JP4312663B2 - Music selection apparatus, music selection method, program, and recording medium - Google Patents

Description

  The present invention relates to a music selection apparatus, a music distribution system and method, a program, and a recording medium that select music based on a user's reproduction history or preference.

Some conventional audio players record the playback time and the number of playbacks of a song that the user has listened to in the past, and select and play back the songs in the order of the playback time and the number of playbacks (see, for example, Patent Document 1). As a result, music that suits the user's preference, that is, music that has a long playback time and number of playbacks, is automatically prioritized and selected, so that the user does not have to perform an operation of skipping unfavorable music. You can listen to music that suits your taste.
JP 2001-272737 A

  Here, in the audio player, the music that the user often listens to is selected. Therefore, with this music player music selection method, there is a high possibility that the same music selection result will be obtained each time. That is, there are few variations in the music to be selected, and the music selection result may be monotonous. When the music selection result becomes monotonous, the user listens to the same music every time, so there is a high possibility that the user will get bored with the selected music.

  Therefore, an object of the present invention is to provide a music selection device that can select music that suits the user's preference and obtain a variety of music selection results.

The present invention has the following configuration in order to solve the above problems. That is, the first invention is a music selection apparatus that selects music to be reproduced and presents it to the user. The music selection apparatus includes a music data storage unit that stores music data, and an impression amount storage that stores an impression amount obtained by quantifying a listener's impression of the music for each music stored in the music data storage unit. A song selection priority storage unit that stores, for each song, a song selection priority that represents at least one of the song playback history and the song preference level, and a song selection process for selecting a song to be played from each song And a music selection processing unit that uses the distance between music pieces and the music selection priority in the impression space with each component of the impression amount as a coordinate axis. The music selection processing unit includes an area determination unit that determines one of the areas obtained by dividing the impression space based on the music selection priority of each music, and an impression amount in the area determined by the reference position determination unit. And a music selection execution unit that performs a music selection process using the music selection priority.

Further, the second inventions, a music song selection device to be presented to the user by selecting the music to be reproduced. The music selection apparatus includes a music data storage unit that stores music data, and an impression amount storage that stores an impression amount obtained by quantifying a listener's impression of the music for each music stored in the music data storage unit. A song selection priority storage unit that stores, for each song, a song selection priority that represents at least one of the song playback history and the song preference level, and a song selection process for selecting a song to be played from each song And a music selection processing unit that uses the distance between music pieces and the music selection priority in the impression space with each component of the impression amount as a coordinate axis. The music selection processing unit calculates a degree of relevance obtained by changing the music selection priority of the other music according to the positional relationship with the other music in the impression space, and a music selection of each music with a probability corresponding to the priority and relevance, that contains a musical piece selecting unit for selecting the music to be reproduced.

Further, the third inventions, a music song selection device to be presented to the user by selecting the music to be reproduced. The music selection apparatus includes a music data storage unit that stores music data, and an impression amount storage that stores an impression amount obtained by quantifying a listener's impression of the music for each music stored in the music data storage unit. A song selection priority storage unit that stores, for each song, a song selection priority that represents at least one of the song playback history and the song preference level, and a song selection process for selecting a song to be played from each song And a music selection processing unit that uses the distance between music pieces and the music selection priority in the impression space with each component of the impression amount as a coordinate axis. The music selection processing unit calculates the music selection priority at an arbitrary position in the impression space based on the music selection priority of each music and the music selection priority calculated by the music selection priority calculation unit. a reference position determination unit for determining a reference position in the impression space probability, based on the positional relationship between the position and the reference position of each song in the impression space, that contains a musical piece determination unit that determines music to be reproduced .

In the fourth invention, the music selection device may further include a reproduction unit that reproduces the music selected by the music selection processing unit using the data stored in the music data storage unit.

In the fifth invention, the music selection priority may include at least one of the number of times the music has been played and the date and time when the music data was last played as information indicating the playback history. At this time, the music selection device receives the input for specifying the music to be reproduced from the user, and when the input is received by the music specification reception unit, the reproduction history stored in the music selection priority storage unit is displayed. And a playback history update unit that updates the input based on the input.

In the sixth invention, the music selection priority may include at least information indicating a preference level. At this time, when the music selection apparatus receives an input from the user for setting an preference for the music reproduced by the reproduction unit from the user, and an input is received by the preference setting reception unit, the music selection priority storage unit And a preference level update unit that updates the preference level stored in the table based on the input.

