JP2008210382A - Music data processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a music data processor allowing a user to grasp a music content visually and intuitively based on a feature amount extracted from a sound signal of a music. <P>SOLUTION: The music data processor comprises: a sound signal input means 61 for inputting the sound signal of a music; a feature amount extraction means 62 for extracting a predetermined feature amount from the sound signal; a character imparting means 63 for selecting a two-dimensional or three-dimensional character shape or character motion to express the feature of the music based on the feature amount; and a character storage means 66 for storing the feature amount and the character shape and/or character motion in association with each other. Thus, the predetermined feature amount is extracted for the sound signal of the music, the character featuring the corresponding music is imparted using the extracted feature amount, and thereby a user is allowed to grasp the content of the music visually. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a music data processing apparatus that enables a user to grasp music contents visually and intuitively during music search reproduction and music data operation.

  When a user searches for a favorite song from an MD player, a DVD player, or a song database in which a large amount of music is stored on the hard disk, the song title or singer name is searched as a keyword, In many cases, the user selects a song that he / she finds fits his / her preference from a list in which a list of text information such as a singer name is displayed. However, as described above, only from text information such as the song name and singer name, the song itself is, for example, a song with a fast tempo, a song with a slow tempo, an intense song, or a gentle song. It is difficult to know the contents. If more text information than bibliographic information is prepared as bibliographic information, it will help at the time of selection, but separate work is required to prepare and register bibliographic information. In the case of software that uses a database that stores the bibliographic information of music such as CDDB, attempts have been made to improve operability and convenience by automatically inserting bibliographic information. It takes time to understand and understand the contents of the music intuitively.

  On the other hand, if it is possible to display a still image that expresses the content or atmosphere of the song itself in association with the song, it becomes easier to grasp the content of the song based on the image information of the still image in addition to the text information. It becomes easy for the user to classify and search for the desired music. However, at present, it is cumbersome to prepare a still image that matches the content of the music, and even if multiple representative still images that express the content are prepared, it is difficult to properly correspond to the music .

  Currently, there is a technique for displaying a character or the like when playing a music piece. However, they may be displayed randomly regardless of the music content, or may display music information such as the song title and lyrics as keywords. In the former method, it is impossible to grasp the contents of the music, and in the latter method, a work for registering music information is separately required and takes time.

  The present invention proposes a means for assigning a still image suitable for a song based on the contents of the acoustic signal of the song, thereby easily searching for or selecting a song desired by the user from a lot of song data. It has been made for the purpose of providing a means for making it possible. In addition, by suggesting a means to operate a character that matches the music based on the contents of the acoustic signal of the music, it is possible for the user to visually grasp the music contents as well as the hearing during music playback. It was made for the purpose of providing a means to make.

  The music data processing device according to claim 1 is an acoustic signal input means for inputting an acoustic signal of music, a feature quantity extracting means for extracting a predetermined feature quantity from the acoustic signal, and a music feature based on the feature quantity. Character adding means for selecting the shape and / or character motion of a two-dimensional or three-dimensional character expressing the character, and character storage means for storing the feature value in association with the character shape and / or character motion. To do.

  The music data processing apparatus according to claim 2, wherein an acoustic signal input means for inputting an acoustic signal of the music every certain period, a feature quantity extraction means for extracting a predetermined feature quantity from the acoustic signal, and a change in the feature quantity And a character providing means for selecting the shape and / or action of the two-dimensional or three-dimensional character in conjunction with the character, and a character storage means for storing the feature value in association with the character's shape and / or character action. .

  According to a third aspect of the present invention, in the music data processing apparatus according to the first or second aspect, the feature amount is extracted at the time of music registration or music reproduction.

  According to a fourth aspect of the present invention, in the music data processing apparatus according to the third aspect, the feature amount extraction is performed at the time of music playback, or is performed in parallel with music playback or before music playback. It is characterized by extracting in advance.

