WO2008117232A2 - Apparatus for creating a multimedia file list - Google Patents

Apparatus for creating a multimedia file list Download PDF

Info

Publication number
WO2008117232A2
WO2008117232A2 PCT/IB2008/051096 IB2008051096W WO2008117232A2 WO 2008117232 A2 WO2008117232 A2 WO 2008117232A2 IB 2008051096 W IB2008051096 W IB 2008051096W WO 2008117232 A2 WO2008117232 A2 WO 2008117232A2
Authority
WO
WIPO (PCT)
Prior art keywords
multimedia file
multimedia
code
characteristic
multimedia files
Prior art date
Application number
PCT/IB2008/051096
Other languages
French (fr)
Other versions
WO2008117232A3 (en
Inventor
Fabio Vignoli
Original Assignee
Koninklijke Philips Electronics N.V.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips Electronics N.V. filed Critical Koninklijke Philips Electronics N.V.
Publication of WO2008117232A2 publication Critical patent/WO2008117232A2/en
Publication of WO2008117232A3 publication Critical patent/WO2008117232A3/en

Links

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/10Indexing; Addressing; Timing or synchronising; Measuring tape travel
    • G11B27/102Programmed access in sequence to addressed parts of tracks of operating record carriers
    • G11B27/105Programmed access in sequence to addressed parts of tracks of operating record carriers of operating discs
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28Databases characterised by their database models, e.g. relational or object models
    • G06F16/284Relational databases
    • G06F16/285Clustering or classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35Clustering; Classification
    • G06F16/353Clustering; Classification into predefined classes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/63Querying
    • G06F16/638Presentation of query results
    • G06F16/639Presentation of query results using playlists
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/683Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73Querying
    • G06F16/735Filtering based on additional data, e.g. user or group profiles
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/02Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
    • G11B27/031Electronic editing of digitised analogue information signals, e.g. audio or video signals
    • G11B27/034Electronic editing of digitised analogue information signals, e.g. audio or video signals on discs

