SELECTING CONTENT PORTIONS FOR ALIGNMENT

• US 20140040713A1
• Filed: 09/05/2012
• Published: 02/06/2014
• Est. Priority Date: 08/02/2012
• Status: Active Grant

• Current Assignee: Audible Incorporated (Amazon.com, Inc.)
• Original Assignee: Guy A. Story, Jr.
• Inventors: Dzik, Steven C., Story, Guy A. Jr.
• Granted Patent: US 10,109,278 B2
• Time in Patent Office: Days
• US Class Current: 715/203
• CPC Class Codes: G06F 16/4393 Multimedia presentations, e.g. slide shows, multimedia albums
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H05K 999/99 dummy group

Abstract

Generally described, a content alignment service may generate content synchronization information to facilitate the synchronous presentation of corresponding audio content and textual content. In some embodiments, portions of body text (as opposed to front matter, such as a table of contents; or back matter, such as an index) in the textual content are identified and synchronized with corresponding audio content. In one example application, an audiobook may be synchronized with an electronic book. As the body text portions of the electronic book are consumed, corresponding words of the audiobook may be audibly presented.

Claims

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

an electronic data store configured to store;

an electronic book comprising a plurality of paragraphs; and

an audiobook that is a companion to the electronic book; and

a computing device in communication with the electronic data store, the computing device configured to;

generate a textual transcription of the audiobook by applying a speech-to-text recognition routine on the audiobook;

identify a portion of the textual transcription that corresponds to at least one paragraph of the electronic book;

generate content synchronization information, wherein the content synchronization information facilitates the synchronous presentation of the at least one paragraph and a portion of the audiobook from which the corresponding portion of the textual transcription was generated.

2. The system of claim 1, wherein the at least one paragraph is a body text paragraph.

3. The system of claim 1, wherein the computing device is further configured to provide the content synchronization information to a separate computing device.

4. The system of claim 1, wherein the computing device is further configured to synchronously present the at least one paragraph and the portion of the audiobook from which the corresponding portion of the textual transcription was generated.

5. A computer-implemented method for aligning content, the computer-implemented method comprising:

as implemented by one or more computing devices configured with specific computer-executable instructions,

obtaining a textual transcription of an item of content comprising audio content;

identifying a plurality of body text portions of a companion item of textual content;

identifying a portion of the textual transcription that corresponds to least one body text portion; and

generating content synchronization information, wherein the content synchronization information facilitates a synchronous presentation of the at least one body text portion and a portion of the audio content that corresponds to the portion of the textual transcription.

6. The computer-implemented method of claim 5, wherein obtaining the textual transcription comprises generating the textual transcription from the audio content.

7. The computer-implemented method of claim 5, wherein identifying the plurality of body text portions comprises:

computing a correlation measure for each block of the companion item of textual content with respect to the textual transcription, each block comprising one or more portions of the companion item of textual content;

wherein a portion of the companion item of textual content is identified as a body text portion if the correlation measure for the block to which that portion belongs satisfies a threshold.

8. The computer-implemented method of claim 5, wherein identifying the plurality of body text portions comprises:

identifying any front matter portions in the companion item of textual content; and

identifying any back matter portions in the companion item of textual content;

wherein all portions of the companion item of textual content that are not front matter portions and that are not back matter portions are identified as body text portions.

9. The computer-implemented method of claim 5, wherein each body text portion of the plurality of body text portions comprises at least one of a word, a phrase, a sentence, a paragraph, and a line of dialogue.

10. The computer-implemented method of claim 5, wherein the companion item of textual content is an electronic book.

11. The computer-implemented method of claim 5, wherein the item of content comprising audio content is an audiobook.

12. The computer-implemented method of claim 5, wherein the item of content comprising audio content further comprises video content.

13. A system for aligning content, the system comprising:

an electronic data store configured to store;

a transcription of an item of content comprising audio content; and

a companion item of textual content; and

a computing device in communication with the electronic data store, the computing device being configured to;

identify at least one body text portion in the companion item of textual content;

identify, in the transcription, a portion of the transcription that substantially corresponds to the at least one body text portion; and

generate content synchronization information, wherein the content synchronization information facilitates the synchronous presentation of the at least one body text portion and a portion of the audio content that corresponds to the identified portion of the transcription.

14. The system of claim 13, wherein the at least one body text portion is identified based at least in part on metadata in the companion item of textual content.

15. The system of claim 13, wherein the at least one body text portion is identified based at least in part on input received from a human interaction task system.

16. The system of claim 13, wherein the at least one body text portion comprises at least one of a word, sentence, or paragraph.

17. The system of claim 13, wherein the portion of the transcription substantially corresponds to the body text portion at least a threshold percentage of words of the portion of the transcription correspond to words of the at least one body text portion.

18. The system of claim 17, wherein a word of the portion of the transcription corresponds to a word of the body text portion if the word of the portion of the transcription substantially matches and chronologically corresponds to the word of the body text portion.

19. A non-transitory computer-readable medium having a computer-executable module, the computer-executable module comprising:

a content alignment module configured to;

select one or words in an item of textual content;

identify, in a companion item of content, a first timestamp at which a first word of the one or more words is audibly presented; and

generate content synchronization information, wherein the content synchronization information comprises instructions to cause a computing device to present the companion item of content starting from the first timestamp as the selected one or more words are presented by the computing device.

20. The non-transitory computer-readable medium of claim 19, wherein the one or more words are included in a body text paragraph.

21. The non-transitory computer-readable medium of claim 20, wherein the body text paragraph occurs before any other body text paragraph in the item of textual content.

22. The non-transitory computer-readable medium of claim 19, wherein the item of textual content is an electronic book.

23. The non-transitory computer-readable medium of claim 19, wherein the companion item of content is an audiobook.

24. The non-transitory computer-readable medium of claim 19, wherein:

the content alignment module is further configured to identify, in the companion item of content, a second timestamp at which a last word of the one or more words is audibly presented; and

the content synchronization information further comprises instructions to cause the computing device to cease presenting the companion item of content when the second timestamp is reached in the companion item of content.

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

the content alignment module is further configured to;

select another one or more words in the item of textual content; and

identify, in the companion item of content, a third timestamp at which a first word of the another one or more other words is audibly presented; and

the content synchronization information further comprises instructions to cause a computing device to present the companion item of content starting from the third timestamp as the another one or more new words are presented by the computing device.