  According to a fifth aspect of the present invention, in the music data processing apparatus according to the first or second aspect, the feature amount extraction unit is an arbitrary combination of the whole music, a part of the music, and a plurality of parts of the music The feature quantity of the acoustic signal of a music is extracted from the area | region which consists of.

  According to a sixth aspect of the present invention, in the music data processing apparatus according to the first or second aspect, the acoustic signal input means inputs a compressed acoustic signal of the music.

  A seventh aspect of the present invention is the music data processing apparatus according to the first or second aspect, further comprising data management means for performing management by associating the music, the feature amount, the character shape and / or the character motion. It is characterized by.

  According to an eighth aspect of the present invention, in the music data processing apparatus according to the first or second aspect, the character storage means stores at least one of the character shape and the character action in association with the feature amount. It is characterized by doing.

  According to the first aspect of the present invention, a predetermined feature value is extracted from an acoustic signal of a music piece, and a character characterizing the corresponding music piece is assigned using the feature quantity, so that the user can obtain the music piece. It becomes possible to grasp the contents of.

  According to the second aspect of the present invention, a predetermined feature amount is extracted several times from the sound signal of the music piece, and the shape of the character characterizing the corresponding music piece and / or By changing the operation, the user can visually grasp the contents of the music.

  According to the third aspect of the present invention, it is possible to select the timing for extracting the feature amount at the time of music registration and at the time of music playback.

  According to the fourth aspect of the present invention, even when a certain amount of time is required to improve the accuracy of feature quantity extraction, it is possible to use the feature quantity with high accuracy by using the feature quantity extracted in advance as an initial value. Become.

  According to the fifth aspect of the present invention, by enabling selection of the feature extraction area, the user can select a song having characteristics close to the intro part, ending part, or favorite phrase part of the character. It can be easily known from the shape and operation.

  According to the invention described in claim 6, not only linear PCM digital audio used in music CDs but also compressed audio data such as AAC, MP3, WMA and the like are extracted from the sound signal of the music. Characters can be associated according to the feature amount.

  According to the seventh aspect of the present invention, it is possible to store and manage data such as music acoustic signals, feature amounts, character shapes and / or actions.

  According to the eighth aspect of the present invention, it becomes possible to fix one of the shape and the motion of the character.

  When displaying a list of songs, the thumbnails of the video data are displayed in correspondence with the music so that the list of video data is displayed as thumbnails. It becomes possible to grasp the contents of the contents visually and intuitively.

(Embodiment 1)
Embodiment 1 of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the overall configuration of a music search apparatus according to Embodiment 1 of the present invention. In FIG. 1, 11 is an acoustic signal input means, 12 is a feature quantity extraction means, 13 is a thumbnail assignment means, 14 is a data management means, 15 is a browsing means, 16 is a browsing requirement input means, and 17 is a bibliographic information input means. ing.

  The operation of the music search apparatus configured as described above will be described below with reference to FIG. This apparatus is roughly divided into a music data registration unit 111 for registering an acoustic signal of the target music and its accompanying data, a data management means 14 for managing the registered music data, and the managed data. The music browsing unit 112 is configured to browse a music desired by a user.

  First, the music data registration unit 111 will be outlined. The data management means 14 records the acoustic signal and associated information described below in association with each piece of music so that it can be searched for. First, the acoustic signal input unit 11 registers the acoustic signal of the music input as a registration target in the data management unit 14 and outputs it to the subsequent feature quantity extraction unit 12 for generating accompanying information. When the input acoustic signal is an analog signal, the acoustic signal input unit 11 digitizes and outputs it to the subsequent stage. In the case of a compressed acoustic signal, the compressed data is registered in the data management unit 14, and after the compressed data is expanded, the expanded data is output to the feature amount extraction unit 12.

