Method and apparatus for identifying an unkown work

• US 20030023852A1
• Filed: 07/09/2002
• Published: 01/30/2003
• Est. Priority Date: 07/10/2001
• Status: Active Grant

• Current Assignee: Audible Magic Corporation
• Original Assignee: Audible Magic Corporation
• Inventors: Wold, Erling H.
• Granted Patent: US 6,968,337 B2
• Time in Patent Office: Days
• US Class Current: 713/176
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: G10L 25/00 Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 muting semiconductor-based amplifiers when some special characteristics of a signal are sensed by a speech detector, e.g. sensing when no signal is present, H03G3/34
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A method and apparatus for identifying an unknown work is disclosed. In one aspect, a method may includes the acts of providing a reference database having a reduced dimensionality containing signatures of sampled works; receiving a sampled work; producing a signature from the work; and reducing the dimensionality of the signature.

Claims

1. A method for identifying an unknown work comprising:

providing a reference database having a reduced dimensionality containing signatures of sampled works;

receiving a sampled work;

producing a signature from said work; and

reducing the dimensionality of said signature.

2. The method claim 1, further including the act of comparing said signature of said received work to said reference database.

3. The method of claim 2, further including the act of generating a list of candidates.

4. The method of claim 3, further including the act of comparing the non-reduced signature to the non-reduced candidate signatures to obtain a result.

5. The method of claim 1, wherein said signature includes a plurality of MFCCs calculated for each said segment.

6. The method of claim 5, wherein said sampled work includes a signature calculated by using a plurality of acoustical features from the group consisting of at least one of loudness, pitch, brightness, bandwidth, spectrum and MFCC coefficients.

7. The method of claim 1, wherein said act of reducing the dimensionality is performed by using an indexing strategy chosen from the group consisting of:

the k-d tree, the SS-tree, the R-tree, and the SR-tree.

8. The method of claim 1, wherein said sampled work signature comprises a plurality of segments and an identification portion.

9. The method of claim 1, wherein said sampled work comprises a segment size of approximately 0.5 to 3 seconds.

10. The method of claim 1, wherein said sampled work comprises a segment size of approximately 1 second.

11. The method of claim 1, wherein said sampled work comprises a hop size of less than 50% of the segment size.

12. The method of claim 1, wherein said sampled work comprises a hop size of approximately 0.1 seconds.

13. The method of claim 1, wherein said signature contains averages taken from the group consisting of:

loudness, pitch, brightness, bandwidth, spectral features, and cepstral coefficients.

14. The method of claim 1, wherein said act of reducing the dimensionality is performed by projecting the features onto a Karhunen-Loeve basis.

15. The method of claim 1, wherein said act of reducing the dimensionality is performed by brute force comparisons.

16. The method of claim 15, wherein 500 candidates are produced.

17. An apparatus for identifying an unknown work comprising:

means for providing a reference database having a reduced dimensionality containing signatures of sampled works;

means for receiving a sampled work;

means for producing a signature from said work; and

means for reducing the dimensionality of said signature.

18. The apparatus of claim 17, further including means for comparing said signature of said received work to said reference database.

19. The apparatus of claim 18, further including means for generating a list of candidates.

20. The apparatus of claim 19, further including means for comparing the non-reduced signature to the non-reduced candidate signatures to obtain a result.