Method and system for speech recognition using social networks

• US 9,117,448 B2
• Filed: 07/27/2009
• Issued: 08/25/2015
• Est. Priority Date: 07/27/2009
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Wu, Sung Fong Solomon, Tong, Aaron, Lee, Sam C.
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US12/509,797
• Publication Number:
  US 20110022388A1
• Time in Patent Office: 2,220 Days
• Field of Search: 704/246
• US Class Current: 1/1
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces

Abstract

In an example embodiment, there is disclosed an apparatus comprising an audio interface configured to receive an audio signal, a data interface is configured to communicate with at least one social graph, and logic is coupled to the audio interface and the data interface. The logic is configured to identify a calling party. The logic is further configured to acquire data representative of a called party from the audio signal. The logic is configured to initiate a search of the at least one social graph for the data representative of the called party to identify the called party responsive to acquiring the data representative of the called party.

Claims

1. An apparatus, comprising:

an audio input configured to receive an audio signal representative of a voice input of an associated calling party;

the audio input further configured to generate audio data corresponding to the received audio signal;

a data interface configured to communicate with one or more associated social graphs, in accordance with an Application Programming Interface (API) corresponding thereto, via an associated internetworking system; and

logic coupled with the audio input and the data interface;

wherein the logic is configured to identify the calling party and a plurality of social graphs associated with the calling party;

wherein the logic is configured to acquire data representative of a called party from the audio data, the data representative of the called party indicating a relationship between the calling party and the called party;

wherein the logic converts the data indicating the relationship between the calling party and called party to a form that is compatible with one or more of the plurality of social graphs;

wherein the logic initiates a session with the plurality of social graphs in accordance with an identity of the calling party;

wherein the logic is configured to initiate, in accordance with a predetermined priority, a prioritized search of the plurality of social graphs associated with the calling party for the data representative of the called party to identify the called party responsive to acquiring the data representative of the called party, wherein the predetermined priority comprises one or more of a time of day, a social characteristic of social graphs of the plurality of social graphs, a business characteristic of social graphs of the plurality of social graphs, a favorite preference of the associated calling party, or a frequency of use by the associated calling party of social graphs of the plurality of social graphs.

2. The apparatus set forth in claim 1, wherein the logic is further configured to place a call to the called party.

3. The apparatus set forth in claim 1, wherein the audio data comprises data representative of one of a group consisting of a name and a relationship.

4. The apparatus set forth in claim 1, wherein the logic is configured to employ one of the group consisting of the Application Programming Interface and a markup language to retrieve the data representative of the called party indicating the relationship between the calling and the called party.

5. The apparatus set forth in claim 1, further comprising a user interface coupled with the logic;

wherein the logic is further configured to identify a plurality of parties; and

wherein the user interface is configured to;

present the plurality of parties for selecting the called party; and

receive a selection of the called party.

6. The apparatus set forth in claim 1, wherein the logic is further configured to access a database to select a social graph from the plurality of social graphs to search for the data representative of the called party to identify the called party.

7. The apparatus set forth in claim 1, wherein the logic is further configured to use one of the group consisting of the calling party'"'"'s phone number and credentials supplied by the calling party to select a social graph from the plurality of social graphs to search for the data representative of the called party to identify the called party.

8. The apparatus set forth in claim 1, wherein the logic is further configured to search the plurality of social graphs for the data representative of the called party to identify the called party responsive to acquiring the data representative of the called party.

9. A method comprising:

receiving an audio signal representative of a voice input of an associated calling party;

generating audio data corresponding to the received audio signal;

identifying the associated calling party and a plurality of associated social graphs associated with the calling party;

acquiring data representative of a called party from the audio data;

searching, by a processor, for the called party;

obtaining, by the processor, a plurality of matches for the called party;

initiating a session with the plurality of associated social graphs in accordance with one or more Application Programming Interfaces (APIs) associated with the plurality of social graphs and the identity of the calling party via an associated internetworking system;

searching by the processor, in accordance with a predetermined priority, the plurality of social graphs for the data representative of the called party responsive to acquiring the data representative of the called party and obtaining a plurality of matches for the called party, wherein the predetermined priority comprises one or more of a time of day, a social characteristic of the plurality of social graphs, a business characteristic of the plurality of social graphs, a preference of the associated calling party, or a frequency of use by the associated calling party of social graphs of the plurality of social graphs; and

eliminating, by the processor, a called party from the plurality of matches for the called party that are not found in the at least one social graph.

10. The method set forth in claim 9, further comprising placing a call to the called party.

11. The method set forth in claim 9, wherein the audio data comprises one of the group consisting of a name and a relationship.

12. The method set forth in claim 9, further comprising retrieving the calling party'"'"'s relationships from the at least one social graph by employing one of the group consisting of an Application Programming Interface and a markup language.

13. The method set forth in claim 9, further comprising:

identifying a plurality of parties;

presenting the plurality of parties for selecting the called party; and

receiving a selection of the called party.

14. The method set forth in claim 9, further comprising accessing a database to select a social graph from a plurality of social graphs to search for the data representative of the called party to identify the called party.

15. The method set forth in claim 9, further comprising using one of the group consisting of the calling party'"'"'s phone number and credentials supplied by the calling party to select a social graph from the plurality of social graphs to search for the data representative of the called party to identify the called party.

16. The method set forth in claim 9, wherein the data representative of the called party is data indicating a relationship between the calling party and the called party, and further comprising:

searching the plurality of social graphs for the data representative of the called party to identify the called party responsive to acquiring the data representative of the called party.

17. An apparatus comprising:

an audio input configured to receive an audio signal representative of a voice input of a calling party;

an audio data generator configured to generate audio data from the received audio signal;

a data interface configured to communicate with a plurality of associated social graphs in accordance with an Application Programming Interface (API) associated therewith;

a calling party module configured to determine an identity of the calling party;

a data retrieval module coupled with the calling party module and the data interface, the data retrieval module being configured to select a set of associated social graphs to be searched from the plurality of associated social graphs associated with the identified calling party;

an audio recognition module coupled with the audio input, the audio recognition module being configured to acquire data representative of a called party from the audio data;

a session module coupled with the data retrieval module, the session module being configured to initiate a session with the selected set of social graphs in accordance with the identity of the calling party via an associated internetworking system; and

a called party module, coupled with the data retrieval module and the audio recognition module, the called party module being configured to identify the called party by searching in accordance, with a predetermined priority, the selected set of social graphs for the data representative of the called party, wherein the predetermined priority comprises one or more of a time of day, a social characteristic of graphs of the plurality of social graphs, a business characteristic of graphs of the plurality of social graphs, a preference of the associated calling party, or a frequency of use by the associated calling party of graphs of the plurality of social graphs.

18. The apparatus set forth in claim 17, wherein the audio data comprises data representative of one of the group consisting of a name and a relationship.