound components with pre-defined sound components associated with a characteristic;

determining a first characteristic based on matching the subset of the set of sound components with the pre-defined sound components;

generating a first voice characteristic profile that is associated with the first caller, the first voice characteristic profile comprising the first characteristic; and

using the first voice characteristic profile associated with the first caller to modify information communicated by the second caller to match the first voice characteristic associated with the first caller to enhance a communication between the first caller'"'"'s device and the second caller'"'"'s device.

7. The method of claim 6, wherein the sound components are provided by the first caller.

8. The method of claim 6, wherein the sound components are provided by a characteristic recognition component.

9. The method of claim 6, wherein matching a subset of the set of sound components with pre-defined sound components associated with a characteristic comprises matching a threshold number of pre-defined sound components associated with the characteristic.

10. The method of claim 9, wherein the subset of the set of sound components match a threshold number of pre-defined sound components associated with a plurality of characteristics.

11. The method of claim 10, wherein the first characteristic of the plurality of characteristics is prioritized above the remaining characteristics of the plurality of characteristics.

12. The method of claim 11, wherein the first characteristic is prioritized based on the subset of the set of sound components matching the greatest number of pre-defined sound components associated with the first characteristic.