Definitions

  • the invention relates to an apparatus for creating a list of multimedia files, to a device comprising the apparatus, to a method for creating a list of multimedia files, to a computer program product for performing the steps of the method and to a medium for storing and comprising the computer program product.
  • Examples of such an apparatus are processors, and examples of such a device are consumer products and non-consumer products. Examples of such a method are services.
  • BACKGROUND OF THE INVENTION WO 2005/071569 discloses a creation of a play- list based on artist and song similarities.
  • an artist selector selects an artist from a list provided by this similarity module.
  • a distance threshold level is set to include distantly similar artists.
  • a song selector a particular song by the selected artist is selected.
  • clustering is used for grouping items having similar characteristics. Each item in a collection is associated with a corresponding point in an N-dimensional space.
  • Attributes of a song determine the corresponding point in the N-dimensional space.
  • This N- dimensional space is partitioned into clusters, and an item is a member of a cluster whose center of gravity is closest to the item.
  • Further objects of the invention are to provide a device, to provide a method, to provide a computer program product and to provide a medium for creating a list of multimedia files in a different way.
  • a first aspect of the invention provides an apparatus for creating a list of multimedia files, the apparatus comprising
  • a memory for storing at least a feature of a first multimedia file and for storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, a comparator for comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and an allocator for, in response to a first comparison result, allocating the first code to the third multimedia file and for, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code.
  • a memory stores at least a feature of a multimedia file and may further store a further part of a multimedia file or an entire multimedia file.
  • a feature is for example an identification such as a title or a name of the multimedia file.
  • a multimedia file comprises audio and/or video such as a song or a movie.
  • a multimedia file is defined by a code such as a classification code.
  • a comparator compares a characteristic of a not yet classified multimedia file with characteristics of differently classified multimedia files and an allocator classifies the not yet classified multimedia file in response to comparison results.
  • a first (second, third) multimedia file may be a first (second, third) set of multimedia files
  • a feature may be a set of features
  • a first (second) code may be a first (second) set of codes
  • a characteristic may be a set of characteristics
  • a first (second) comparison result may be a first (second) set of comparison results
  • further multimedia files, further codes and further comparison results are not to be excluded.
  • an apparatus is defined by the list of multimedia files belonging to a group of multimedia files and the first code defining a first sub-group of the group and the second code defining a second sub-group of the group, the first and second sub-groups being different groups.
  • a group of multimedia files is classified into different sub-groups of the group.
  • an apparatus is defined by the group of multimedia files being songs from an artist or a band or a composer or an entity and the first and second sub-groups being typical and non-typical songs from the artist or the band or the composer or the entity.
  • typical songs such as common songs
  • non-typical songs such as non-common songs
  • an apparatus is defined by the characteristic being at least one of a tempo, a rhythm, a timbre, an instrumentation, a beat strength and a multimedia characteristic.
  • an apparatus comprising a calculator for calculating a first distance between the characteristic of the third multimedia file and the characteristic of the first multimedia file and for calculating a second distance between the characteristic of the third multimedia file and the characteristic of the second multimedia file, the first comparison result defining the first distance to be respectively smaller than or equal to the second distance and the second comparison result defining the first distance to be respectively equal to or larger than the second distance.
  • distances between characteristics of a not yet classified multimedia file and classified multimedia files are used for classifying the not yet classified multimedia file.
  • an apparatus is defined by the first code being allocated to the first multimedia file and the second code being allocated to the second multimedia file in response to a user action or in response to a search engine action.
  • a user may classify at least first and second multimedia files, after which classification the apparatus may classify a third multimedia file etc.
  • a search engine Via a search engine, a search in a local database or in a non-local database or in a global database will result in a search result that may be used for classifying at least first and second multimedia files, after which classification the apparatus may classify a third multimedia file etc.
  • a second aspect of the invention provides a device comprising the apparatus and further comprising a first port to be coupled to a source and/or further comprising a second port to be coupled to a destination.
  • the source such as a database or an interface to be coupled to a database may form part of the device or not.
  • the destination such as a loudspeaker or a display may form part of the device or not.
  • a third aspect of the invention provides a method for creating a list of multimedia files, the method comprising the steps of storing at least a feature of a first multimedia file and storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and in response to a first comparison result, allocating the first code to the third multimedia file and, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code.
  • a fourth aspect of the invention provides a computer program product for performing the steps of the method and a fifth aspect of the invention provides a medium for storing and comprising the computer program product.
  • Embodiments of the device and the method and the computer program product and the medium correspond with the embodiments of the apparatus.
  • An insight might be, that previous classifications of first and second multimedia files may be used for a future classification of a third multimedia file.
