Content identification system
The content of a media program is recognized by analyzing its audio content to extract therefrom prescribed features, which are compared to a database of features associated with identified content. The identity of the content within the database that has features that most closely match the features of the media program being played is supplied as the identity of the program being played. The features are extracted from a frequency domain version of the media program by a) filtering the coefficients to reduce their number, e.g., using triangular filters; b) grouping a number of consecutive outputs of triangular filters into segments; and c) selecting those segments that meet prescribed criteria, such as those segments that have the largest minimum segment energy with prescribed constraints that prevent the segments from being too close to each other. The triangular filters may be log-spaced and their output may be normalized.
1. An apparatus for use in recognizing the content of a media program, the apparatus comprising:
processing circuitry configured to digitize an audio representation of a media program to form a digitized audio representation of the media program;
a memory configured to store the digitized audio representation of the media program; and
a processor communicatively connected to the memory, the processor configured to:
divide the digitized audio representation into time domain blocks of a prescribed number of samples;
smooth the time domain blocks using a filter;
convert the smoothed time domain blocks into frequency domain blocks, wherein the smoothed time domain blocks are represented by frequency coefficients;
filter each first frequency domain representation of blocks of the media program using a plurality of filters to develop a respective second frequency domain representation of each of the blocks of the media program, the second frequency domain representation of each of the blocks having a reduced number of frequency coefficients with respect to the first frequency domain representation;
group frequency coefficients of the second frequency domain representation of the blocks to form frequency coefficient segments;
select a plurality of the segments as representing the media program;
compare the selected segments to frequency coefficient segments of stored programs to provide corresponding matching scores; and
determine the media program using the matching scores.
2. An apparatus for developing database information adapted for use in identifying an unidentified media program, the apparatus comprising:
processing circuitry configured to digitize an audio representation of an identified media program to form a digitized audio representation of the identified media program;
a memory configured to store the digitized audio representation of the identified media program; and
a processor communicatively connected to the memory, the processor configured to:
divide the digitized audio representation into time domain blocks of a prescribed number of samples;
smooth the time domain blocks using a filter;
convert the smoothed time domain blocks into first frequency domain blocks of respective frequency coefficients;
filter the first frequency domain blocks using a plurality of filters to develop respective second frequency domain blocks, the second frequency domain blocks having a reduced number of frequency coefficients with respect to the first frequency domain blocks;
group frequency coefficients of the second frequency domain blocks to form frequency coefficient segments;
select a plurality of the frequency coefficient segments as representing the identified media program; and
store the selected frequency coefficient segments in a database so as to relate the selected frequency coefficients segments with an identifier of the identified media program.
3. The apparatus of claim 2 , wherein a grouping of frequency coefficients of the second frequency domain blocks to form a respective one of the frequency coefficient segments includes second frequency domain blocks that are consecutive in time in the identified media program.
4. The apparatus of claim 2 , wherein the selected frequency coefficient segments correspond to portions of the identified media program that are not contiguous in time.
5. The apparatus of claim 2 , wherein the plurality of filters includes at least a set of triangular filters or at least a set of log-spaced triangular filters.
6. The apparatus of claim 2 , wherein the selected frequency coefficient segments are those that have largest minimum segment energy.
7. The apparatus of claim 2 , wherein the selected frequency coefficient segments are selected in accordance with prescribed constraints such that the selected frequency coefficient segments are separated by a minimum distance.
8. The apparatus of claim 2 , wherein the selected frequency coefficient segments are selected to contain portions of the identified media program that correspond in time to prescribed search windows that are separated by gaps.
9. The apparatus of claim 2 , wherein the selected frequency coefficient segments are those having a maximum entropy.
10. The apparatus of claim 2 , wherein the processor is configured to:
normalize frequency coefficients forming the selected frequency coefficient segments.
11. The apparatus of claim 2 , wherein the processor is configured to:
store, in the database, timing information associated with the selected frequency coefficient segments.
12. An apparatus for identifying an unidentified media program, the apparatus comprising:
processing circuitry configured to digitize an audio representation of the unidentified media program to form a digitized audio representation of the unidentified media program;
a memory configured to store the digitized audio representation of the unidentified media program; and
a processor communicatively connected to the memory, the processor configured to:
divide the digitized audio representation into time domain blocks of a prescribed number of samples;
smooth the time domain blocks using a filter;
convert the smoothed time domain blocks into first frequency domain blocks of respective frequency coefficients;
filter the first frequency domain blocks using a plurality of filters to develop respective second frequency domain blocks, the second frequency domain blocks having a reduced number of frequency coefficients with respect to the first frequency domain blocks;
group frequency coefficients of the second frequency domain blocks to form frequency coefficient segments;
select a plurality of the frequency coefficient segments as representing the unidentified media program; and
compare the selected frequency coefficient segments of the unidentified media program to entries in a database of selected frequency coefficient segments that have been identified to relate to a media program to identify thereby the unidentified media program.
13. The apparatus of claim 12 , wherein a grouping of frequency coefficients of the second frequency domain blocks to form a respective one of the frequency coefficient segments includes second frequency domain blocks that are consecutive in time in the unidentified media program.
14. The apparatus of claim 12 , wherein the selected frequency coefficient segments correspond to portions of the unidentified media program that are not contiguous in time.
15. The apparatus of claim 12 , wherein the plurality of filters includes at least a set of triangular filters or at least a set of log-spaced triangular filters.
16. The apparatus of claim 12 , wherein the selected frequency coefficient segments are those that have largest minimum segment energy.
17. The apparatus of claim 12 , wherein the selected frequency coefficient segments are selected in accordance with prescribed constraints such that the selected frequency coefficient segments are separated by a minimum distance.
18. The apparatus of claim 12 , wherein the selected frequency coefficient segments are selected to contain portions of the unidentified media program that correspond in time to prescribed search windows that are separated by gaps.
19. The apparatus of claim 12 , wherein the selected frequency coefficient segments are those having a maximum entropy.
20. The apparatus of claim 12 , wherein the processor is configured to:
normalize frequency coefficients forming the selected frequency coefficient segments.
21. The apparatus of claim 12 , wherein the processor is configured to:
compare the selected frequency coefficient segments of the unidentified media program to entries in the database to determine corresponding matching scores; and
identify the unidentified media program based on the matching scores.