Content customization

• US 8,849,676 B2
• Filed: 03/29/2012
• Issued: 09/30/2014
• Est. Priority Date: 03/29/2012
• Status: Active Grant

• Current Assignee: Audible Incorporated (Amazon.com, Inc.)
• Original Assignee: Audible Incorporated (Amazon.com, Inc.)
• Inventors: Goldstein, Douglas S., Arora, Ajay, Hwang, Douglas, Story, Guy A. Jr., Yang, Shirley C.
• Primary Examiner(s): COLUCCI, MICHAEL C
• Application Number:
  US13/434,574
• Publication Number:
  US 20130262127A1
• Time in Patent Office: 915 Days
• Field of Search: 704/278, 704/2, 715/760, 715/203, 709/223, 709/204, 709/203, 707/767, 707/723, 705/319
• US Class Current: 704/278
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser

Abstract

A content processing service may analyze an item of original content and identify several objects, attributes of those objects, and relationships between those objects present in the item of original content. The content processing service may also analyze a source graph, such as a social graph or supplemental graph, and identify several objects, attributes of those objects, and relationships between objects present in the source graph. The content processing service may customize the item of original content by selecting an original object and selecting a source graph object. One or more of the attributes or relationships of the selected original object in the item of original content may be replaced by one or more of the attributes or relationships of the selected source graph object. To customize items of audio content, audio content associated with the source graph object may replace audio content associated with the target graph object.

Claims

1. A system for processing content, the system comprising:

a content data store configured to store an audiobook, the audiobook comprising an audio narration that references a plurality of original objects and includes a recitation of at least one original attribute of each original object;

a narration data store configured to store a plurality of audio clips, each audio clip affiliated with a speaker; and

a computing device in communication with the content data store and the narration data store, the computing device configured to;

access a social graph comprising a plurality of social graph objects;

select a social graph object from the plurality of social graph objects, the selected social graph object comprising at least one social graph attribute;

select an original object; and

replace, in the audio narration of the audiobook, the recitation of at least one original attribute of the selected original object with a replacement recitation of the at least one social graph attribute to form a modified audiobook;

wherein the replacement recitation is generated from one or more audio clips stored in the narration data store.

2. The system for processing content of claim 1, wherein the generated replacement recitation is synthesized from one or more audio clips.

3. The system for processing content of claim 1, wherein the social graph is associated with a consumer of the audiobook.

4. The system for processing content of claim 1, wherein the selected social graph object corresponds to the speaker of one or more audio clips.

5. The system for processing content of claim 1, wherein the selected original object corresponds to the speaker of one or more audio clips.

6. A computer-implemented method for customizing an item of original content comprising an audio narration and a plurality of original objects, the computer-implemented method comprising:

under control of one or more computing devices configured with specific computer executable instructions,

identifying, in the audio narration, at least one original recitation affiliated with a first speaker;

selecting an object from a source graph, wherein the selected object corresponds to a second speaker;

obtaining one or more audio clips affiliated with the second speaker; and

replacing, in the audio narration, the at least one original recitation with at least one replacement recitation generated from one or more audio clips affiliated with the second speaker to form an item of modified content.

7. The computer-implemented method of claim 6, wherein the at least one original recitation comprises the same words as the at least one replacement recitation.

8. The computer-implemented method of claim 6, wherein the at least one replacement recitation is synthesized from the one or more audio clips affiliated with the second speaker.

9. The computer-implemented method of claim 6, wherein:

the source graph comprises a social graph comprising a plurality of social graph objects; and

the object selected from the source graph is a social graph object corresponding to the second speaker.

10. The computer-implemented method of claim 9 further comprising identifying, in the item of original content, a relationship between an original object corresponding to the first speaker and a second original object of the plurality of original objects.

11. The computer-implemented method of claim 10 further comprising identifying, in the social graph, a relationship between the social graph object corresponding to the second speaker and a second social graph object of the plurality of social graph objects.

12. The computer-implemented method of claim 11, wherein the relationship between the original object corresponding to the first speaker and the second original object is identical to the relationship between the social graph object corresponding to the second speaker and the second social graph object.

13. The computer-implemented method of claim 6, wherein:

the source graph comprises a supplemental graph comprising a plurality of supplemental objects; and

the object selected from the source graph is a supplemental object corresponding to the second speaker.

