Noise based interest point density pruning
View Patent ↗Systems and methods for noise based interest point density pruning are disclosed herein. The systems include determining an amount of noise in an audio sample and adjusting the amount of interest points within an audio sample fingerprint based on the amount of noise. Samples containing high amounts of noise correspondingly generate fingerprints with more interest points. The disclosed systems and methods allow reference fingerprints to be reduced in size while increasing the size of sample fingerprints. The benefits in scalability do not compromise the accuracy of an audio matching system using noise based interest point density pruning.
1. A system, comprising:
a processor; and
a memory communicatively coupled to the processor, the memory having stored therein computer executable components comprising:
a noise detection component that determines a noise level in an audio sample based upon a determination of a source device of the audio sample and an association of the source device to a corresponding noise level;
an interest point detection component that generates a set of interest points for the audio sample;
a density component that selects a subset of the interest points, wherein an amount of the interest points to include in the subset is determined as a function of the noise level, the amount of interest points to include in the subset increases as the noise level increases; and
a fingerprint component that generates a set of sample fingerprints for the audio sample based on the subset of the interest points.
2. The system of claim 1 , wherein the association of the source device to the corresponding noise level is determined based upon whether the source device is expected to have recorded the audio sample or expected to have obtained the audio sample from a storage library.
3. The system of claim 2 , wherein the expectation to have recorded the audio sample is indicative of associated with a high noise level.
4. The system of claim 2 , wherein the expectation to have obtained the audio sample from a storage library is associated with a low noise level.
5. The system of claim 1 , wherein the source device of the audio sample includes at least one of a phone, a tablet, a personal digital assistant, a server, a personal computer, or a media player.
6. The system of claim 1 , further comprising a matching component that:
determines whether there is a reference sample in a set of reference samples that matches the audio sample based upon comparing the set of sample fingerprints with respective sets of reference fingerprints associated with the set of reference samples.
7. The system of claim 6 , wherein the amount of interest points in the set of sample fingerprints is greater than the amount of interest points in respective reference fingerprints in the set of reference fingerprints.
8. A method, comprising:
identifying, by a system including a processor, a noise level in an audio sample based upon a determination of a source device of the audio sample and an association of the source device to a corresponding noise level;
creating, by the system, a set of interest points for the audio sample;
selecting, by the system, a subset of the interest points, wherein an amount of the interest points to include in the subset is determined as a function of the noise level, the amount of interest points to include in the subset increases as the noise level increases; and
creating, by the system, a set of sample fingerprints for the audio sample based on the subset of the interest points.
9. The method of claim 1 , wherein the association of the source device to the corresponding noise level is determined based upon whether the source device is expected to have recorded the audio sample or expected to have obtained the audio sample from a storage library.
10. The method of claim 9 , wherein the expectation to have recorded the audio sample is indicative of associated with a high noise level.
11. The method of claim 9 , wherein the expectation to have obtained the audio sample from a storage library is associated with a low noise level.
12. The method of claim 8 , wherein the source device of the audio sample includes at least one of a phone, a tablet, a personal digital assistant, a server, a personal computer, or a media player.
13. The method of claim 8 , further comprising a matching component that:
determining, by the system, whether there is a reference sample in a set of reference samples that matches the audio sample based upon comparing the set of sample fingerprints with respective sets of reference fingerprints associated with the set of reference samples.
14. The method of claim 13 , wherein the amount of interest points in the set of sample fingerprints is greater than the amount of interest points in respective reference fingerprints in the set of reference fingerprints.
15. A non-transitory computer-readable medium having instructions stored thereon that, in response to execution, cause a system including a processor to perform operations comprising:
ascertaining a noise level in an audio sample based upon a determination of a source device of the audio sample and an association of the source device to a corresponding noise level;
producing a set of interest points for the audio sample;
selecting a subset of the interest points, wherein an amount of the interest points to include in the subset is determined as a function of the noise level, the amount of interest points to include in the subset increases as the noise level increases; and
producing a set of sample fingerprints for the audio sample based on the subset of the interest points.
16. The non-transitory computer-readable medium of claim 15 , wherein the association of the source device to the corresponding noise level is determined based upon whether the source device is expected to have recorded the audio sample or expected to have obtained the audio sample from a storage library.
17. The non-transitory computer-readable medium of claim 16 , wherein the expectation to have recorded the audio sample is indicative of associated with a high noise level.
18. The non-transitory computer-readable medium of claim 16 , wherein the expectation to have obtained the audio sample from a storage library is associated with a low noise level.
19. The non-transitory computer-readable medium of claim 15 , wherein the source device of the audio sample includes at least one of a phone, a tablet, a personal digital assistant, a server, a personal computer, or a media player.
20. The non-transitory computer-readable medium of claim 15 , further comprising a matching component that:
determining, by the system, whether there is a reference sample in a set of reference samples that matches the audio sample based upon comparing the set of sample fingerprints with respective sets of reference fingerprints associated with the set of reference samples.