  Next, the feature amount extraction unit 12 extracts some feature amounts representing the physical features of the acoustic signal from the input acoustic signal, registers the feature amount in the data management unit 14 as accompanying information, and assigns a subsequent thumbnail. The feature value is output to the means 13. The thumbnail assigning unit 13 generates a thumbnail still image from the input feature amount, and registers it in the data management unit 14 as accompanying information. In association with the above procedure, the bibliographic information input unit 17 registers the input bibliographic information such as the music title, singer name, and genre name in the data management unit 14 as accompanying information.

  A user inputs a singer name, a genre name, or the like as a keyword from the browsing requirement input means 16, and the data management means 14 searches data managed based on the keyword and extracts a corresponding music candidate. The result is output to the browsing means 15. In response to the result obtained from the data management unit 14, the browsing unit 15 displays a list of thumbnail still images associated with the corresponding music together with bibliographic information such as a music name. From the displayed list, the user selects a desired piece of music through the browse requirement input means 16 based on information obtained from the still image for thumbnail or bibliographic information. The browsing requirement input means 16 causes the audio management signal of the selected music to be output from the data management means 14 to the browsing means 15, and the browsing means 15 reproduces the music.

As feature quantities extracted by the feature quantity extraction means 12, the degree of spectrum change P1 (the degree of spectrum change between frames), the average number of pronunciations P2 (the pronunciation frequency of sounds produced in the music), the pronunciation aperiodicity P3 The degree of non-periodicity of the sound generated in the music) and the beat period P4 (the length of time corresponding to a quarter note of the music), but in addition to the above four types of feature quantities, the beat period ratio P5 Parameters such as the beat intensity P6 and the beat intensity ratio P7 may also be used as feature quantities. Details of the calculation method of these feature amounts are described in Japanese Patent Application No. 2001-082150. In extracting the feature value from the target music, the feature value extraction range includes the entire music, a part of the music, and
You may extract from the area | region which consists of arbitrary combinations with respect to several parts of a music.

  Next, the operation of the thumbnail providing unit will be described. The thumbnail assigning unit inputs N values of feature values P1 to PN extracted in the previous stage, and generates a still image for thumbnails based on these values. As an example of thumbnail still image generation, generation of a two-dimensional or three-dimensional shape still image and generation of a color still image will be described.

  For generating a two-dimensional or three-dimensional shape still image, M feature amounts are selected in advance from N feature amounts, and a graph having M values is generated. The graph may be a still image displayed as a pie chart or a bar graph, or may be generated as a still image expressing M-dimensional values in a two-dimensional or three-dimensional shape.

  For generation of a color still image, M feature values are selected in advance from N feature values and converted to a color space using M values. (Equation 1) is an example of conversion to a color space value using the three primary colors based on the above-described three feature amounts.

The beat period P4 represents the time length of the quarter note of the music, and is a numerical value representing the tempo of the music. Therefore, according to (Equation 2), a music with a fast tempo using the beat period P4 has a red color. In addition, a still image may be displayed in a blue color for music with a slow tempo.

  Through the above operation, the thumbnail assigning unit 13 generates a thumbnail still image for the music, and the data management unit 14 manages this in association with the music. Further, bibliographic information such as a music title, singer name, and genre name is input from the bibliographic information input means 17, and the data management means 14 manages these in association with the music. Data such as music audio signals, bibliographic information, feature amounts, and still images for thumbnails are associated with each other and managed by the data management means 14 in the form of records. FIG. 3 shows an example of a record format managed by the data management means 14. 32 indicates the content of the feature amount, and 31 indicates a record in which the bibliographic information and the feature amount are arranged as the accompanying information 1 before the sound signal of the music. Reference numeral 33 denotes accompanying information 2, which includes thumbnail still image data. The acoustic signals 33 of the accompanying information 2 and 31 are associated with each other based on the music ID uniquely given to each music.

