System and method for audio fingerprinting

• US 6,963,975 B1
• Filed: 08/10/2001
• Issued: 11/08/2005
• Est. Priority Date: 08/11/2000
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Weare, Christopher Bruce
• Primary Examiner(s): Caldwell, Andrew
• Assistant Examiner(s): Teslovich, Tamara
• Application Number:
  US09/928,004
• Time in Patent Office: 1,551 Days
• Field of Search: 708/200, 708/212, 708/404, 708/405, 704/200.1, 380/37, 713/176, 705/51
• US Class Current: 713/176
• CPC Class Codes: G10L 19/018 Audio watermarking, i.e. embedding inaudible data in the audio signal

Abstract

A system and methods for the creation, management, and distribution of media entity fingerprinting are provided. In connection with a system that convergently merges perceptual and digital signal processing analysis of media entities for purposes of classifying the media entities, various means are provided to a user for automatically processing fingerprints for media entities for distribution to participating users. Techniques for providing efficient calculation and distribution of fingerprints for use in satisfying copyright regulations and in facilitating the association of meta data to media entities are included. In an illustrative implementation, the fingerprints may be generated and stored allowing for persistence of media from experience to experience. In various non-limiting embodiments, the processing of fingerprints includes calculating the average information density of the media entities, determining the standard deviation of the calculated information of the media entities, calculating the average critical band energy of the media entities, calculating the average standard deviation of the critical band energy of the media entities, determining the play-time of the media entities and processing the information density, the standard deviation of the information density, the critical band energy, the standard deviation of the critical band, and the play time to produce a bit-sequence representative of the fingerprint.

Claims

1. A method to create a fingerprint for media entities, comprising:

reading data indicative of a media entity desiring at least one fingerprint, said media entity data containing a sequence of random bits having a length N;

processing said media entity data in accordance with at least one fingerprinting algorithm, said fingerprinting algorithm employing bit-to-bit comparisons and at least one approximation technique to process fingerprints, wherein said processing further comprises;

calculating the average information density of said media entities;

determining the standard deviation of the calculated information of said media entities;

calculating the average critical band energy of the said media entities;

calculating the average standard deviation of the critical band energy of said media entities;

determining the play-time of said media entities; and

processing said information density, said standard deviation of said information density, said critical band energy, said standard deviation of said critical band, and said play time to produce a bit-sequence representative of said fingerprint.

2. The method as recited in claim 1, further comprising the step of comparing said bit sequence of said created fingerprint with said bit sequence of said data indicative of said media entities.

3. The method as recited in claim 2, wherein said comparing step contemplates the use of the Hamming distance between the fingerprint bit and the media entity bit to determine the probability that said fingerprint and said media entity bits differ by Hamming distance according to the relation,

P(M)=e⁻

(M−

N/2)<sup>2/2σ

2</sup>/σ

√

{square root over (2π

)},

wherein σ

is the standard deviation of the distribution expressed as,

σ

=√

{square root over (N/2)}.

4. The method as recited in claim 3, further comprising the step of calculating the probability that the Hamming distance between two sequences of random bits is less than a value M′

according to the relation, $P<br><br>\left(M<<br><br>M^{\mathrm{\prime <br><br>}}\right)={\int <br><br>}_0^{M^{\mathrm{\prime <br><br>}}-1}<br><br>{\mathrm{e<br><br>}}^{-{\left(x-N<br><br>\text{/}<br><br>2\right)}^2<br><br>\text{/}<br><br>2<br><br><br><br>{\mathrm{\sigma <br><br>}}^2}<br><br>\text{/}<br><br>\mathrm{\sigma <br><br>}<br><br><br><br>\sqrt{2<br><br><br><br>\mathrm{\pi <br><br>}}<br><br><br><br>d<br><br>x.$

5. The method as recited in claim 1, wherein the average information density is taken to be the average entropy per processing frame of said media entities.

6. The method as recited in claim 5, wherein said average information density is determined by the relation, $S_{\mathrm{ave}}=\frac{\sum _{}}{}$

j⁢

SjN

wherein, N is the total number of processing frames.

7. The method as recited in claim 6, wherein S_j is determined by the relation, $S_j=-\sum _{}$

n⁢

bn⁢

⁢

log⁢

⁢

2⁢

(bn),

where b_n is the absolute value of the nth bin of the normalized real FFT of the processing frame.

8. The claim as recited in claim 7, where in the average standard deviation of the information density of said media entities is determined by the relation, $S_{\mathrm{std}}=\frac{\sqrt{\sum _{}}}{}$

j⁢

(Save-Sj)2N.

9. The method as recited in claim 1, wherein the average critical band energy is determined by the relation, ${\overrightarrow{C}}_{}$

ave=∑

j⁢

C→

jN

wherein, {overscore (C)}_j is a vector of values consisting of the critical band energy in each critical band and N is the total number of processing frames.

10. The method as recited in claim 1, wherein the average standard deviation of the critical band energy is determined by the relation, $C_{\mathrm{std}}=\frac{\sqrt{\sum _{}}}{}$

j⁢

(Cave-Cj)2N

wherein, N is the total number of processing frames.

11. A computer readable medium bearing computer executable instructions for carrying out the method of claim 1.

12. A modulated data signal carrying computer executable instructions for carrying out the method of claim 1.

13. A computing device comprising means for carrying out each of the steps of the method of claim 1.

21. A method to identify media entities using fingerprints, comprising the steps of:

calculating a fingerprint in accordance with the steps of claim 1 of said media entities;

comparing said calculated fingerprint to already calculated fingerprints found in a cooperating fingerprint data store; and

evaluating the results of the comparison.

22. The method as recited in claim 21, further comprising the step of communicating the results of said evaluation step to participating users, said participating users comprising any of:

cooperating media entity processing systems, end-users, regulatory agencies.

23. A method to authenticate media entities to ensure compliance with copyright regulations by employing fingerprints, comprising the steps of:

calculating a fingerprint in accordance the steps of claim 1 of said media entities;

comparing said calculated fingerprint to fingerprints of authorized media entities stored in a cooperating data store; and

evaluating the results of the comparison to return a response indicative whether authorization was granted.

24. The method as recited in claim 23, further comprising the step of denying distribution access to media entities that are determined to be unauthorized.

14. A system to create a fingerprint for media entities comprising:

a sampling system;

a processing system cooperating with said sampling system to generate said fingerprints, said processing system comprising means to calculate the information density of said media entities, standard deviation of the information density of said media entities, average critical band energy of said media entities, standard deviation of the critical band energy of said media entities, and the play-time of said media entities; and

a communications interface, said communications interface cooperating with said processing system to communicate created fingerprints to participating users.

15. The system as recited in claim 14, wherein said sampling system prepares at least one sampling portion of said media entities for communication to said processing system.

16. The system as recited in claim 15, wherein said processing system cooperates with said sampling system to process said sampling portion when generating said fingerprint.

17. The system as recited in claim 14, wherein said processing system comprises a computing environment capable of performing said calculations.

18. The system as recited in claim 17, wherein said computing environment comprises any of a stand-alone or networked computing environments.

19. The system as recited in claim 14, wherein said communications interface comprises any of a fixed-wire LAN, a wireless LAN, a fixed-wire WAN, a wireless WAN, a fixed-wire extranet, a wireless extranet, a fixed-wire intranet, a wireless intranet, peer-to-peer computer network, the wireless Internet, and the fixed-wire Internet.

20. The system as recited in claim 14, wherein said processing system is a component of a media content analysis and distribution system.