Methods and apparatus to improve detection of audio signatures
Methods, apparatus, systems and articles of manufacture are disclosed to improve detection of audio signatures. An example apparatus includes at least one memory, instructions in the apparatus, and processor circuitry to execute the instructions to: determine a first time difference of arrival for a first audio sensor of a meter and a second audio sensor of the meter based on a first audio recording from the first audio sensor and a second audio recording from the second audio sensor; determine a second time difference of arrival for the first audio sensor and a third audio sensor of the meter based on the first audio recording and a third audio recording from the third audio sensor; determine a match by comparing the first time difference of arrival to i) a first virtual source time difference of arrival and ii) a second virtual source time difference of arrival; in response to determining that the first time difference of arrival matches the first virtual source time difference of arrival, identify a first virtual source location as the location of a media presentation device presenting media; and remove the second audio recording to reduce a computational burden on the processor.
1. A non-transitory computer readable medium comprising instructions that, when executed, cause at least one processor to at least:
generate a plurality of virtual sources at a plurality of angles relative to a meter;
determine a virtual source time difference of arrival for a particular virtual source of the plurality of the virtual sources;
determine a first time difference of arrival for a first audio recording associated with a first pair of audio sensors of the meter;
compare the first time difference of arrival to the virtual source time difference of arrival; and
in response to a determination that the first time difference of arrival aligns with the virtual source time difference of arrival, identify a particular virtual source location associated with the virtual source time difference of arrival as a location of a media presentation device presenting media captured by the first pair of the audio sensors.
2. The non-transitory computer readable medium of claim 1 , wherein the plurality of angles relative to the meter comprises a plurality of radially spaced different angles relative to the meter.
3. The non-transitory computer readable medium of claim 1 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least access audio recordings from audio sensors of the meter.
4. The non-transitory computer readable medium of claim 1 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least:
determine a second time difference of arrival for a second audio recording associated with a second pair of audio sensors of the meter; and
compare the second time difference of arrival to the virtual source time difference of arrival.
5. The non-transitory computer readable medium of claim 4 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least remove the second audio recording to reduce noise.
6. The non-transitory computer readable medium of claim 4 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least remove the second audio recording to reduce computational burden.
7. The non-transitory computer readable medium of claim 4 , wherein the first pair of audio sensors includes a first audio sensor and a second audio sensor, and the second pair of audio sensors includes the first audio sensor and a third audio sensor.
8. The non-transitory computer readable medium of claim 7 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least:
calculate a first time for a first virtual signal to reach the first audio sensor based on a distance of the particular virtual source from the first audio sensor and speed of sound;
calculate a second time for the first virtual signal to reach the second audio sensor based on a distance of the particular virtual source from the second audio sensor and the speed of sound; and
determine a particular virtual source time difference of arrival based on a difference between the first time and the second time.
9. The non-transitory computer readable medium of claim 4 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to process at least one of the first audio recording and the second audio recording from the audio sensors using a short-time Fourier transform to obtain audio transforms with time-frequency bins.
10. The non-transitory computer readable medium of claim 9 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least calculate an inter-channel time difference between a first transform and a second transform, wherein the inter-channel time difference is representative of the first time difference of arrival.
11. A meter comprising:
a plurality of audio sensors;
at least one memory;
instructions; and
at least one processor to execute the instructions to at least:
generate a plurality of virtual sources at a plurality of angles relative to the meter;
determine a virtual source time difference of arrival for a particular virtual source of the plurality of the virtual sources;
determine a first time difference of arrival for a first audio recording associated with a first pair of audio sensors of the meter;
compare the first time difference of arrival to the virtual source time difference of arrival; and
in response to a determination that the first time difference of arrival aligns with the virtual source time difference of arrival, identify a particular virtual source location associated with the-virtual source time difference of arrival as a location of a media presentation device presenting media captured by the first pair of the audio sensors.
12. The meter of claim 11 , wherein the plurality of angles relative to the meter comprises a plurality of radially spaced different angles relative to the meter.
13. The meter of claim 11 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least access audio recordings from audio sensors of the meter.
14. The meter of claim 11 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least:
determine a second time difference of arrival for a second audio recording associated with a second pair of audio sensors of the meter; and
compare the second time difference of arrival to the virtual source time difference of arrival.
15. The meter of claim 14 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least remove the second audio recording to reduce noise.
16. The meter of claim 14 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least remove the second audio recording to reduce computational burden.
17. The meter of claim 14 , wherein the first pair of audio sensors includes a first audio sensor and a second audio sensor, and the second pair of audio sensors includes the first audio sensor and a third audio sensor.
18. The meter of claim 17 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to at least:
calculate a first time for a first virtual signal to reach the first audio sensor based on a distance of the particular virtual source from the first audio sensor and a speed of sound;
calculate a second time for the first virtual signal to reach the second audio sensor based on a distance of the particular virtual source from the second audio sensor and the speed of sound; and
determine a particular virtual source time difference of arrival based on a difference between the first time and the second time.
19. The meter of claim 14 , wherein the instructions further comprise instructions that, when executed, cause at least one processor to process at least one of the first audio recording and the second audio recording from the audio sensors using a short-time Fourier transform to obtain audio transforms with time-frequency bins.
20. A method comprising:
generating a plurality of virtual sources at a plurality of angles relative to a meter;
determining a virtual source time difference of arrival for a particular virtual source of the plurality of the virtual sources;
determining a first time difference of arrival for a first audio recording associated with a first pair of audio sensors of the meter;
comparing the first time difference of arrival to the virtual source time difference of arrival; and
in response to a determination that the first time difference of arrival aligns with the virtual source time difference of arrival, identifying a particular virtual source location associated with the virtual source time difference of arrival as a location of a media presentation device presenting media captured by the first pair of the audio sensors.