  Next, when the user inputs, for example, a genre name as a search keyword from the browse requirement input unit 16, the browse requirement input unit 16 instructs the data management unit 14 to search for music data based on the keyword. The data management means 14 outputs a list of music corresponding to the specified genre as a search result to the browsing means 15 together with the bibliographic information associated with each music and the still image for thumbnail and the music ID indicated by 31. 15 displays bibliographic information and thumbnail still images associated with the music. The user selects a desired piece of music from the list of bibliographic information and thumbnail still images displayed by the browsing unit 15 using the browsing requirement input unit 16, and the browsing requirement input unit 16 selects the selected piece of music. A search is instructed to the data management means 14 based on the song ID. The data management unit 14 outputs the acoustic signal data of the music corresponding to the music ID to the browsing unit 15, and the browsing unit 15 reproduces the music.

  In the above, the first embodiment of the present invention has been described as an apparatus for realizing the above-described operation. However, the respective units may be configured as a program for realizing the above-described operation on a computer. In addition, the program may be a program that executes the music data registration unit 111 and the data management unit on a server, and the music browsing unit 112 may be a program that is executed on a terminal computer via the Internet. Alternatively, a program for realizing data management means is executed on the server, and the programs for realizing the music data registration unit 111 and the music browsing unit 112 are executed on the same computer or different computers via the Internet. It is good.

(Embodiment 2)
Embodiment 2 of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing the overall configuration of the music search apparatus according to Embodiment 2 of the present invention. In FIG. 2, 11 is an acoustic signal input means, 12 is a feature amount extraction means, 23 is a thumbnail assignment means, 24 is a data management means, 25 is a browsing means, 26 is a browsing requirement input means, and 27 is a still image input means. ing.

  The operation of the music search apparatus configured as described above will be described below with reference to FIG. This apparatus is roughly divided into a music data registration unit 211 for registering an acoustic signal of the target music and its associated data, a data management means 24 for managing the registered music data, and the managed data. The music browsing unit 212 browses the music desired by the user. This apparatus uses each means indicated by each block in the first embodiment, and a part thereof is replaced with a new function. The contents will be described below.

  The acoustic signal input unit 11 and the feature amount input unit 12 perform the same operations as the acoustic signal input unit 11 and the feature amount input unit 12 of the first embodiment, and output the feature amount extracted from the music to the thumbnail providing unit 23. To do. At this time, if a still image for thumbnail is input from the still image input means 27, the thumbnail providing means 23 outputs the thumbnail still image and the feature quantity to the data management means 24, and the data management means 24 Are associated with the sound signal data of the music and managed.

  When a still image for thumbnail is not input from the still image input unit 27, the thumbnail providing unit 23 is managed by the data management unit 24 using the feature amount P extracted from the input music as a key. The feature amount PX in the accompanying information is searched, the Euclidean distance between each feature amount is calculated in (Equation 3), and its reciprocal L is obtained as a value representing the similarity.

The thumbnail assigning means 23 obtains a feature amount PXm having the largest value Lm in the calculated similarity, and extracts thumbnail still image data managed in association with PXm. Thereafter, the thumbnail still image, the feature amount P, and the similarity Lm are output to the data management unit 24 in association with the music, and the data management unit 24 manages these in association with each other. 3 shows an example of the accompanying information 3 which is a record having thumbnail still image data and similarity data.

  The user designates the list display of the music data by the browse requirement input means 26, and the data management means 24, in response to an instruction from the browse requirement input means 26, searches for the music ID, thumbnail still image data, feature amount as a search result. The similarity list is output to the browsing means 25. The browsing means 25 displays a list of thumbnail still images.

