AD detection using ID code and extracted signature

• US 7,483,835 B2
• Filed: 12/23/2002
• Issued: 01/27/2009
• Est. Priority Date: 12/23/2002
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: Arbitron Inc. (MOOD Media Corporation)
• Inventors: White, Thomas W., Neuhauser, Alan R.
• Primary Examiner(s): Chawan; Vijay B
• Application Number:
  US10/328,201
• Publication Number:
  US 20040122679A1
• Time in Patent Office: 2,227 Days
• Field of Search: 704270-275, 382/100, 380/54, 725/22, 713/176
• US Class Current: 704/273
• CPC Class Codes: H04H 20/14 for monitoring programmes
• CPC Class Codes: H04H 20/33 by plural channels
• CPC Class Codes: H04H 2201/90 characterised by the use of signatures
• CPC Class Codes: H04H 60/372 Programme
• CPC Class Codes: H04H 60/375 Commercial
• CPC Class Codes: H04H 60/44 for identifying broadcast stations
• 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

Systems and methods are provided for gathering audience measurement data relating to receipt of and/or exposure to audio data by an audience member. A signature characterizing the audio data and additional data are obtained, and the audio data is identified based on both.

Claims

1. A method of identifying audio data received at an audience member'"'"'s location, comprising:

obtaining signature data from the received audio data, wherein said signature data characterizes the received audio data;

obtaining additional data from the received audio data, wherein the additional data comprises at least one ancillary code encoded within the received audio data, and wherein at least a portion of the additional data is independent of the signature data; and

producing an identification of the received audio data based both on the signature data and the additional data.

2. The method of claim 1, wherein obtaining the signature data comprises forming a signature data set reflecting time-domain variations of the received audio data.

3. The method of claim 2, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the received audio data in a plurality of frequency sub-bands of the received audio data.

4. The method of claim 1, wherein obtaining the signature data comprises forming a signature data set reflecting frequency-domain variations in the received audio data.

5. The method of claim 1, wherein the additional data comprises a plurality of substantially single-frequency code components.

6. The method of claim 5, further comprising processing the received audio data to produce signal-to-noise ratios for the plurality of components.

7. The method of claim 1, wherein obtaining the signature data comprises forming a signature data set comprising signal-to-noise ratios for frequency components of the audio data and/or data representing characteristics of the audio data.

8. The method of claim 7, wherein obtaining the signature data further comprises combining selected ones of the signal-to-noise ratios.

9. The method of claim 7, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the signal-to-noise ratios.

10. The method of claim 9, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the received audio data.

11. The method of claim 10, wherein the sub-bands are substantially single-frequency sub-bands.

12. The method of claim 7, wherein obtaining the signature data further comprises forming a signature data set reflecting frequency-domain variations of the signal-to-noise ratios.

13. The method of claim 12, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

14. The method of claim 1, wherein the signature data comprises data obtained from the additional data and/or a source identification code included in the audio data.

15. The method of claim 14, wherein the additional data and the source identification code occur simultaneously in the audio data.

16. The method of claim 14, wherein the additional data and the source identification code occur in different time segments of the audio data.

17. The method of claim 1, wherein the step of identifying the received audio data comprises comparing the obtained signature data to reference signature data of identified audio data.

18. The method of claim 1, wherein identifying the received audio data comprises:

selecting a signature subset of reference audio data signatures from a library of reference audio data signatures, each which signatures characterizes identified audio data, based on the additional data; and

comparing the signature data to at least one reference audio data signature in the signature subset to identify the received audio data.

19. The method of claim 18, wherein obtaining the signature data comprises forming a signature data set reflecting time-domain variations of the received audio data.

20. The method of claim 19, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the received audio data in a plurality of frequency sub-bands of the received audio data.

21. The method of claim 18, wherein obtaining the signature data comprises forming a signature data set reflecting frequency-domain variations in the received audio data.

22. The method of claim 18, wherein the additional data comprises a plurality of substantially single-frequency code components.

23. The method of claim 22, further comprising processing the received audio data to produce signal-to-noise ratios for the plurality of components.

24. The method of claim 18, wherein obtaining the signature data comprises forming a signature data set comprising signal-to-noise ratios for frequency components of the audio data and/or data representing characteristics of the audio data.

25. The method of claim 24, wherein obtaining the signature data further comprises combining selected ones of the signal-to-noise ratios.

26. The method of claim 24, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the signal-to-noise ratios.

27. The method of claim 26, wherein obtaining the signature data further comprises forming a signature data set reflecting time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the received audio data.

28. The method of claim 27, wherein the sub-bands are substantially single-frequency sub-bands.

29. The method of claim 24, wherein obtaining the signature data further comprises forming a signature data set reflecting frequency-domain variations of the signal-to-noise ratios.

30. The method of claim 29, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

31. The method of claim 18, wherein the signature data comprises data obtained from the additional data and/or a source identification code included in the audio data.

32. The method of claim 31, wherein the additional data and the source identification code occur simultaneously in the audio data.

33. The method of claim 31, wherein the additional data and the source identification code occur in different time segments of the audio data.