In the seventh invention, the music selection apparatus uses a plurality of music pieces whose distances in the impression space are larger than a predetermined distance from the music pieces stored in the music data storage unit for testing. You may further provide the test music selection part selected as a music. At this time, the playback unit plays back the test music in response to the test music selected by the test music selection unit.

In the eighth invention, the music selection device according to the ratio of the portion received by the playback unit with respect to the whole of the music, and the stop reception unit that receives an input to stop the playback of the music by the playback unit from the user, You may further provide the music selection priority update part which updates the music selection priority stored in the music selection priority storage part.

  Moreover, this invention may be provided as a music selection method used in the said music selection apparatus. Moreover, the present invention may be provided as a program that realizes each function of the music selection device or a recording medium that records the program by being executed by a computer.

  According to the present invention, music selection is performed using the music selection priority including the reproduction history or the preference level, so that the music can be selected reflecting the user's preference. Further, according to the present invention, the distance between music pieces in the impression space is taken into consideration when selecting music. Here, since the distance indicates the similarity of the music, the music selection is performed based on the two factors of the music selection priority and the music similarity by selecting the music in consideration of the distance. As a result, the music selection result is not uniquely determined based on the music selection priority, so that the music selection result rich in variations can be obtained.

(Embodiment 1)
First, the music selection apparatus according to Embodiment 1 of the present invention will be described. FIG. 1 is a block diagram showing a music selection apparatus according to the first embodiment. In FIG. 1, the music selection apparatus includes a music database 1, an update unit 2, a reproduction unit 3, a control unit 4, a music selection processing unit 5, an input reception unit 6, and a display unit 7. This music selection device is used as, for example, a car audio system or a jukebox.

  The music database 1 stores music data and related information related to the music (see FIG. 2 described later). The related information is, for example, the song title, artist name, performance time, etc. of the song. The music database 1 stores the music selection priority and the impression amount for each music for which data is stored. The contents of the music database 1 may be acquired by any method. The content of the music database 1 may be acquired, for example, by a recording medium that can be read by the music selection device, or may be acquired by communication when the music selection device can communicate with another information processing device. Good.

  The music selection priority is an index used for a process (music selection process) for selecting a music to be reproduced from music stored in the music database 1. Specifically, the music selection priority represents at least one of a music reproduction history and a music preference. The music playback history is, for example, the number of times the music has been played, the date and time (playback time) of the music. The music preference level is a degree representing the user's preference for music. The preference level of the music may be a value set directly by the user or a value calculated from the reproduction history. The preference degree of the music may be expressed by a numerical value, or may be expressed by binary of likes / dislikes.

  The impression amount is a quantified impression of the music. The impression amount is expressed by a numerical value set for each item such as “intensity” and “exhilaration”, for example. The impression amount preferably has a plurality of items, that is, is expressed as a multi-dimensional vector, but may have only one item, that is, a one-dimensional scalar amount. The impression amount is calculated from a musical feature amount (for example, tempo, average number of sounds, etc.) obtained as a result of analyzing the music signal of the music. For example, the impression amount can be calculated by performing a predetermined weighted addition to the musical feature amount.

  The music selection processing unit 5 shown in FIG. 1 performs music selection processing. In the music selection process by the music selection processing unit 5, an impression amount is used in addition to the music selection priority. For example, when the music selection process is performed using only the music selection priority such as the reproduction history, only the music that is often reproduced is selected. As a result, the same music selection result is obtained every time, and the user may get bored with the music selection result. Therefore, in the present invention, by selecting music using the music selection priority and impression amount, it is possible to obtain a variety of music selection results that are rich in variations and reflect user preferences. In the first embodiment, the music selection processing unit 5 performs a music selection process (automatic music selection process) that is automatically performed using the music selection priority and the impression amount, and a music selection process that is performed in accordance with a song name specified by the user. (Normal music selection process).

  The update unit 2 updates the contents of the music database 1 in response to an instruction from the control unit 4. Specifically, the update unit 2 updates the music selection priority stored in the music database 1. The reproduction unit 3 reads out and reproduces music data stored in the music database 1 in response to an instruction from the control unit 4. The control unit 4 instructs the music selection processing unit 5 to execute the music selection process or instructs the update unit 2 to update the music database 1 in accordance with a user instruction received via the input reception unit 6.

  The input receiving unit 6 includes an input device such as a button or a touch panel. The input receiving unit 6 receives various instruction inputs from the user. For example, the input receiving unit 6 is for the user to instruct the music selection processing unit 5 to perform music selection processing, to input the preference of music, or to instruct to skip the music being played back. Used.

  The display unit 7 is configured by a display device such as a CRT or a liquid crystal display device. The display unit 7 displays a list of music (reproduced music list) obtained as a result of the music selection process, an input screen used for normal music selection, and the like.