  Further, when displaying the thumbnail still image, the browsing means 25 adjusts the display brightness of the thumbnail still image or the color tone based on the similarity data associated with the thumbnail still image. Thus, the degree of similarity between music pieces assigned the same thumbnail still image may be expressed. Also, the similarity is calculated from all the feature values corresponding to each music using the Euclidean distance. However, the similarity may be calculated as a Euclidean distance between values composed of combinations of arbitrary feature values. However, instead of the distance, a value calculated using a simple value difference or a coefficient that emphasizes a specific feature amount may be used as the similarity. Further, the bibliographic information input unit according to the first embodiment of the invention is added to the second embodiment, and bibliographic information can be input. The browsing unit 25 includes a bibliography such as a song title of a song together with a still image for thumbnails. Information may be displayed. FIG. 4 shows an example of a list display presented by the browsing means 25. Reference numeral 41 denotes a browse screen, and reference numeral 42 denotes an example of a thumbnail still image. As in the first embodiment, the entire program may be configured as a program for realizing the operation of the second embodiment on a computer. The program may be executed on a server and via the Internet. It is also possible to adopt a form having a part executed on one or a plurality of computers.

(Embodiment 3)
Embodiment 3 of the present invention will be described below with reference to the drawings. FIG. 5 is a block diagram showing the overall configuration of the music data processing apparatus according to Embodiment 3 of the present invention. In FIG. 5, 51 is an acoustic signal input means, 52 is a feature quantity extraction means, 53 is a character provision means, 54 is a data management means, 55 is a display operation means, and 56 is a character storage means.

  The operation of the music data processing apparatus configured as described above will be described below with reference to FIG. This apparatus is roughly divided into a music data registration unit 511 for registering an acoustic signal of the target music and its associated data, a data management means 54 for managing the registered music data, and the managed data. The display operation means 55 is configured to display and operate characters corresponding to the characteristics of the music. First, the music data registration unit 511 will be outlined. The data management means 54 records the sound signal and accompanying information described below in association with each piece of music so that it can be searched for.

  First, the acoustic signal input unit 51 registers the acoustic signal of the music input as a registration target in the data management unit 54 and outputs it to the subsequent feature amount extraction unit 52 for generating accompanying information. If the input acoustic signal is an analog signal, the acoustic signal input means 51 digitizes it and outputs it to the subsequent stage. In the case of a compressed acoustic signal, the compressed data is registered in the data management unit 54, and after the compressed data is expanded, the expanded data is output to the feature amount extraction unit 52.

Next, the feature quantity extraction unit 52 extracts some feature quantities representing the physical features of the acoustic signal from the input acoustic signal, registers them as accompanying information in the data management unit 54, and assigns the character in the subsequent stage. The feature value is output to the means 56. Note that, as in the first embodiment, the feature amount includes a spectrum change degree P1, an average pronunciation number P2, a pronunciation non-periodicity P3, and a beat period P4. Parameters such as the ratio P5, the beat intensity P6, and the beat intensity ratio P7 may also be used as feature amounts. Moreover, when extracting the feature value from the target music, the feature value extraction range may be extracted from an area composed of any combination of the entire music, a part of the music, and a plurality of parts of the music.

  Next, the character provision unit 53 will be described. The character assigning means 53 inputs N values of the feature quantity P1 to PN extracted in the previous stage, and the shape and motion of the two-dimensional or three-dimensional character associated with the feature quantity closest to the feature quantity. Is selected from the subsequent character storage means and registered in the data management means 54 as accompanying information. Hereinafter, examples of character selection will be described.

  The character storage unit 56 associates the feature amount of the music with the character shape and the character action in advance. It may correspond to each parameter of a specific or arbitrary combination of feature amounts, or may correspond to statistics of N values from P1 to PN. When the feature amount is input to the pre-stage character assigning means 53, the character assigning means 53 determines the character shape and the character associated with the feature amount that is closest to the feature amount of the music from the character storage means 56 by the above method. Select the action. Further, regarding the tempo of the character's motion, if the beat cycle P4 is used as a time parameter for designating the motion cycle, the beat cycle P4 represents the time length of the four-part note of the music, so that the character corresponds to the corresponding music It works with movement that matches the tempo. In addition to this, the content of the music may be expressed by more movements by using other feature amounts as the motion parameters of the character. The pre-stage character assigning means 53 designates the character shape, motion, and motion cycle by the above means and registers them in the data management means.

