Methods and systems for generating automated tags for video files

• US 8,396,878 B2
• Filed: 09/26/2011
• Issued: 03/12/2013
• Est. Priority Date: 09/22/2006
• Status: Active Grant

• Current Assignee: Limelight Networks, Inc.
• Original Assignee: Limelight Networks, Inc.
• Inventors: Acharya, Soam
• Primary Examiner(s): Chbouki, Tarek
• Application Number:
  US13/245,840
• Publication Number:
  US 20120278337A1
• Time in Patent Office: 533 Days
• Field of Search: None
• US Class Current: 707/749
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/739 in form of a video summary, e.g. the video summary being a video sequence, a composite still image or having synthesized frames
• CPC Class Codes: G06F 16/78 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data

Abstract

Techniques for generating automated tags for a video file are described. The method includes receiving one or more manually generated tags associated with a video file, based at least in part on the one or more manually entered tags, determining a preliminary category for the video file, and based on the preliminary category, generating a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words. The method further includes generating an ontology of the plurality of words based on the targeted transcript, ranking the plurality of words in the ontology based on a plurality of scoring factors, and based on the ranking of the plurality of words, generating one or more automated tags associated with the video file.

Claims

1. A computer-implemented method of generating automated tags for a video file, the method comprising:

receiving one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determining a preliminary category for the video file;

based on the preliminary category, generating a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generating an ontology of the plurality of words based on the targeted transcript;

ranking the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generating one or more automated tags associated with the video file; and

generating a heat map for the video file, wherein the heat map comprises a graphical display which indicates offset locations of words within the video file with the highest rankings,

wherein the plurality of scoring factors consists of two or more of;

distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

2. The computer-implemented method of generating automated tags for a video file as in claim 1, wherein the plurality of scoring factors further includes one or more of:

frequency of words and proximity of words relative to other words.

3. The computer-implemented method of generating automated tags for the video file as in claim 1, further comprising:

determining if the rankings for each of the plurality of words exceed a threshold ranking value; and

excluding any of the plurality of words that have a ranking value lower than the threshold value.

4. The computer-implemented method of generating automated tags for the video file as in claim 1, further comprising determining a score for each of the plurality of words, wherein the score includes word frequency, word distribution, and word variety.

5. The computer-implemented method of generating automated tags for the video file as in claim 1, further comprising cross-referencing words with the plurality of words to determine correlations between words or to construct phrases, wherein the cross-referencing of the word or words is configured to increase the ranking of the word or words.

6. The computer-implemented method of generating automated tags for the video file as in claim 1, further comprising:

establishing a top concepts threshold value;

determining that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associating information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file.

7. The computer-implemented method of generating automated tags for the video file as in claim 1, wherein the ontology is configured to provide categorical models of the plurality of words.

8. The computer-implemented method of generating automated tags for the video file as in claim 1, wherein the video file includes an associated audio file, wherein the targeted transcript comprise the plurality of words extracted from the audio file, and wherein each of the plurality of words has an associated offset value which designates the occurrence position of each word within the video file.

9. The computer-implemented method of generating automated tags for the video file as in claim 1, further comprising:

receiving a second video file;

receiving one or more manually generated tags associated with the second video file;

based at least in part on the one or more manually entered tags associated with the second video file, determining a preliminary category for the second video file;

based on the preliminary category of the second video file, generating a targeted transcript of the second video file, wherein the targeted transcript of the second video file includes a second plurality of words;

generating an ontology of the second plurality of words based on the targeted transcript of the second video file;

ranking the plurality of words and the second plurality of words based on the plurality of scoring factors; and

based on the ranking of the plurality of words and the second plurality of words, generating one or more automated tags associated with both the video file and the second video file.

10. A system for generating automated tags for a video file, the system comprising:

a storage memory; and

a processor in communication with the storage memory, wherein the storage memory includes sets of instructions which, when executed by the processor, cause the processor to;

receive one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determine a preliminary category for the video file;

based on the preliminary category, generate a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generate an ontology of the plurality of words based on the targeted transcript;

rank the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generate one or more automated tags associated with the video file; and

generate a heat map for the video file, wherein the heat map comprises a graphical display which indicates offset locations of words within the video file with the highest rankings,

wherein the plurality of scoring factors consists of two or more of;

frequency of words, distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

11. The system of claim 10, wherein the sets of instructions when further executed by the processor cause the processor to cross-reference words with the plurality of words to determine correlations between words or to construct phrases, wherein the cross-referencing of the word or words is configured to increase the ranking of the word or words.