34. The method of claim 1, wherein identification of the received audio data comprises encoding the ancillary data to allow selection of a signature subset of reference audio data signatures from a library of reference audio data signatures characterizing identified audio data, said ancillary data occurring at least one of (a) simultaneously in the audio data, and (b) in different time segments of the audio data.

35. A system for identifying audio data received at an audience member'"'"'s location, comprising:

a first means to obtain signature data from the received audio data, wherein said signature data characterizes the received audio data;

a second means to obtain additional data from the received audio data, wherein the additional data comprises at least one ancillary code encoded within the received audio data, and wherein at least a portion of the additional data is independent of the signature data; and

a third means to produce an identification of the received audio data based both on the signature data and the additional data.

36. The system of claim 35, wherein the first means is operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the received audio data.

37. The system of claim 36, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the received audio data in a plurality of frequency sub-bands of the received audio data.

38. The system of claim 35, wherein the first means is operative to obtain the signature data by forming a signature data set reflecting frequency-domain variations in the received audio data.

39. The system of claim 35, wherein the additional data comprises a plurality of substantially single-frequency code components.

40. The system of claim 39, wherein the first means is operative to process the received audio data to produce signal-to-noise ratios for the plurality of components.

41. The system of claim 35, wherein the first means is operative to obtain the signature data by forming a signature data set comprising signal-to-noise ratios for frequency components of the audio data and/or data representing characteristics of the audio data.

42. The system of claim 41, wherein the first means is further operative to obtain the signature data by combining selected ones of the signal-to-noise ratios.

43. The system of claim 41, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the signal-to-noise ratios.

44. The system of claim 43, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the received audio data.

45. The system of claim 44, wherein the sub-bands are substantially single-frequency sub-bands.

46. The system of claim 41, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting frequency-domain variations of the signal-to-noise ratios.

47. The system of claim 46, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

48. The system of claim 35, wherein the signature data comprises data obtained from the additional data and/or a source identification code included in the audio data.

49. The system of claim 48, wherein the additional data and the source identification code occur simultaneously in the audio data.

50. The system of claim 48, wherein the additional data and the source identification code occur in different time segments of the audio data.

51. The system of claim 35, wherein the third means is operative to compare the obtained signature data to reference signature data of identified audio data.

52. The system of claim 35, wherein the third means comprises:

a first means to select a signature subset of reference audio data signatures from a library of reference audio data signatures, each of which signatures characterizes identified audio data, based on the additional data; and

a second means to compare the signature data to at least one reference audio data signature in the signature subset to identifier the received audio data.

53. The system of claim 52, wherein the first means is operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the received audio data.

54. The system of claim 53, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the received audio data in a plurality of frequency sub-bands of the received audio data.

55. The system of claim 52, wherein the first means is operative to obtain the signature data by forming a signature data set reflecting frequency-domain variations in the received audio data.

56. The system of claim 52, wherein the additional data comprises a plurality of substantially single-frequency code components.

57. The system of claim 56, wherein the first means is operative to process the received audio data to produce signal-to-noise ratios for the plurality of components.

58. The system of claim 52, wherein the first means is operative to obtain the signature data by forming a signature data set comprising signal-to-noise ratios for frequency components of the audio data anchor data representing characteristics of the audio data.

59. The system of claim 58, wherein the first means is further operative to obtain the signature data by combining selected ones of the signal-to-noise ratios.

60. The system of claim 58, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the signal-to-noise ratios.

61. The system of claim 60, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the received audio data.

62. The system of claim 61, wherein the sub-bands are substantially single-frequency sub-bands.

63. The system of claim 58, wherein the first means is further operative to obtain the signature data by forming a signature data set reflecting frequency-domain variations of the signal-to-noise ratios.

64. The system of claim 63, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

65. The system of claim 52, wherein the signature data comprises data obtained from the additional data and/or a source identification code included in the audio data.

66. The system of claim 65, wherein the additional data and the source identification code occur simultaneously in the audio data.

67. The system of claim 65, wherein the additional data and the source identification code occur in different time segments of the audio data.

68. The system of claim 35, wherein the third means comprises means for encoding the ancillary code to allow selection of a signature subset of reference audio data signatures from a library of reference audio data signatures characterizing identified audio data, said ancillary data occurring at least one of (a) simultaneously in the audio data, and (b) in different time segments of the audio data.

69. A method of encoding audio data for gathering data reflecting receipt of and/or exposure to the audio data, comprising:

forming a database having a plurality of reference signature data sets, wherein each of the signature data sets characterizes the audio data for identification;

grouping the reference signature data sets into a plurality of respective signature data groups; and

encoding audio data with additional data comprising an ancillary code that denotes at least one of the signature data groups, wherein the encoded audio data is identifiable based on the ancillary code and the at least one denoted signature data group.

70. The method of claim 69, wherein forming the database comprises forming the plurality of signature data sets, wherein each of the sets reflects time-domain variations of identified audio data.

71. The method of claim 70, wherein forming the database further comprises forming the plurality of signature data sets, wherein each of the sets reflects time-domain variations of identified audio data in a plurality of frequency sub-bands of the identified audio data.

