Music content characteristic identification and matching

• US 6,545,209 B1
• Filed: 07/05/2001
• Issued: 04/08/2003
• Est. Priority Date: 07/05/2000
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Deeds, Paul D., Schrock, Christian E., Flannery, Frank P.
• Primary Examiner(s): Witkowski, Stanley J.
• Application Number:
  US09/899,490
• Time in Patent Office: 642 Days
• Field of Search: 846/09--14, 434/307.A
• US Class Current: 84/609
• 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 16/637 Administration of user profiles, e.g. generation, initialization, adaptation or distribution
• CPC Class Codes: G06F 16/639 using playlists
• 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: G10H 1/0008 Associated control or indicating means
• CPC Class Codes: G10H 1/0058 Transmission between separate instruments or between individual components of a musical system G10H1/0083 takes precedence
• CPC Class Codes: G10H 2240/125 Library distribution, i.e. distributing musical pieces from a central or master library
• CPC Class Codes: G10H 2240/135 Library retrieval index, i.e. using an indexing scheme to efficiently retrieve a music piece
• CPC Class Codes: G10H 2240/305 Internet or TCP/IP protocol use for any electrophonic musical instrument data or musical parameter transmission purposes

Abstract

A system and methods for matching media entities to associate closely related media entities 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 match media entities using aggregation and non-aggregation matching. In an illustrative implementation, various factors representative of inherent characteristics of the media entities are employed and processed to generate data sets having closely related and/or similarly situated media entities. Once a media matching is performed on a library of media entities, the results of the media match may be persisted for the user from experience to experience.

Claims

1. A method for selecting media entities that closely relate to a reference media entity from a data set having at least one media entity comprising the steps of:

producing media entities match scores for said at least one media entity of said data set by comparing media entity affinity data of reference song with media affinity data of said at least one media entity of said data set, wherein said media affinity data comprises at least one factor from a similarity matrix;

storing media entity match scores for media entities having scores nearest the score to the reference media entity for selecting the closest set of media entities to said reference media entity; and

generating a resultant data set of media entities using said produced and stored match scores, wherein said resultant data set containing closely related media entities to said reference media entity.

2. The method as recited in claim 1, wherein the producing step further comprising the step of pruning down said data set having at least one media entity using sub-genre factors to eliminate dissimilar media entities from being processed in said selection method.

3. The method as recited in claim 2, further comprising the steps of determining at least one sub-genre for said reference media entity and comparing said sub-genre with said sub-genre factors of said at least one media of said data set, said sub-genre factors comprising at least one of rhythm time, rhythm type, style, vocal voices, flavor, mood, emotion, density, weight, consonance, melodic movement, tempo, or rhythm activity.

4. The method as recited in claim 3, further comprising the step of applying at least one scoring-based algorithm on said data set having said at least one media entity to produce said matching scores, said algorithm employing said sub-genre factors to produce said matching scores.

5. The method as recited in claim 1, wherein said producing step further comprises the step of performing a look-up on at least one similarity matrix to obtain said matching score.

6. The method as recited in claim 1, further comprising the step of associating N number of media entities having the closest relationship based on said matching scores with said reference media entity.

7. The method as recited in claim 1, further comprising the step of producing a roll-up table, said roll-up table having at least one look-up index to facilitate searching of similarly matched songs.

8. The method as recited in claim 1, further comprising the step of performing aggregate matching to produce a data set of matched media entities matched at least one combination of various media entity characteristics, said characteristics comprising at least one of media entity author, media entity title, or user-defined search parameters.

9. The method as recited in claim 8, wherein aggregate matching further comprises the steps of:

creating a graph with an edge connecting each media entity having a pre-determined range of match to another media entity, wherein each edge has a direction going from a source (the media entity being matched against) to the sink (the second media entity, which matches the first media entity);

assigning a weight to each edge relative to its ranking within said range;

summing the edge weights; and

categorizing said summed edge weights for said media entities according to their sinks.

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

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

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

13. A system for matching media entities to provide associations between closely related media entities comprising:

a scoring system that accepts as input data sets representative of media entity data and similarity matrices data, wherein said scoring system applies at least one weighting algorithm on said media entity data and said similarity matrices data to calculate matched scores, said algorithm employing inherent media element characteristics when calculate said matched scores; and

a data store cooperating with said scoring system to store said matched scores for media entities having scores nearest the score to a selected reference media entity.

14. The system as recited in claim 13, further comprising an aggregation system, said aggregation system employing at least one node counting algorithm to assign weights to media entities, wherein said media entities are sorted according to said assigned weights to create associations between relevant and closely related media entities.

15. The system as recited in claim 14, wherein said at least one node counting algorithm creates a graph with an edge connecting each media entity having a pre-determined range of match to another media entity, wherein each edge has a direction going from a source (the media entity being matched against) to the sink (the second media entity, which matches the first media entity), assigning a weight to each edge relative to its ranking within said range, summing the edge weights, and categorizing said summed edge weights for said media entities according to their sinks.

16. The system as recited in claim 13, wherein said media element characteristics comprise at least one of rhythm time, rhythm type, style, vocal voices, flavor, mood, emotion, density, weight, consonance, melodic movement, tempo, or rhythm activity.

17. The system as recited in claim 13, wherein said data store stores said media elements in a roll-up table having at least one index that is accessible by participating users.

18. The system as recited in claim 13, wherein said scoring system compares a reference media entity to a library of media entities, said reference entity being compared to said library of media entities using media entity characteristics to calculate said matched scores, said scoring system associating media entities as closely related if said calculated matched score is below a pre-defined threshold score.

19. The system recited in claim 13, further comprising a communications network cooperating with said scoring system and said data store to communicate data representative of matched media entities and media entities to participating users, said communications network comprising any of a fixed-wire public network, a wireless public network, a fixed wire private network, and a wireless private network.

20. A method to enhance the experience of participating users of a media entity distribution system comprising the steps of:

providing a library of media entities;

applying media entity matching processes on said library of media elements to create associations between the media entities of said library using inherent characteristics of said media elements, wherein said media entity matching processes employ at least one of a score-based matching or node counting matching;

storing for distribution said associations in a roll-up table having one or more indices, said roll-up table provided to said participating users.