Multimedia object grouping, selection, and playback system

• US 20080168390A1
• Filed: 01/05/2007
• Published: 07/10/2008
• Est. Priority Date: 01/05/2007
• Status: Active Grant

• Current Assignee: Harman International Industries Incorporated (Samsung Electronics Co. Ltd.)
• Original Assignee: Harman International Industries Incorporated (Samsung Electronics Co. Ltd.)
• Inventors: Benyamin, Daniel
• Granted Patent: US 7,842,876 B2
• Time in Patent Office: Days
• US Class Current: 715/810
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/64 Browsing; Visualisation therefor generation of a list or set of audio data G06F16/638
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: G11B 27/105 of operating discs
• CPC Class Codes: G11B 27/322 used signal is digitally coded

Abstract

A multimedia organization and playback system intelligently organizes media objects, such as music files, and plays back their contents. The system considers and analyzes multiple media object attributes to determine groups of similar songs. As a result, the system delivers a consistent selection of media to the listener despite wide variations in media characteristics and without overburdening the listener with complicated configuration input. The listener may play back grouped songs based on the similarities between songs. Successive selections may be more or less similar to a current selection based on the organization of the music objects.

Claims

1. A music object organization system, comprising:

music object organization logic operable to;

determine a first music object group and a second music object group based on a music object characteristic;

identify a heuristic distance measure applicable to the music objects; and

assign the music objects among the first music object group and the second music object group such that the heuristic distance measure between the music objects within the group satisfies a clustering criterion.

2. The music object organization system of claim 1 where the clustering criterion comprises a minimization of the heuristic distance measure between the music objects in the first group and a minimization of the heuristic distance measure between the music objects in the second group.

3. The music object organization system of claim 2 further comprising a memory operably coupled with the music object organization logic and comprising an n-tuple ordering of the music objects in a characteristic-based vector space, where the heuristic distance measure comprises a geometric distance between the music objects.

4. The music object organization system of claim 3 where the heuristic distance measure comprises a similarity measure between the music objects.

5. The music object organization system of claim 2 where the music object organization logic is operable to initiate execution of an expectation-maximization process that assigns the music objects based on the ordered n-tuples.

6. The music object organization system of claim 5 where the expectation-maximization process comprises a k-means process.

7. The music object organization system of claim 1 where the music object characteristic comprises a genre characteristic, an artist characteristic, an album characteristic, a composer characteristic, a period characteristic, a beat count characteristic, a listener-defined characteristic related to the music object, or any combination thereof.

8. The music object organization system of claim 1 where the music object organization logic is operable to extract the music object characteristic from an ID3 tag associated with the music object.

9. The music object organization system of claim 1 further comprising user interface generation logic operable to display a music object group selector.

29. The multimedia object organization system of claim 1 where the clustering criterion comprises a minimization of the heuristic distance measure between the multimedia objects in the first group and a minimization of the heuristic distance measure between the multimedia objects in the second group.

30. The multimedia object organization system of claim 29 further comprising a memory operably coupled with the multimedia object organization logic and comprising an n-tuple ordering of the multimedia objects in a characteristic-based vector space, where the heuristic distance measure comprises a geometric distance between the multimedia objects.

31. The multimedia object organization system of claim 30 where the heuristic distance measure comprises a similarity measure between the multimedia objects.

32. The multimedia object organization system of claim 29 where the multimedia object organization logic is operable to initiate execution of an expectation-maximization process that assigns the multimedia objects based on the ordered n-tuples.

33. The multimedia object organization system of claim 32 where the expectation-maximization process comprises a k-means process.

10. A music object playback system, comprising:

a computer-readable medium configured to store a group of music objects, the group organized based on a music object characteristic associated with the music objects such that a heuristic distance measure between the music objects satisfies a clustering criterion in a vector space determined by the music object characteristic;

user interface generation logic operable to display a music object group selector;

music object playback logic operable to select a next music object to play from the group of objects by adjusting a next music object selection space based on a determined maximum heuristic distance measure

based on an input from the user interface; and

a loudspeaker that outputs a currently playing music object.