12. The system of claim 10, wherein the sets of instructions when further executed by the processor cause the processor to:

establish a top concepts threshold value;

determine that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associate information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file.

13. The system of claim 10, wherein the ontology is configured to provide categorical models of the plurality of words.

14. The system of claim 10, wherein the sets of instructions when further executed by the processor cause the processor to:

receive a second video file;

receive one or more manually generated tags associated with the second video file;

based at least in part on the one or more manually entered tags associated with the second video file, determine a preliminary category for the second video file;

based on the preliminary category of the second video file, generate a targeted transcript of the second video file, wherein the targeted transcript of the second video file includes a second plurality of words;

generate an ontology of the second plurality of words based on the targeted transcript of the second video file;

rank the plurality of words and the second plurality of words based on the plurality of scoring factors; and

based on the ranking of the plurality of words and the second plurality of words, generate one or more automated tags associated with both the video file and the second video file.

15. A non-transitory computer-readable medium having sets of instructions stored thereon which, when executed by a computer, cause the computer to:

receive one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determine a preliminary category for the video file;

based on the preliminary category, generate a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generate an ontology of the plurality of words based on the targeted transcript;

rank the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generate one or more automated tags associated with the video file; and

generate a heat map for the video file, wherein the heat map comprises a graphical display which indicates offset locations of words within the video file with the highest rankings,

wherein the plurality of scoring factors consists of two or more of;

proximity of words relative to other words, distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

16. The computer-readable medium of claim 15, wherein the sets of instructions when further executed by the computer cause the computer to:

generate a web page which includes an embedded widget;

based on the one or more automated tags searching web content to return a list of videos, blogs, audio, and web pages that match the one or more automated tags; and

providing within the widget a view of each of the returned results.

17. The non-transitory computer-readable medium of claim 15, wherein the sets of instructions when further executed by the computer cause the computer to:

determine if the rankings for each of the plurality of words exceed a threshold ranking value; and

exclude any of the plurality of words that have a ranking value lower than the threshold value.

18. The non-transitory computer-readable medium of claim 15, wherein the sets of instructions when further executed by the computer cause the computer to determine a score for each of the plurality of words, wherein the score includes word frequency, word distribution, and word variety.

19. A computer-implemented method of generating automated tags for a video file, the method comprising:

receiving one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determining a preliminary category for the video file;

based on the preliminary category, generating a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generating an ontology of the plurality of words based on the targeted transcript;

ranking the plurality of words in the ontology based on a plurality of scoring factors; and

based on the ranking of the plurality of words, generating one or more automated tags associated with the video file;

establishing a top concepts threshold value;

determining that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associating information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file,

wherein the video file includes an associated audio file,

wherein the targeted transcript comprise the plurality of words extracted from the audio file, and

wherein each of the plurality of words has an associated offset value which designates the occurrence position of each word within the video file.

20. A computer-implemented method of generating automated tags for a video file, the method comprising:

receiving one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determining a preliminary category for the video file;

based on the preliminary category, generating a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generating an ontology of the plurality of words based on the targeted transcript;

ranking the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generating one or more automated tags associated with the video file;

establishing a top concepts threshold value;

determining that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associating information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file,

wherein the plurality of scoring factors consists of two or more of;

distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

21. The computer-implemented method of generating automated tags for a video file as in claim 20, wherein the plurality of scoring factors further includes one or more of:

frequency of words and proximity of words relative to other words.

22. The computer-implemented method of generating automated tags for the video file as in claim 20, further comprising:

determining if the rankings for each of the plurality of words exceed a threshold ranking value; and

excluding any of the plurality of words that have a ranking value lower than the threshold value.

23. The computer-implemented method of generating automated tags for the video file as in claim 20, further comprising determining a score for each of the plurality of words, wherein the score includes word frequency, word distribution, and word variety.

24. The computer-implemented method of generating automated tags for the video file as in claim 20, further comprising cross-referencing words with the plurality of words to determine correlations between words or to construct phrases, wherein the cross-referencing of the word or words is configured to increase the ranking of the word or words.

25. The computer-implemented method of generating automated tags for the video file as in claim 20, wherein the ontology is configured to provide categorical models of the plurality of words.