  • a basic idea might be, that first and second multimedia files are to be classified in view of each other and that a third multimedia file is to be classified in view of the classifications of the first and second multimedia files.
  • the invention solves a problem to provide an apparatus for creating a list of multimedia files in a different way.
  • the invention is further advantageous in that, after first and second numbers of multimedia files have been classified into at least two classes in a relatively subjective way (for example via a user action) or in a relatively objective way (for example via a search machine action), a third number of multimedia files can be classified automatically.
  • Fig. 1 shows a schematic block diagram of an apparatus according to the invention
  • Fig. 2 shows a schematic block diagram of a device according to the invention comprising an apparatus according to the invention.
  • the apparatus 1 shown schematically in the Fig. 1 for creating a list of multimedia files comprises a memory 2 for storing at least a feature Al of a first multimedia file Bl and for storing at least a feature A2 of a second multimedia file B2.
  • These features Al and A2 are for example stored in a column A of the memory 2.
  • a column B of the memory for example further parts of the multimedia files or the entire multimedia files Bl and B2 may be stored.
  • a column C of the memory 2 for example a first code Cl defining the first multimedia file Bl is stored and a second code C2 defining the second multimedia file B2 is stored.
  • a column D of the memory 2 for example a characteristic Dl of the first multimedia file Bl is stored and a characteristic D2 of the second multimedia file B2 is stored.
  • the apparatus 1 further comprises a module 10 with a comparator 3 for comparing a characteristic D3 of a third multimedia file B3 with a characteristic Dl of the first multimedia file Bl and with a characteristic D2 of the second multimedia file B2.
  • the list of multimedia files comprises multimedia files defined by a same code (or code value).
  • the module 10 for example further comprises a controller 6 coupled to the comparator 3 and to the allocator 4 and to a memory interface 7.
  • the memory interface 7 is further coupled to the memory 2 and to a data in/output 9, and the controller 6 is further coupled to a control in/output 8.
  • a feature is for example an identification such as a title or a name of the multimedia file.
  • a multimedia file comprises audio and/or video such as a song or a movie.
  • a code such as a classification code defines a multimedia file.
  • the comparator 3 compares a characteristic D3 of a not yet classified multimedia file B3 ("not yet classified” means that C3 has not yet been determined) with characteristics D1,D2 of differently classified multimedia files B1,B2 ("differently classified” means that Cl and C2 have got different values) and the allocator 4 classifies the not yet classified multimedia file B3 in response to comparison results.
  • the list of multimedia files belongs to a group of multimedia files such as songs from an artist or a band or a composer or an entity or such as movies from a director or a producer and the first code defines a first sub-group of the group and the second code defines a second sub-group of the group.
  • the first sub-group for example comprises typical songs for the artist or the band or the composer or the entity or typical movies for the director or the producer and second sub-group for example comprises non-typical songs for the artist or the band or the composer or the entity or non-typical movies for the director or the producer.
  • Typical may be common or usual, and non-typical may be non-common or non-usual.
  • the characteristic is at least one of a tempo, a rhythm, a timbre, an instrumentation, a beat strength and a multimedia characteristic in case the multimedia file is a song.
  • Other characteristics may be defined in case the multimedia file is a movie.
  • the comparator 3 comprises a calculator 5 for calculating a first distance between the characteristic D3 of the third multimedia file B3 and the characteristic Dl of the first multimedia file Bl and for calculating a second distance between the characteristic D3 of the third multimedia file B3 and the characteristic D2 of the second multimedia file B2.
  • the first comparison result may define the first distance to be respectively smaller than or equal to the second distance and the second comparison result may define the first distance to be respectively equal to or larger than the second distance.
  • the first code is allocated to the first multimedia file and the second code is allocated to the second multimedia file in response to a user action.
  • the first and second multimedia files are classified in a relatively subjective way (by letting a user make a selection), and the third multimedia file is then classified automatically by using the classifications of the first and second multimedia files.
  • the first code is allocated to the first multimedia file and the second code is allocated to the second multimedia file in response to a search engine action.
  • the first and second multimedia files are classified in a relatively objective way (by letting a search engine generate search results that for example indicate per multimedia file a number of web sites that cite this multimedia file), and the third multimedia file is then classified automatically by using the classifications of the first and second multimedia files.
  • Further options are not to be excluded, such as using popularity information (by for example letting a group of users vote) or statistics (by for example retrieving chart statistics or by for example converting commercial results).
  • the device 11 shown schematically in the Fig. 2 comprises the apparatus 1 and further comprises an input/output interface 12 coupled to the control in/output 8 and to the data in/output 9 of the apparatus 1.
  • the input/output interface 12 is further coupled to a first port 13 that is to be coupled to a source 21 and is further coupled to a second port 14 that is to be coupled to a destination 22.
  • the source 21 may for example comprise a database or an interface to be coupled to a database and will usually be located outside the device 11. Alternatively such a source 21 may form part of the device 11.
  • the destination 22 may for example comprise a loudspeaker or a display and may form part of the device 11 or not.
  • the comparator 3 may compare characteristics in an analog way or in a digital way and may be realized through hardware or through software or through a mixture of both.
