Methods and apparatus for generating signatures

• US 8,458,737 B2
• Filed: 04/28/2008
• Issued: 06/04/2013
• Est. Priority Date: 05/02/2007
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Topchy, Alexander, Srinivasan, Venugopal
• Primary Examiner(s): Smits, Talivaldis Ivars
• Assistant Examiner(s): Roberts, Shaun
• Application Number:
  US12/110,951
• Publication Number:
  US 20080276265A1
• Time in Patent Office: 1,863 Days
• Field of Search: 704/205, 725/22, 725/9, 725/18, 713/176
• US Class Current: 725/18
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: H04H 20/14 for monitoring programmes
• CPC Class Codes: H04H 2201/90 characterised by the use of signatures
• CPC Class Codes: H04H 60/58 of audio determination or detection of speech characteristics in general G10L25/00; speech recognition in general G10L15/00

Abstract

Methods and apparatus for characterizing media are described. In one example, a method of characterizing media includes capturing a block of audio; converting at least a portion of the block of audio into a frequency domain representation; dividing the frequency domain representation into a plurality of bands; determining a characteristic difference of a first band of the plurality of bands based on a comparison of a characteristic of the first band and a characteristic of a second band different from the first band, wherein the characteristic of the first band occurs at a time that is different than a time at which the characteristic of the second band takes place; and determining a signature bit based on a characteristic difference. Other examples are shown and described.

Claims

1. A method of characterizing media comprising:

converting at least a portion of a block of audio into a frequency domain representation;

dividing the frequency domain representation into a plurality of bands;

determining a characteristic difference of a first band of the plurality of bands based on a comparison of a first characteristic of the first band and a second characteristic of a second band different from the first band, wherein the first characteristic of the first band occurs at a time that is different than a time at which the second characteristic of the second band takes place, and wherein the characteristic difference is determined differently based on whether a count value associated with the first band has a first property; and

determining a signature bit based on the characteristic difference.

2. A method as defined in claim 1, further comprising capturing a block of audio by obtaining audio via a hardwired connection.

3. A method as defined in claim 1, further comprising capturing a block of audio by obtaining audio via a wireless connection.

4. A method as defined in claim 1, wherein converting at least a portion of the block of audio into the frequency domain representation comprises the use of a Fourier transformation.

5. A method as defined in claim 1, wherein dividing the frequency domain representation into the plurality of bands comprises grouping frequency components that are adjacent in the frequency domain representation.

6. A method as defined in claim 5, wherein the frequency components have frequencies in an audible frequency range.

7. A method as defined in claim 1, wherein determining the characteristic of a selected band comprises determining an energy in the selected band.

8. A method as defined in claim 1, wherein the first band has a first frequency that is higher than a second frequency of the second band.

9. A method as defined in claim 8, wherein the comparison comprises a subtraction.

10. A method as defined in claim 1, wherein the first band has a first frequency that is higher than a second frequency of the second band, and wherein the first characteristic of the first band occurred later in time than the second characteristic of the second band.

11. A method as defined in claim 1, wherein the comparison comprises a subtraction.

12. A method as defined in claim 1, wherein the first band has a highest frequency that is lower than a lowest frequency of the second band.

13. A method as defined in claim 12, wherein the first characteristic of the first band occurred later in time than the second characteristic of the second band.

14. A method as defined in claim 1, wherein determining the signature bit based on the characteristic difference comprises comparing the characteristic difference to a value of zero.

15. A method as defined in claim 14, wherein determining the signature bit based on the characteristic difference further comprises comparing the characteristic difference to another characteristic difference.

16. A method as defined in claim 1, wherein the count value has the first property when the count value is an even number and does not have the first property when the count value is an odd number.

17. A signature generation system comprising:

a transformer to convert at least a portion of a block of audio into a frequency domain representation;

a divider to divide the frequency domain representation into a plurality of bands;

a characteristic determiner to determine a first characteristic of a first band of the plurality of bands and a second characteristic of a second band different from the first band, wherein the first characteristic of the first band occurs at a time that is different than a time at which the second characteristic of the second band takes place;

a subtractor to determine a characteristic difference between the first characteristic and the second characteristic, wherein the characteristic difference is determined differently based on whether a count value associated with the first band has a first property; and

a signature determiner to determine a signature bit based on the characteristic difference.

18. A signature generation system as defined in claim 17, further comprising a sample generator to capture a block of audio by obtaining audio via a hardwired connection.

19. A signature generation system as defined in claim 17, further comprising a sample generator to capture a block of audio by obtaining audio via a wireless connection.

20. A signature generation system as defined in claim 17, wherein converting at least a portion of the block of audio into the frequency domain representation comprises the use of a Fourier transformation.

21. A signature generation system as defined in claim 17, wherein dividing the frequency domain representation into the plurality of bands comprises grouping frequency components that are adjacent in the frequency domain representation.

22. A signature generation system as defined in claim 21, wherein the frequency components have frequencies in an audible frequency range.

23. A signature generation system as defined in claim 17, wherein determining the characteristic of a selected band comprises determining an energy in the selected band.

24. A signature generation system as defined in claim 23, wherein the first band has a lowest frequency that is higher than a highest frequency of the second band.

25. A signature generation system as defined in claim 24, wherein the first characteristic of the first band occurred later in time than the second characteristic of the second band.

26. A signature generation system as defined in claim 17, wherein the count value has the first property when the count value is an even number and does not have the first property when the count value is an odd number.

27. A tangible machine readable medium comprising instructions that, when executed, cause a machine to at least:

convert at least a portion of a block of audio into a frequency domain representation;

divide the frequency domain representation into a plurality of bands;

determine a characteristic difference of a first band of the plurality of bands based on a comparison of a first characteristic of the first band and a second characteristic of a second band different from the first band, wherein the first characteristic of the first band occurs at a time that is different than a time at which the second characteristic of the second band takes place, and wherein the characteristic difference is determined differently based on whether a count value associated with the first band has a first property; and

determine a signature bit based on the characteristic difference.

28. A machine readable medium as defined in claim 27, wherein the first band has a first frequency that is lower than a second frequency of the second band, and wherein the first characteristic of the first band occurred later in time than the second characteristic of the second band.

29. A machine readable medium as defined in claim 27, wherein determining the signature bit based on the characteristic difference comprises comparing the characteristic difference to a value of zero.

30. A machine readable medium as defined in claim 29, wherein determining the signature bit based on the characteristic difference further comprises comparing the characteristic difference to another characteristic difference.

31. A machine readable medium as defined in claim 27, wherein the count value has the first property when the count value is an even number and does not have first property when the count value is an odd number.