  FIG. 2 is a diagram showing an example of data stored in the music database shown in FIG. The music database 1 includes a music number, music data, related information, impression amount, and music selection priority for each music. The music number is a number for which a unique value is set for each music in order to identify the music. The music data is music data, that is, data representing a music signal of the music. The music data is, for example, waveform data obtained by linear PCM of music or compressed data such as AAC. The related information is information related to the music as described above. The related information includes, for example, a song title (song title), artist name, album title, songwriter, composer, arranger, genre, performance time, and the like. In the first embodiment, it is assumed that the related information includes the song title and the performance time of the song.

  In the first embodiment, the impression amount includes two components (items) and is expressed as a two-dimensional vector. Therefore, each music in the music database 1 (each music for which data is stored in the music database 1) can be expressed as a point in a space having each component of the impression amount as a coordinate axis. Hereinafter, the space is referred to as an impression space. In Embodiment 1, since the impression amount is a two-dimensional vector, the impression space is a two-dimensional plane. FIG. 3 is a diagram illustrating an example of an impression space in the first embodiment. Each point shown in FIG. 3 represents each song in the song database 1. Since the impression amount is a numerical value representing the impression of listening to the music, if a point indicating the impression amount of a plurality of music is placed in the impression space, the point indicating each music is placed according to the similarity of the impression. It becomes. That is, the two music pieces indicated by the two points located close to each other in the impression space are music pieces that are similar to each other (music pieces that receive a similar impression). As shown in FIG. 2, in the music database 1, one of the two components of the impression amount is stored as an X coordinate value, and the other component is stored as a Y coordinate value.

  In FIG. 2, the music database 1 includes the number of reproductions, the last reproduction date, and the preference score as the music selection priority. The number of times played is the number of times the music has been played so far. The last playback date is the date when the music was last played. The preference score is an index indicating the degree of user preference. The preference score changes at the timing when the input by the user and the number of reproductions and the last reproduction date are updated.

  In addition, about the content of the music database 1, it is possible to add and delete suitably. In the first embodiment, when a music piece is added, a music number is automatically assigned. Further, when the music data of the music is registered, the impression amount of the music is automatically calculated, and the calculated impression amount is stored. Further, when the music data of the music is registered, the number of reproductions is set to “0”, the last reproduction date is set to blank (“(blank)” in FIG. 2), and the preference score is “5” as an initial value. Set to

  Hereinafter, the operation of the music selection apparatus will be described. FIG. 4 is a flowchart showing a process flow of the music selection apparatus according to the first embodiment. The process shown in FIG. 4 is started when, for example, the music selection apparatus is turned on. First, in step S1, the control unit 4 determines the music selection mode. The music selection mode includes two modes, an automatic music selection mode for performing the above automatic music selection processing and a normal music selection mode for performing normal music selection processing. In step S1, it is determined whether or not the automatic music selection process is to be executed, that is, whether the music selection apparatus is to execute the automatic music selection process or the normal music selection process. Here, selection of the music selection mode is performed by the user. The input receiving unit 6 is provided with a button for selecting a music selection mode. FIG. 5 is a diagram illustrating an example of a button provided in the input receiving unit 6 for selecting a music selection mode. As shown in FIG. 5, the input receiving unit 6 includes an automatic music selection button 11 and a normal music selection button 12. The user selects a music selection mode by pressing either the automatic music selection button 11 or the normal music selection button 12. In the first embodiment, one of the automatic music selection mode and the normal music selection mode is always selected. The control unit 4 determines whether or not to execute the automatic music selection process according to the user's selection received by the input reception unit 6. If it is determined in step S1 that automatic music selection processing is to be executed, processing in step S2 is performed. On the other hand, if it is determined in step S1 that the automatic music selection process is not executed (that is, the normal music selection process is executed), the processes in steps S3 and S4 are performed.