  • the allocator 4 may allocate codes by generating codes in response to comparison results or by converting comparison results into codes and may be realized through hardware or through software or through a mixture of both.
  • the comparator 3 and the allocator 4 may be integrated into the controller 6.
  • the module 10 may be a processor for running software functions for realizing one or more of said comparator 3, allocator 4, calculator 5 and controller 6, whereby the software functions could be stored in the memory 2.
  • the module 10 may alternatively be a processor with dedicated hardware for realizing one or more of said comparator 3, allocator 4, calculator 5 and controller 6, or the module 10 may be dedicated hardware, without having excluded further options.
  • the memory interface 7 is an embodiment only and may for example be replaced by a bus or a switch. Alternatively, separate data connections and control connections are to be used between the controller 6 and the memory 2.
  • the memory 2 will at least store one feature per multimedia file in the column A and, after being classified, one code per multimedia file in the column C.
  • the memory 2 may further store further parts of the multimedia files or the entire multimedia files in the column B, alternatively these further parts or entire multimedia files may be downloaded and/or retrieved when necessary.
  • the memory 2 may further store characteristics of the multimedia files in the column D, alternatively these characteristics may be downloaded and/or retrieved when necessary.
  • the memory 2 may further store other features, other characteristics and/or other codes per multimedia file, for example in the columns G and H and/or in columns not shown.
  • the content of the rows may be placed in the columns and vice versa.
  • a memory based on rows and columns other kinds of memories may be used, such as a memory based on fields and/or tables and/or addresses.
  • Music augmentation may be employed to obtain one or more lists of most famous (most typical or most common) songs from a specific artists (e.g. songs that are mentioned more frequently on the web) and of least typical (uncommon) or non-typical (non-common) songs.
  • the result is analyzed to extract relevant features (e.g. album names) and each song is analyzed to extract characteristics such as tempo/rhythm, timbre etc.
  • relevant features e.g. album names
  • each song is analyzed to extract characteristics such as tempo/rhythm, timbre etc.
  • a characteristic or classifier is trained to discriminate between typical and non-typical songs.
  • the songs in a collection e.g.
  • a service with hundreds of thousands of songs may be analyzed and a flag to discriminate between typical and non-typical songs for each artist can be set.
  • the flag can be used to generate a play- list based on typical songs / non-typical songs at a request of one or more users or as a part of another service.
  • a play-list generation tool is able to create a play-list based on the most or the least typical songs recorded by for example a specific artist.
  • a music download server has available about 1.000.000 songs of many artists (and its corresponding audio).
  • One possible approach, without excluding other approaches, may be the following: 1) By using web searches is possible to create a set S+ of the most frequently mentioned songs of an artist.
  • Information about every song in the set S+ is obtained (e.g. album name, album year, etc). This data can be obtained by cross checking the title of the song with a database of all songs from the specific artist. 3) The audio of the songs in set S+ is analyzed to extract relevant characteristics such as tempo/rhythm, timbre, instrumentation, beat strength.
  • An algorithm may compute the distances between all clusters in the two groups of clusters. There will be songs close to the typical song clusters and songs more far away from the typical song clusters.
  • a flag can be set in a database to distinguish between typical songs from an artist and non-typical songs from the artist.
  • a user interface may let a user create a play-list by specifying: 1) artists, and 2) typical / non-typical songs from the chosen artists.
  • the system can generate a play-list based on the user requests.
  • the invention may take advantage of the web to generate a list of most typical songs for an artist. This may give an advantage with respect to a DSP method.
  • the list of web-generated songs contains the most prototypical songs for that artist. To find these songs a simple way could be based on the following: For each song of an artist, search the web for artist + title (of the song), then check the number of pages returned. The songs with the most pages returned may be the most typical for that artist.
  • a first overview is obtained of the most famous songs from an artist (e.g. songs that are mentioned more frequently on the web).
  • a second overview is obtained of least famous songs.
  • the first overview may be analyzed to extract relevant characteristics, e.g. tempo/rhythm, timbre etc.
  • the songs in the first overview are clustered and distances between each song of the first and second overviews and the cluster are calculated.
  • Fifthly "typical” songs are close to the cluster, and "non- typical” songs are remote from the cluster.
  • Sixthly a flag "typical” or “non-typical” is set in a song database for each song based on its distance to the cluster.
  • apparatuses 1 for creating lists of multimedia files are provided with memories 2 for storing features of first and second multimedia files defined by first and second codes and with comparators 3 for comparing a characteristic of a third multimedia file with characteristics of the first and second multimedia files and with allocators 4 for, in response to first / second comparison results, allocating the first / second code to the third multimedia file.
  • the list of multimedia files comprises multimedia files defined by same codes. Features may be identifications, titles, names.
  • Multimedia files comprise audio, video, songs, movies.
  • Codes may be classification codes. Characteristics may be tempos, rhythms, timbres, instrumentations, beat strengths.
  • the comparators 3 may comprise calculators 5 for calculating distances between characteristics of the multimedia files.
  • the first and second codes for the first and second multimedia files are generated in response to user actions or search engine actions.
  • a computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Abstract