14. The computer-implemented method of claim 13, wherein the supplemental object corresponding to the second speaker is selected based at least in part on a user profile.

15. The computer-implemented method of claim 13, wherein the supplemental object corresponding to the second speaker is selected based at least in part on user input.

16. The computer-implemented method of claim 13, wherein the supplemental object corresponding to the second speaker is selected based at least in part on input received from a human interaction task system.

17. The computer-implemented method of claim 13, wherein the supplemental graph comprises a graph representing at least one of a setting or theme.

18. The computer-implemented method of claim 17, wherein the at least one replacement recitation is modified based at least in part on the setting or theme.

19. The computer-implemented method of claim 6, wherein all recitations affiliated with the first speaker in the audio narration are replaced with replacement recitations generated from the one or more audio clips affiliated with the second speaker.

20. A non-transitory, computer-readable medium having a computer-executable component for processing content, the computer-executable component comprising:

a content processing component configured to;

from an item of original content comprising audio content and a plurality of original objects, select an original object corresponding to a first speaker;

identify, in the audio content, at least one original recitation affiliated with the first speaker;

extract a plurality of objects from a source graph;

select an object from the source graph corresponding to a second speaker;

obtain one or more audio clips affiliated with the second speaker; and

replace, in the audio content, the at least one original recitation with at least one replacement recitation based at least in part on the one or more audio clips affiliated with the second speaker to form an item of modified content.

21. The non-transitory, computer-readable medium of claim 20, wherein the content processing component is further configured to:

identify, in the audio content of the item of modified content, the at least one replacement recitation;

obtain one or more new audio clips affiliated with the second speaker; and

replace, in the audio content of the item of modified content, the at least one replacement recitation with at least one updated recitation based at least in part on the one or more new audio clips.

22. The non-transitory, computer-readable medium of claim 20, wherein the content processing component is further configured to:

identify, in the audio content of the item of modified content, the at least one replacement recitation;

select an object from the source graph corresponding to a third speaker;

obtain one or more audio clips affiliated with the third speaker; and

replace, in the audio content of the item of modified content, the at least one replacement recitation with at least one updated recitation generated from the one or more audio clips affiliated with the third speaker.

23. The non-transitory, computer-readable medium of claim 20, wherein:

the source graph comprises a social graph comprising a plurality of social graph objects; and

the object selected from the source graph is a social graph object corresponding to the second speaker.

24. The non-transitory, computer-readable medium of claim 20, wherein:

the source graph comprises a supplemental graph comprising a plurality of supplemental objects; and

the object selected from the source graph is a supplemental object corresponding to the second speaker.

25. A system for processing content, the system comprising:

a content data store configured to store a plurality of original items of content, each item of original content comprising audio narration and a plurality of original objects, each original object comprising at least one original attribute;

a narration data store comprising one or more audio clips; and

a computing device in communication with the content data store and the narration data store, the computing device configured to;

select an object from a source graph comprising a plurality of objects, the selected object comprising at least one attribute;

access an item of original content;

select an original object; and

replace, in the item of original content, at least one original attribute of the selected original object with at least one attribute of the object selected from the source graph to form an item of modified content.

26. The system for processing content of claim 25, wherein the computing device is further configured to:

identify, in the audio narration, a recitation of the at least one original attribute of the selected original object;

generate, from the one or more audio clips, a recitation of at least one attribute of the object selected from the source graph; and

replace, in the audio narration, the recitation of the at least one original attribute of the selected original object with the generated recitation of the at least one attribute of the selected object selected from the source graph to form a modified audio narration.

27. The system for processing content of claim 26, wherein the computing device is further configured to:

present at least one of a portion of the item of modified content or a portion of the modified audio narration to a user; and

receive, from the user, a desired further modification to at least one of the item of modified content or the modified audio narration.

28. The system for processing content of claim 25, wherein:

the source graph comprises a social graph;

the object selected from the source graph is a social graph object comprising at least one social graph attribute; and

the at least one attribute replacing the at least one original attribute of the selected original object is the at least one social graph attribute.

29. The system for processing content of claim 25, wherein:

the source graph comprises a supplemental graph;

the object selected from the source graph is a supplemental object comprising at least one supplemental attribute; and

the at least one attribute replacing the at least one original attribute of the selected original object is the at least one supplemental attribute.