  Further, when associating the feature amount of the music with the shape or motion of the character, as described in Japanese Patent Application No. 2001-082150, the user with respect to the music is performed by a sensory evaluation experiment using an SD (semantic differential) method or the like. If the shape and movement of the character are determined based on the subjective impression of the song, it will be more suitable for the content of the music.

  With the above operation, the character giving means 53 selects a character for the music piece, and the data management means 54 manages it in association with the music piece. The data management means 54 manages the sound signal, feature quantity, character shape, character motion, and other data of the music piece in association with each other. FIG. 7 shows an example of a record format managed by the data management means 54. 72 indicates the contents of the feature amount, and 73 indicates character information. Reference numeral 71 denotes a record in which feature amounts and character information are arranged as accompanying information before an acoustic signal of music. 74 is an example of character storage, and x is a feature value. In addition to 74, some music records may be registered in advance in the character storage means. 75 is a character shape information list, and 76 is a character motion information list.

At the time of music reproduction, the data management means 54 extracts the corresponding music information from the managed data and outputs the result to the display operation means 55. The display operation unit 55 operates the character associated with the corresponding music in response to the result obtained from the data management unit 54. Also, in the above embodiment, once registered music data is stored by the data management means 54, and the data is read at the time of music playback, but the feature quantity is extracted for each music playback without using the data management means, and the character and The display operation may be performed in association with each other. In addition, when extracting the feature amount for each music reproduction, the feature amount accuracy may be increased by using the feature amount extracted in advance as an initial value. Further, as in the first embodiment, the whole may be configured as a program for realizing the operation of the third embodiment on a computer. The program may be executed on a server and via the Internet. It is good also as a form with the part performed on one or several computers.

(Embodiment 4)
Embodiment 4 of the present invention will be described below with reference to the drawings. FIG. 6 is a block diagram showing the overall configuration of the music data processing apparatus according to Embodiment 4 of the present invention. In FIG. 6, 61 is an acoustic signal input means, 62 is a feature amount extraction means, 63 is a character provision means, 64 is a data management means, 65 is a display operation means, and 66 is a character storage means.

  The operation of the music data processing apparatus configured as described above will be described below with reference to FIG. The apparatus is roughly divided into a music data registration unit 611 for registering an acoustic signal of the target music and its associated data, a data management means 64 for managing the registered music data, and the managed data. It comprises display operation means 65 for displaying and operating characters corresponding to the characteristics of the music. This apparatus uses each means indicated by each block in the third embodiment, and a part thereof is replaced with a new function. The contents will be described below.

  The acoustic signal input means 61 outputs the acoustic signal of the music to the feature quantity extraction means 62 and the data management means 64 at a time T (T1, T2,... Tn,...) Every fixed time delta T. The feature quantity extracting means 62 extracts some feature quantities representing the physical characteristics of the input acoustic signal at the time Tn, registers them in the data management means 64 as accompanying information, and features them in the subsequent character adding means 63. Output quantity. The character imparting means 63 selects the shape and / or action of the two-dimensional or three-dimensional character associated with the feature quantity closest to the feature quantity extracted every fixed time T from the subsequent character storage means. Then, it is registered in the data management means 64 as accompanying information together with the time Tn. During music reproduction, the data management means 64 extracts the corresponding music information from the managed data and outputs the result to the display operation means 65. The display operation unit 65 operates the character associated with the corresponding music in response to the result obtained from the data management unit 64.