26. The computer-implemented method of generating automated tags for the video file as in claim 20, wherein the video file includes an associated audio file, wherein the targeted transcript comprise the plurality of words extracted from the audio file, and wherein each of the plurality of words has an associated offset value which designates the occurrence position of each word within the video file.

27. The computer-implemented method of generating automated tags for the video file as in claim 20, further comprising:

receiving a second video file;

receiving one or more manually generated tags associated with the second video file;

based at least in part on the one or more manually entered tags associated with the second video file, determining a preliminary category for the second video file;

based on the preliminary category of the second video file, generating a targeted transcript of the second video file, wherein the targeted transcript of the second video file includes a second plurality of words;

generating an ontology of the second plurality of words based on the targeted transcript of the second video file;

ranking the plurality of words and the second plurality of words based on the plurality of scoring factors; and

based on the ranking of the plurality of words and the second plurality of words, generating one or more automated tags associated with both the video file and the second video file.

28. A system for generating automated tags for a video file, the system comprising:

a storage memory; and

a processor in communication with the storage memory, wherein the storage memory includes sets of instructions which, when executed by the processor, cause the processor to;

receive one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determine a preliminary category for the video file;

based on the preliminary category, generate a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generate an ontology of the plurality of words based on the targeted transcript;

rank the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generate one or more automated tags associated with the video file;

establish a top concepts threshold value;

determine that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associate information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file,

wherein the plurality of scoring factors consists of two or more of;

frequency of words, distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

29. The system of claim 28, wherein the sets of instructions when further executed by the processor cause the processor to cross-reference words with the plurality of words to determine correlations between words or to construct phrases, wherein the cross-referencing of the word or words is configured to increase the ranking of the word or words.

30. The system of claim 28, wherein the ontology is configured to provide categorical models of the plurality of words.

31. The system of claim 28, wherein the sets of instructions when further executed by the processor cause the processor to:

receive a second video file;

receive one or more manually generated tags associated with the second video file;

based at least in part on the one or more manually entered tags associated with the second video file, determine a preliminary category for the second video file;

based on the preliminary category of the second video file, generate a targeted transcript of the second video file, wherein the targeted transcript of the second video file includes a second plurality of words;

generate an ontology of the second plurality of words based on the targeted transcript of the second video file;

rank the plurality of words and the second plurality of words based on the plurality of scoring factors; and

based on the ranking of the plurality of words and the second plurality of words, generate one or more automated tags associated with both the video file and the second video file.

32. A non-transitory computer-readable medium having sets of instructions stored thereon which, when executed by a computer, cause the computer to:

receive one or more manually generated tags associated with a video file;

based at least in part on the one or more manually entered tags, determine a preliminary category for the video file;

based on the preliminary category, generate a targeted transcript of the video file, wherein the targeted transcript includes a plurality of words;

generate an ontology of the plurality of words based on the targeted transcript;

rank the plurality of words in the ontology based on a plurality of scoring factors;

based on the ranking of the plurality of words, generate one or more automated tags associated with the video file,

establish a top concepts threshold value;

determine that one or more of the rankings of the plurality of words exceeds the top concepts threshold; and

associate information about the one or more of the plurality of words with rankings that exceeds the top concepts with the video file to designate the top concepts of the video file,

wherein the plurality of scoring factors consists of two or more of;

proximity of words relative to other words, distribution of words throughout the targeted transcript of the video file, words related to the plurality of words throughout the targeted transcript of the video file, occurrence age of the related words, information associated with the one or more manually entered tags, vernacular meaning of the plurality of words, or colloquial considerations of the meaning of the plurality of words.

33. The non-transitory computer-readable medium of claim 32, wherein the sets of instructions when further executed by the computer cause the computer to:

generate a web page which includes an embedded widget;

based on the one or more automated tags searching web content to return a list of videos, blogs, audio, and web pages that match the one or more automated tags; and

providing within the widget a view of each of the returned results.

34. The non-transitory computer-readable medium of claim 32, wherein the sets of instructions when further executed by the computer cause the computer to:

determine if the rankings for each of the plurality of words exceed a threshold ranking value; and

exclude any of the plurality of words that have a ranking value lower than the threshold value.

35. The non-transitory computer-readable medium of claim 32, wherein the sets of instructions when further executed by the computer cause the computer to determine a score for each of the plurality of words, wherein the score includes word frequency, word distribution, and word variety.