Apparatuses (1) for creating lists of multimedia files are provided with memories (2) for storing features of first and second multimedia files defined by first and second codes and with comparators (3) for comparing a characteristic of a third multimedia file with characteristics of the first and second multimedia files and with allocators (4) for, in response to first / second comparison results, allocating the first / second code to the third multimedia file. The list of multimedia files comprises multimedia files defined by same codes. Features may be identifications, titles, names. Multimedia files comprise audio, video, songs, movies. Codes may be classification codes. Characteristics may be tempos, rhythms, timbres, instrumentations, beat strengths. The comparators (3) may comprise calculators (5) for calculating distances between characteristics of the multimedia files. The first and second codes for the first and second multimedia files are generated in response to user actions or search engine actions.

Description

Apparatus for creating a multimedia file list
FIELD OF THE INVENTION
The invention relates to an apparatus for creating a list of multimedia files, to a device comprising the apparatus, to a method for creating a list of multimedia files, to a computer program product for performing the steps of the method and to a medium for storing and comprising the computer program product.
Examples of such an apparatus are processors, and examples of such a device are consumer products and non-consumer products. Examples of such a method are services.
BACKGROUND OF THE INVENTION WO 2005/071569 discloses a creation of a play- list based on artist and song similarities. By using a similarity module, an artist selector selects an artist from a list provided by this similarity module. Thereto, a distance threshold level is set to include distantly similar artists. By using a song selector, a particular song by the selected artist is selected. Thereto, clustering is used for grouping items having similar characteristics. Each item in a collection is associated with a corresponding point in an N-dimensional space.
Attributes of a song determine the corresponding point in the N-dimensional space. This N- dimensional space is partitioned into clusters, and an item is a member of a cluster whose center of gravity is closest to the item.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an apparatus for creating a list of multimedia files in a different way.
Further objects of the invention are to provide a device, to provide a method, to provide a computer program product and to provide a medium for creating a list of multimedia files in a different way.
A first aspect of the invention provides an apparatus for creating a list of multimedia files, the apparatus comprising
- a memory for storing at least a feature of a first multimedia file and for storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, a comparator for comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and an allocator for, in response to a first comparison result, allocating the first code to the third multimedia file and for, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code.
A memory stores at least a feature of a multimedia file and may further store a further part of a multimedia file or an entire multimedia file. A feature is for example an identification such as a title or a name of the multimedia file. A multimedia file comprises audio and/or video such as a song or a movie. A multimedia file is defined by a code such as a classification code. A comparator compares a characteristic of a not yet classified multimedia file with characteristics of differently classified multimedia files and an allocator classifies the not yet classified multimedia file in response to comparison results.
A first (second, third) multimedia file may be a first (second, third) set of multimedia files, a feature may be a set of features, a first (second) code may be a first (second) set of codes, a characteristic may be a set of characteristics, a first (second) comparison result may be a first (second) set of comparison results, and further multimedia files, further codes and further comparison results are not to be excluded.
According to an embodiment, an apparatus is defined by the list of multimedia files belonging to a group of multimedia files and the first code defining a first sub-group of the group and the second code defining a second sub-group of the group, the first and second sub-groups being different groups. Preferably, a group of multimedia files is classified into different sub-groups of the group.
According to an embodiment, an apparatus is defined by the group of multimedia files being songs from an artist or a band or a composer or an entity and the first and second sub-groups being typical and non-typical songs from the artist or the band or the composer or the entity. Preferably, typical songs, such as common songs, and non-typical songs, such as non-common songs, are different classes of songs of an artist or a band or a composer or an entity. According to an embodiment, an apparatus is defined by the characteristic being at least one of a tempo, a rhythm, a timbre, an instrumentation, a beat strength and a multimedia characteristic.
According to an embodiment, an apparatus is defined by the comparator comprising a calculator for calculating a first distance between the characteristic of the third multimedia file and the characteristic of the first multimedia file and for calculating a second distance between the characteristic of the third multimedia file and the characteristic of the second multimedia file, the first comparison result defining the first distance to be respectively smaller than or equal to the second distance and the second comparison result defining the first distance to be respectively equal to or larger than the second distance. Preferably, distances between characteristics of a not yet classified multimedia file and classified multimedia files are used for classifying the not yet classified multimedia file.
According to an embodiment, an apparatus is defined by the first code being allocated to the first multimedia file and the second code being allocated to the second multimedia file in response to a user action or in response to a search engine action. Via a user action, a user may classify at least first and second multimedia files, after which classification the apparatus may classify a third multimedia file etc. Via a search engine, a search in a local database or in a non-local database or in a global database will result in a search result that may be used for classifying at least first and second multimedia files, after which classification the apparatus may classify a third multimedia file etc.
A second aspect of the invention provides a device comprising the apparatus and further comprising a first port to be coupled to a source and/or further comprising a second port to be coupled to a destination. The source such as a database or an interface to be coupled to a database may form part of the device or not. The destination such as a loudspeaker or a display may form part of the device or not.
A third aspect of the invention provides a method for creating a list of multimedia files, the method comprising the steps of storing at least a feature of a first multimedia file and storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and in response to a first comparison result, allocating the first code to the third multimedia file and, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code. A fourth aspect of the invention provides a computer program product for performing the steps of the method and a fifth aspect of the invention provides a medium for storing and comprising the computer program product.
Embodiments of the device and the method and the computer program product and the medium correspond with the embodiments of the apparatus. An insight might be, that previous classifications of first and second multimedia files may be used for a future classification of a third multimedia file. A basic idea might be, that first and second multimedia files are to be classified in view of each other and that a third multimedia file is to be classified in view of the classifications of the first and second multimedia files. The invention solves a problem to provide an apparatus for creating a list of multimedia files in a different way. The invention is further advantageous in that, after first and second numbers of multimedia files have been classified into at least two classes in a relatively subjective way (for example via a user action) or in a relatively objective way (for example via a search machine action), a third number of multimedia files can be classified automatically.