  With the above operation, the character giving means 63 selects a character for the music, and the data management means 64 manages this in association with the music. The data management means 64 manages the acoustic signal of the music, the feature amount, the character shape, and the data such as the character motion at a predetermined time T in the form of records. In the above embodiment, once registered music data is stored by the data management means 64, and the data is read at the time of music reproduction. However, the feature quantity is extracted for each music reproduction without using the data management means, and is associated with the character. A display operation may be performed. In the above-described embodiment, the data is registered in the data management means 64 at every fixed time T regardless of whether the character shape and the movement are changed. However, at time Tn, the previous time Tn−1 is registered. When there is a change in the shape and action of the character, only the changed data may be registered in the data management means 64. Further, by shortening the time period delta T, it becomes more suitable for the contents of the music. Further, as in the first embodiment, the whole may be configured as a program for realizing the operation of the fourth embodiment on a computer. The program may be executed on a server and via the Internet. It is good also as a form with the part performed on one or several computers.

  The music data processing apparatus according to the present invention can automatically display a two-dimensional or three-dimensional character according to the characteristics of the music and display it in conjunction with the reproduction of the music. As a result, the contents of the favorite music can be viewed visually as well as audibly, so that it can be applied to an entertainment terminal in a moving space such as a home, a theater or a car.

The block diagram showing schematic structure of the music search device by Embodiment 1 of this invention. The block diagram showing schematic structure of the music search apparatus by Embodiment 2 of this invention. The figure which shows the example of the record which the data management means by Embodiment 1 and 2 of this invention manages The figure which shows the example of a screen which the browsing means by Embodiment 2 of this invention shows. The block diagram showing the schematic structure of the music data processing apparatus by Embodiment 3 of this invention. Block diagram showing a schematic configuration of a music data processing apparatus according to Embodiment 4 of the present invention. The figure which shows the example of the record which the data management means by Embodiment 3 and 4 of this invention manages, and the example of character information storage

Explanation of symbols

11, 51, 61 Acoustic signal input means 12, 52, 62 Feature amount extraction means 13, 23 Thumbnail assignment means 14, 24, 54, 64 Data management means 15, 25 Browse means 16, 26 Browse requirement input means 17 Bibliographic input means 17 27 Still image input means 111, 211, 511, 611 Music data registration unit 112, 212 Music data browsing unit 31 Accompanying information 1
32 feature amount 33 additional information 2
34 Accompanying information 3
41 Browse screen 42 Still image for thumbnail 53, 63 Character giving means 55, 65 Display operation means 56, 66 Character storage means 71 Music record 72 Feature value 73 Character information 74 Character information storage example 75 Character shape information 76 Character action information

Claims (8)

An acoustic signal input means for inputting an acoustic signal of a song, a feature quantity extracting means for extracting a predetermined feature quantity from the acoustic signal, and a two-dimensional or three-dimensional representation of the feature of the song based on the feature quantity A music data processing apparatus comprising: character giving means for selecting a shape and / or action of a character; and character storage means for storing the feature quantity in association with the shape and / or action of the character. An acoustic signal input means for inputting an acoustic signal of a music every certain period, a feature quantity extraction means for extracting a predetermined feature quantity from the acoustic signal, and two-dimensional or three-dimensional in conjunction with the change of the feature quantity A music data processing apparatus, comprising: character giving means for selecting a character shape and / or action; and character storage means for storing the feature quantity in association with the character shape and / or character action. 3. The music data processing apparatus according to claim 1, wherein the feature amount extraction unit extracts the music at the time of the music registration or the music reproduction. 4. 4. The music data processing apparatus according to claim 3, wherein the feature amount extraction means is performed in parallel with the reproduction of the music or is extracted in advance before the reproduction of the music. The feature amount extraction unit extracts a feature amount of an acoustic signal of a song from an area composed of an arbitrary combination with respect to the entire song, a part of the song, and a plurality of portions of the song. Or the music data processing apparatus of Claim 2. The music data processing apparatus according to claim 1, wherein the acoustic signal input unit inputs a compressed acoustic signal of music. The music data processing apparatus according to claim 1 or 2, further comprising data management means for performing management by associating a music, a feature value, a character shape, and / or a character action. The music data processing apparatus according to claim 1 or 2, wherein the character storage means stores at least one of a character shape and a character action in association with a feature amount.

Images