11. The music object playback system of claim 10 where the clustering criterion comprises a minimization of the heuristic distance measure between the music objects in the group.

12. The music object playback system of claim 10 where the music object playback logic is further operable to select the next music object to play by randomly selecting the next music object from the group of music objects or to play the currently playing music object until an end of the currently playing music object.

13. The music object playback system of claim 11 where the heuristic distance measure comprises a geometric distance between the music objects within the vector space.

14. The music object playback system of claim 10 where the determined maximum heuristic distance measure is related to a heuristic distance scaling factor, the heuristic distance scaling factor determined based on input from the user interface.

15. The music object playback system of claim 14 where music object playback logic selects a next music object that is more like the currently playing music object by reducing the next music object selection space.

16. The music object playback system of claim 15 where the music object playback logic reduces the next music object selection space by increasing the heuristic distance scaling factor.

17. The music object playback system of claim 16 where the music object playback logic randomly selects the next music object within the determined maximum heuristic distance measure in the next music object selection space.

18. The music object playback system of claim 14 where music object playback logic selects a next music object that is less like the currently playing music object by increasing the next music object selection space.

19. The music object playback system of claim 18 where the music object playback logic increases the next music object selection space by decreasing the heuristic distance scaling factor.

20. The music object playback system of claim 18 where the music object playback logic randomly selects the next music object within the determined maximum heuristic distance measure in the next music object selection space.

21. A product for selecting a next object for playback from a group of music objects, the product comprising:

a machine-readable medium;

display instructions stored in the machine-readable medium that initiate display of a “

more like this”

user interface element through which an operator reduces a next music object selection space; and

display instructions stored in the machine-readable medium that initiate display of a “

less like this”

user interface element through which an operator expands the next music object selection search space.

22. The product of claim 21 where the group is organized based on a music object characteristic associated with the music objects such that a heuristic distance measure between the music objects satisfies a clustering criterion for an ordering of the music objects in a vector space determined by the music object characteristic, the heuristic distance measure comprising a geometric distance between the music objects in the vector space.

23. The product of claim 22, where the display instructions that initiate display of a “

more like this”

user interface element further comprise selection instructions that reduce the next music object selection space by increasing a heuristic distance scaling factor.

24. The product of claim 23 where the selection instructions are further operable to randomly select a next music object to play from the group of music objects within a determined maximum heuristic distance measure from a current music object, the determined maximum heuristic distance based on the heuristic distance scaling factor.

25. The product of claim 22, where the display instructions that initiate display of a “

more like this”

user interface element further comprise selection instructions that reduce the next music object selection space by increasing a heuristic distance scaling factor.

26. The product of claim 25 where the selection instructions are further operable to randomly select a next music object to play from the group of music objects within a determined maximum heuristic distance measure from a current music object, the determined maximum heuristic distance based on the heuristic distance scaling factor.

27. The product of claim 21 where the product further comprises:

display instructions stored in the machine-readable medium that initiate display of a “

track to next”

user interface element configured to randomly select a next music object from the group of music objects; and

display instructions stored in the machine-readable medium that initiate display of a “

play to end of track”

user interface element configured to play a current music object until completion of the current music object.

28. A multimedia object organization system, comprising:

multimedia object organization logic operable to;

determine a first multimedia object group and a second multimedia object group based on a multimedia object characteristic;

identify a heuristic distance measure applicable to the multimedia objects; and

assign the multimedia objects among the first multimedia object group and the second multimedia object group such that the heuristic distance measure between the multimedia objects within the group satisfies a clustering criterion.

34. The multimedia object organization system of claim 28, where the multimedia objects comprise a video object, a picture object, a phonebook entry object, an address book object, or a combination thereof.