These and other aspects of the invention are apparent from and will be elucidated with reference to the embodiments described hereinafter.
BRIEF DESCRIPTION OF THE DRAWINGS In the drawings:
Fig. 1 shows a schematic block diagram of an apparatus according to the invention, and
Fig. 2 shows a schematic block diagram of a device according to the invention comprising an apparatus according to the invention.
DETAILED DESCRIPTION
The apparatus 1 shown schematically in the Fig. 1 for creating a list of multimedia files comprises a memory 2 for storing at least a feature Al of a first multimedia file Bl and for storing at least a feature A2 of a second multimedia file B2. These features Al and A2 are for example stored in a column A of the memory 2. In a column B of the memory 2, for example further parts of the multimedia files or the entire multimedia files Bl and B2 may be stored. In a column C of the memory 2, for example a first code Cl defining the first multimedia file Bl is stored and a second code C2 defining the second multimedia file B2 is stored. In a column D of the memory 2, for example a characteristic Dl of the first multimedia file Bl is stored and a characteristic D2 of the second multimedia file B2 is stored.
The apparatus 1 further comprises a module 10 with a comparator 3 for comparing a characteristic D3 of a third multimedia file B3 with a characteristic Dl of the first multimedia file Bl and with a characteristic D2 of the second multimedia file B2. The module 10 further comprises an allocator 4 for, in response to a first comparison result, allocating the first code Cl (or code value) to the third multimedia file B3 (in other words, in that case, C3 = Cl) and for, in response to a second comparison result, allocating the second code C2 (or code value) to the third multimedia file B3 (in other words, in that case, C3 = C2). The list of multimedia files comprises multimedia files defined by a same code (or code value).
Thereto, the module 10 for example further comprises a controller 6 coupled to the comparator 3 and to the allocator 4 and to a memory interface 7. The memory interface 7 is further coupled to the memory 2 and to a data in/output 9, and the controller 6 is further coupled to a control in/output 8.
A feature is for example an identification such as a title or a name of the multimedia file. A multimedia file comprises audio and/or video such as a song or a movie. A code such as a classification code defines a multimedia file. In this case, the comparator 3 compares a characteristic D3 of a not yet classified multimedia file B3 ("not yet classified" means that C3 has not yet been determined) with characteristics D1,D2 of differently classified multimedia files B1,B2 ("differently classified" means that Cl and C2 have got different values) and the allocator 4 classifies the not yet classified multimedia file B3 in response to comparison results.
Preferably, the list of multimedia files belongs to a group of multimedia files such as songs from an artist or a band or a composer or an entity or such as movies from a director or a producer and the first code defines a first sub-group of the group and the second code defines a second sub-group of the group. The first sub-group for example comprises typical songs for the artist or the band or the composer or the entity or typical movies for the director or the producer and second sub-group for example comprises non-typical songs for the artist or the band or the composer or the entity or non-typical movies for the director or the producer. Typical may be common or usual, and non-typical may be non-common or non-usual.
Preferably, the characteristic is at least one of a tempo, a rhythm, a timbre, an instrumentation, a beat strength and a multimedia characteristic in case the multimedia file is a song. Other characteristics may be defined in case the multimedia file is a movie.
Preferably, the comparator 3 comprises a calculator 5 for calculating a first distance between the characteristic D3 of the third multimedia file B3 and the characteristic Dl of the first multimedia file Bl and for calculating a second distance between the characteristic D3 of the third multimedia file B3 and the characteristic D2 of the second multimedia file B2. The first comparison result may define the first distance to be respectively smaller than or equal to the second distance and the second comparison result may define the first distance to be respectively equal to or larger than the second distance. Preferably, according to a first option, the first code is allocated to the first multimedia file and the second code is allocated to the second multimedia file in response to a user action. In this case, the first and second multimedia files are classified in a relatively subjective way (by letting a user make a selection), and the third multimedia file is then classified automatically by using the classifications of the first and second multimedia files. According to a second option, the first code is allocated to the first multimedia file and the second code is allocated to the second multimedia file in response to a search engine action. In this case, the first and second multimedia files are classified in a relatively objective way (by letting a search engine generate search results that for example indicate per multimedia file a number of web sites that cite this multimedia file), and the third multimedia file is then classified automatically by using the classifications of the first and second multimedia files. Further options are not to be excluded, such as using popularity information (by for example letting a group of users vote) or statistics (by for example retrieving chart statistics or by for example converting commercial results).
The device 11 shown schematically in the Fig. 2 comprises the apparatus 1 and further comprises an input/output interface 12 coupled to the control in/output 8 and to the data in/output 9 of the apparatus 1. The input/output interface 12 is further coupled to a first port 13 that is to be coupled to a source 21 and is further coupled to a second port 14 that is to be coupled to a destination 22. The source 21 may for example comprise a database or an interface to be coupled to a database and will usually be located outside the device 11. Alternatively such a source 21 may form part of the device 11. The destination 22 may for example comprise a loudspeaker or a display and may form part of the device 11 or not.
The comparator 3 may compare characteristics in an analog way or in a digital way and may be realized through hardware or through software or through a mixture of both. The allocator 4 may allocate codes by generating codes in response to comparison results or by converting comparison results into codes and may be realized through hardware or through software or through a mixture of both. The comparator 3 and the allocator 4 may be integrated into the controller 6. The module 10 may be a processor for running software functions for realizing one or more of said comparator 3, allocator 4, calculator 5 and controller 6, whereby the software functions could be stored in the memory 2. The module 10 may alternatively be a processor with dedicated hardware for realizing one or more of said comparator 3, allocator 4, calculator 5 and controller 6, or the module 10 may be dedicated hardware, without having excluded further options.
The memory interface 7 is an embodiment only and may for example be replaced by a bus or a switch. Alternatively, separate data connections and control connections are to be used between the controller 6 and the memory 2. The memory 2 will at least store one feature per multimedia file in the column A and, after being classified, one code per multimedia file in the column C. The memory 2 may further store further parts of the multimedia files or the entire multimedia files in the column B, alternatively these further parts or entire multimedia files may be downloaded and/or retrieved when necessary. The memory 2 may further store characteristics of the multimedia files in the column D, alternatively these characteristics may be downloaded and/or retrieved when necessary. The memory 2 may further store other features, other characteristics and/or other codes per multimedia file, for example in the columns G and H and/or in columns not shown. The content of the rows may be placed in the columns and vice versa. Instead of a memory based on rows and columns, other kinds of memories may be used, such as a memory based on fields and/or tables and/or addresses.