72. The method of claim 69, wherein forming the database comprises forming the plurality of signature data sets, wherein each of the sets reflects frequency-domain variations in the identified audio data.

73. The method of claim 69, wherein the data denoting one of the signature data groups comprises a plurality of substantially single-frequency code components.

74. The method of claim 69, wherein forming the database comprises forming the plurality of signature data sets, wherein each of the sets comprises signal-to-noise ratios for frequency components of the audio data and/or data representing characteristics of the audio data.

75. The method of claim 74, wherein forming the signature data sets further comprises combining selected ones of the signal-to-noise ratios.

76. The method of claim 74, wherein forming the database further comprises forming the plurality of signature data sets, wherein each of the sets reflects time-domain variations of the signal-to-noise ratios.

77. The method of claim 76, wherein forming the database further comprises forming a plurality of signature data sets, wherein each of the sets reflects time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the identified audio data.

78. The method of claim 77, wherein the sub-bands are substantially single-frequency sub-bands.

79. The method of claim 74, wherein forming the database further comprises forming a plurality of signature data sets, wherein each of the sets reflects frequency-domain variations of the signal-to-noise ratios.

80. The method of claim 79, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

81. The method of claim 69, wherein the signature data comprises data obtained from the data denoting one of the signature data groups and/or a source identification code included in the audio data.

82. The method of claim 81, wherein the data denoting one of the signature data groups and the source identification code occur simultaneously in the audio data.

83. The method of claim 81, wherein the data denoting one of the signature data groups and the source identification code occur in different time segments of the audio data.

84. The method of claim 69, further comprising further grouping the reference signature data sets into a plurality of signature data subgroups.

85. The method of claim 69, wherein encoding audio data comprises encoding the ancillary code to allow selection of a signature subset of reference audio data signatures from a library of reference audio data signatures characterizing identified audio data, said ancillary data occurring at least one of (a) simultaneously in the audio data, and (b) in different time segments of the audio data.

86. A system for encoding audio data for gathering data reflecting receipt of and/or exposure to the audio data, comprising:

a database having a plurality of signature data groups, wherein each of the signature data groups has at least one reference signature data set that characterizes audio data for identification; and

an encoder to encode audio data to be monitored with additional data comprising ancillary code that denotes at least one of the signature data groups, wherein the encoded audio data is identifiable based on the ancillary code and the denoted signature data group.

87. The system of claim 86, wherein each reference signature data set reflects time-domain variations of identified audio data.

88. The system of claim 87, wherein each reference signature data set reflects time-domain variations of identified audio data in a plurality of frequency sub-bands of the identified audio data.

89. The system of claim 86, wherein each reference signature data set reflects frequency-domain variations in the identified audio data.

90. The system of claim 86, wherein the data denoting one of the signature data groups comprises a plurality of substantially single-frequency code components.

91. The system of claim 86, wherein each reference signature data set comprises signal-to-noise ratios for frequency components of the audio data and/or data representing characteristics of the audio data.

92. The system of claim 91, wherein each reference signature data set comprises a combination of selected ones of the signal-to-noise ratios.

93. The system of claim 91, wherein each reference signature data set reflects frequency-domain variations of the signal-to-noise ratios.

94. The system of claim 93, wherein the signal-to-noise ratios reflect the ratios of the magnitudes of substantially single-frequency components data to noise levels.

95. The system of claim 90, wherein each reference signature data set reflects time-domain variations of the signal-to-noise ratios.

96. The system of claim 95, wherein each reference signature data set reflects time-domain variations of the signal-to-noise ratios in a plurality of frequency sub-bands of the identified audio data.

97. The system of claim 96, wherein the sub-bands are substantially single-frequency sub-bands.

98. The system of claim 86, wherein the signature data comprises data obtained from the data denoting one of the signature groups and/or a source identification code included in the audio data.

99. The system of claim 98, wherein the data denoting one of the signature data groups and the source identification code occur simultaneously in the audio data.

100. The system of claim 98, wherein the data denoting one of the signature data groups and the source identification code occur in different time segments of the audio data.

101. The system of claim 86, wherein the reference signatures are further grouped into reference signature data subgroups.

102. A system for identifying audio data received at an audience member'"'"'s location, comprising:

a monitoring device comprising an input that receives audio data and is configured to obtain (1) signature data from the received audio data, wherein said signature data characterizes the received audio data, and (2) additional data from the received audio data, wherein the additional data comprises an ancillary code encoded within the received audio data, and wherein at least a portion of the additional data is independent of the signature data; and

a reporting system, coupled to the monitoring device, wherein the reporting system is configured to process both the signature data and the additional data to produce an identification of a program segment.

103. The system as recited in claim 102, wherein the monitoring device comprises a processor for obtaining the signature data and the additional data.

104. The system as recited in claim 102, wherein the reporting system comprises one or more processors to select a signature subset of reference audio data signatures, each of which signatures characterizes identified audio data, based on the additional data, and a database for comparison of the signature data to at least one reference audio data signature in the signature subset to identify the received audio data.