In other words, for example a way to create a play- list of typical / non-typical songs from a specific artist is described. Music augmentation may be employed to obtain one or more lists of most famous (most typical or most common) songs from a specific artists (e.g. songs that are mentioned more frequently on the web) and of least typical (uncommon) or non-typical (non-common) songs. The result is analyzed to extract relevant features (e.g. album names) and each song is analyzed to extract characteristics such as tempo/rhythm, timbre etc. From an overview of the most typical songs, for example a characteristic or classifier is trained to discriminate between typical and non-typical songs. The songs in a collection (e.g. a service with hundreds of thousands of songs) may be analyzed and a flag to discriminate between typical and non-typical songs for each artist can be set. The flag can be used to generate a play- list based on typical songs / non-typical songs at a request of one or more users or as a part of another service.
So, a play-list generation tool is able to create a play-list based on the most or the least typical songs recorded by for example a specific artist. A music download server has available about 1.000.000 songs of many artists (and its corresponding audio). One possible approach, without excluding other approaches, may be the following: 1) By using web searches is possible to create a set S+ of the most frequently mentioned songs of an artist.
2) Information about every song in the set S+ is obtained (e.g. album name, album year, etc). This data can be obtained by cross checking the title of the song with a database of all songs from the specific artist. 3) The audio of the songs in set S+ is analyzed to extract relevant characteristics such as tempo/rhythm, timbre, instrumentation, beat strength.
4) The other music of the artist is also analyzed to extract relevant characteristics.
5) The music in the set S+ is clustered and the other music is also clustered.
6) An algorithm may compute the distances between all clusters in the two groups of clusters. There will be songs close to the typical song clusters and songs more far away from the typical song clusters.
With the following pre-processing a flag can be set in a database to distinguish between typical songs from an artist and non-typical songs from the artist. A user interface may let a user create a play-list by specifying: 1) artists, and 2) typical / non-typical songs from the chosen artists. The system can generate a play-list based on the user requests.
The invention may take advantage of the web to generate a list of most typical songs for an artist. This may give an advantage with respect to a DSP method. The list of web-generated songs contains the most prototypical songs for that artist. To find these songs a simple way could be based on the following: For each song of an artist, search the web for artist + title (of the song), then check the number of pages returned. The songs with the most pages returned may be the most typical for that artist.
In yet other words, to create a play- list of typical / non-typical songs for a specific artist, firstly a first overview is obtained of the most famous songs from an artist (e.g. songs that are mentioned more frequently on the web). Secondly a second overview is obtained of least famous songs. Thirdly the first overview may be analyzed to extract relevant characteristics, e.g. tempo/rhythm, timbre etc. Fourthly the songs in the first overview are clustered and distances between each song of the first and second overviews and the cluster are calculated. Fifthly "typical" songs are close to the cluster, and "non- typical" songs are remote from the cluster. Sixthly a flag "typical" or "non-typical" is set in a song database for each song based on its distance to the cluster. Seventhly the flag may be used to generate a play- list based on typical songs / uncommon songs at the request of the users or as a part of another service. A typical song is for example defined as most frequently retrieved on the web (number of pages returned in a search engine - popularity). Summarizing, apparatuses 1 for creating lists of multimedia files are provided with memories 2 for storing features of first and second multimedia files defined by first and second codes and with comparators 3 for comparing a characteristic of a third multimedia file with characteristics of the first and second multimedia files and with allocators 4 for, in response to first / second comparison results, allocating the first / second code to the third multimedia file. The list of multimedia files comprises multimedia files defined by same codes. Features may be identifications, titles, names. Multimedia files comprise audio, video, songs, movies. Codes may be classification codes. Characteristics may be tempos, rhythms, timbres, instrumentations, beat strengths. The comparators 3 may comprise calculators 5 for calculating distances between characteristics of the multimedia files. The first and second codes for the first and second multimedia files are generated in response to user actions or search engine actions.
While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:
1. An apparatus (1) for creating a list of multimedia files, the apparatus (1) comprising a memory (2) for storing at least a feature of a first multimedia file and for storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, a comparator (3) for comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and an allocator (4) for, in response to a first comparison result, allocating the first code to the third multimedia file and for, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code.
2. An apparatus (1) as claimed in claim 1, the list of multimedia files belonging to a group of multimedia files and the first code defining a first sub-group of the group and the second code defining a second sub-group of the group, the first and second sub-groups being different groups.
3. An apparatus (1) as claimed in claim 2, the group of multimedia files being songs from an artist or a band or a composer or an entity and the first and second sub-groups being typical and non-typical songs from the artist or the band or the composer or the entity.
4. An apparatus (1) as claimed in claim 1, the characteristic being at least one of a tempo, a rhythm, a timbre, an instrumentation, a beat strength and a multimedia characteristic.
5. An apparatus (1) as claimed in claim 1, the comparator (3) comprising a calculator (5) for calculating a first distance between the characteristic of the third multimedia file and the characteristic of the first multimedia file and for calculating a second distance between the characteristic of the third multimedia file and the characteristic of the second multimedia file, the first comparison result defining the first distance to be respectively smaller than or equal to the second distance and the second comparison result defining the first distance to be respectively equal to or larger than the second distance.
6. An apparatus (1) as claimed in claim 1, the first code being allocated to the first multimedia file and the second code being allocated to the second multimedia file in response to a user action or in response to a search engine action.
7. A device comprising the apparatus (1) as defined in claim 1 and further comprising a first port (13) to be coupled to a source (21) and/or further comprising a second port (14) to be coupled to a destination (22).
8. A method for creating a list of multimedia files, the method comprising the steps of storing at least a feature of a first multimedia file and storing at least a feature of a second multimedia file, which first multimedia file is defined by a first code and which second multimedia file is defined by a second code, the first and second multimedia files being different multimedia files and the first and second codes being different codes, comparing a characteristic of a third multimedia file with a characteristic of the first multimedia file and with a characteristic of the second multimedia file, and - in response to a first comparison result, allocating the first code to the third multimedia file and, in response to a second comparison result, allocating the second code to the third multimedia file, the first and second comparison results being different comparison results and the list of multimedia files comprising multimedia files defined by a same code.
9. A computer program product for performing the steps of the method as claimed in claim 8.
10. A medium for storing and comprising the computer program product as claimed in claim 9.
PCT/IB2008/051096 2007-03-27 2008-03-25 Apparatus for creating a multimedia file list WO2008117232A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP07105008 2007-03-27
EP07105008.2 2007-03-27

Publications (2)

Publication Number Publication Date
WO2008117232A2 true WO2008117232A2 (en) 2008-10-02
WO2008117232A3 WO2008117232A3 (en) 2008-11-20

Family

ID=39673005

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2008/051096 WO2008117232A2 (en) 2007-03-27 2008-03-25 Apparatus for creating a multimedia file list

Country Status (1)

Country Link
WO (1) WO2008117232A2 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012134854A1 (en) * 2011-03-29 2012-10-04 Bose Corporation Multimedia analysis and cluster formation

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5918223A (en) * 1996-07-22 1999-06-29 Muscle Fish Method and article of manufacture for content-based analysis, storage, retrieval, and segmentation of audio information
EP1073272A1 (en) * 1999-02-15 2001-01-31 Sony Corporation Signal processing method and video/audio processing device
WO2001020483A2 (en) * 1999-09-14 2001-03-22 Cantametrix, Inc. Finding perceptually similar music with precomputation
EP1437738A1 (en) * 2003-01-06 2004-07-14 Thomson Licensing S.A. Method for creating and accessing a menu for audio content without using a display
US20050092165A1 (en) * 2000-07-14 2005-05-05 Microsoft Corporation System and methods for providing automatic classification of media entities according to tempo
WO2005071569A1 (en) * 2004-01-20 2005-08-04 Koninklijke Philips Electronics, N.V. Playlist based on artist and song similarity
EP1750269A2 (en) * 2005-08-03 2007-02-07 Delphi Technologies, Inc. Reducing genre metadata indicating the type of music

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5918223A (en) * 1996-07-22 1999-06-29 Muscle Fish Method and article of manufacture for content-based analysis, storage, retrieval, and segmentation of audio information
EP1073272A1 (en) * 1999-02-15 2001-01-31 Sony Corporation Signal processing method and video/audio processing device
WO2001020483A2 (en) * 1999-09-14 2001-03-22 Cantametrix, Inc. Finding perceptually similar music with precomputation
US20050092165A1 (en) * 2000-07-14 2005-05-05 Microsoft Corporation System and methods for providing automatic classification of media entities according to tempo
EP1437738A1 (en) * 2003-01-06 2004-07-14 Thomson Licensing S.A. Method for creating and accessing a menu for audio content without using a display
WO2005071569A1 (en) * 2004-01-20 2005-08-04 Koninklijke Philips Electronics, N.V. Playlist based on artist and song similarity
EP1750269A2 (en) * 2005-08-03 2007-02-07 Delphi Technologies, Inc. Reducing genre metadata indicating the type of music

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JAIN A K ET AL: "Data clustering: a review" 1 September 1999 (1999-09-01), ACM COMPUTING SURVEYS, ACM, NEW YORK, NY, US, US, PAGE(S) 264 - 323 , XP002165131 ISSN: 0360-0300 abstract *
PAUWS S ET AL: "PATS: Realization and User Evaluation of an Automatic Playlist Generator" 13 October 2002 (2002-10-13), PROCEEDINGS ANNUAL INTERNATIONAL SYMPOSIUM ON MUSIC INFORMATIONRETRIEVAL, XX, XX, PAGE(S) 1 - 9 , XP002325429 page 3 *
WELSH M ET AL: "QUERYING LARGE COLLECTIONS OF MUSIC FOR SIMILARITY" 1 November 1999 (1999-11-01), UC BERKELEY TECHNICAL REPORT, XX, XX , XP008027813 the whole document *
WOLD E., BLUM T., KEISLAR D., WHEATON J.: "Content-Based Classification, Search, and Retrieval of Audio" IEEE MULTIMEDIA, [Online] vol. 3, no. 3, 1996, pages 27-36, XP002492745 Retrieved from the Internet: URL:www.musclefish.com> [retrieved on 2008-08-15] *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012134854A1 (en) * 2011-03-29 2012-10-04 Bose Corporation Multimedia analysis and cluster formation

Also Published As

Publication number Publication date
WO2008117232A3 (en) 2008-11-20

Similar Documents

Publication Publication Date Title
Oramas et al. Multi-label music genre classification from audio, text, and images using deep features
KR101648204B1 (en) Generating metadata for association with a collection of content items
JP4378646B2 (en) Information processing apparatus, information processing method, and program
US7953735B2 (en) Information processing apparatus, method and program
US8438168B2 (en) Scalable music recommendation by search
US6987221B2 (en) Auto playlist generation with multiple seed songs
US20070106405A1 (en) Method and system to provide reference data for identification of digital content
Cai et al. Scalable music recommendation by search
US20100217755A1 (en) Classifying a set of content items
WO2001020483A2 (en) Finding perceptually similar music with precomputation
CN1910582A (en) Hierarchical playlist generator
JP2010541092A5 (en)
WO2013185107A1 (en) Systems and methods for recognizing ambiguity in metadata
WO2010151421A1 (en) Methods and apparatus for determining a mood profile associated with media data
CN1950908A (en) Methods and apparatus for selecting items from a collection of items
CN1771498A (en) Method and apparatus for grouping content items
KR20090033750A (en) Method and apparatus for recommending playlist of contents
Stupar et al. Picasso-to sing, you must close your eyes and draw
JP2008522311A (en) Automatic content organization based on content item association
Rhodes et al. Investigating music collections at different scales with audiodb
CN105164671A (en) Using hierarchical scoring for disambiguation in an information retrieval system
WO2008117232A2 (en) Apparatus for creating a multimedia file list
Myna et al. Hybrid recommender system for music information retrieval
Yu et al. Local summarization and multi-level LSH for retrieving multi-variant audio tracks
Cui et al. Quest: querying music databases by acoustic and textual features

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08719814

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08719814

Country of ref document: EP

Kind code of ref document: A2