  Next, the automatic music selection process in step S2 will be described. FIG. 6 is a flowchart showing detailed processing of step S2 of FIG. Automatic music selection processing is performed by the music selection processing unit 5 in response to an instruction from the control unit 4. First, in step S11, the music selection processing unit 5 divides the impression space into a plurality of areas. FIG. 7 is a diagram illustrating an example of a state in which the impression space is divided into a plurality of areas. In the first embodiment, the impression space is divided into a plurality of areas by a straight line perpendicular to the X axis or the Y axis. Information indicating each area is stored in a storage unit (not shown in FIG. 1) built in the music selection apparatus. As shown in FIG. 7, each piece of music included in a single area is a piece of music having a short distance between the pieces of music in the impression space. Therefore, it can be said that music included in the same area is similar music (music with a similar impression). In FIG. 7, the area is set by dividing the impression space by a straight line at equal intervals, but any method may be used for setting the area. For example, the impression space may be divided by a curve, or the impression space may be divided so that the number of songs included in the area is the same. Further, the music selection processing unit 5 may set the same area each time in step S11, or may set different areas. Note that area numbers are assigned in order from 1 to the plurality of divided areas. Here, j = BA (i) where j is an area j (area number is “j”) (j is an arbitrary integer) to which the music number is i (i is an arbitrary integer). It shall be expressed as

  The music selection processing unit 5 performs a process of determining a music to be reproduced after step S11. In the first embodiment, first, a music selection target area is selected (steps S12 and S13), and then a music that is a music selection result is selected from the music included in the music selection target area (steps S14 and S15). ). The music selection target area is an area in the impression space, in which music in the area is selected as a music selection result. Details of steps S12 to S15 will be described below.

  In step S12, the music selection processing unit 5 calculates an area selection value EA (j) for each area. The area selection value is a numerical value used for selecting a music selection target area from each area obtained by dividing the impression space. The area selection value is an index representing the probability that the area is selected as the music selection target area. Hereinafter, a specific method for calculating the area selection value will be described.

In order to calculate the area selection value, the music selection processing unit 5 first calculates the listening level LA (i) and the preference level TA (i) for each music in the music database 1. The listening level LA (i) of the music i (music having the music number “i”) is calculated based on the music selection priority of the music i. Specifically, if the number of times the music i is played is PC (i), the last playback date of the music i is LT (i), and today's date is NT, the listening degree LA (i) of the music i is (1).
LA (i) = wLA1 * PPC (i) + wLA2 * PLT (i) (1)
However,
PPC (i) = PC (i) (when PC (i) <10)
= 10 (Other cases)
PLT (i) = 365− (NT−LT (i)) (when NT−LT (i) <365)
= 0 (other cases)
In equation (1), wLA1 is a coefficient representing the degree of reflection of the number of reproductions to the listening degree LA (j), and wLA2 is a coefficient representing the degree of reflection of the last reproduction date to the degree of listening LA (j). It is. From equation (1), it can be seen that the value of the listening level LA (i) increases as the number of times of reproduction of the music increases, and the value of the listening level LA (i) increases as the last playback date becomes more recent.

If the preference score of music i is TP (i), the music preference TA (i) is calculated by the following equation (2).
TA (i) = PTP (i) (2)
However,
PTP (i) = TP (i) (when TP (i) <10)
= 10 (Other cases)
From equation (2), it can be seen that the degree of preference TA increases with an increase in preference score TP (i) with 10 as the upper limit.

  In equations (1) and (2), the maximum PPC value is 10, the maximum PLT value is 365, and the maximum PTP value is 10. However, the maximum value may be other values.

The music selection processing unit 5 calculates the listening degree LA (i) and the preference degree TA (i) for all the music pieces in the music database 1. Thereafter, the music selection processing unit 5 calculates an area selection value EA for each area based on the calculated listening level LA (i) and preference level TA (i). Specifically, the area selection value EA (j) for the area j is calculated by performing the following (3) for all musical pieces i where j = BA (i).
EA (j) = Σ (wEA1 * LA (i) + wEA2 * TA (i)) (3)
In equation (3), wEA1 is a coefficient representing the degree of reflecting the listening degree LA in area selection, and wEA2 is a coefficient representing the degree of reflecting the preference TA in area selection. The music selection processing unit 5 calculates the area selection value of each area by calculating Expression (3) for each area.

  After step S12, in step S13, the music selection processing unit 5 selects a music selection target area using each area selection value EA. Specifically, the music selection processing unit 5 selects an area with a probability proportional to the area selection value EA, that is, selects an area such that an area with a larger area selection value EA is selected with a higher probability. Thereby, an area with a high music selection priority (an area with many music with a high music selection priority) is preferentially selected as a music selection target area. That is, it is easy to select a song that has an impression similar to a song that the user often listens to.

  The music selection target area is selected by the above steps S12 and S13. In steps S14 and S15, the music to be selected is determined from the music pieces belonging to the selected music selection target area. For example, assuming that an area 15 indicated by diagonal lines in FIG. 7 is a music selection target area, one piece is selected from the points in the music selection target area 15 to determine the music to be selected. Specifically, first, in step S14, a music selection value is calculated for each piece of music belonging to the music selection target area selected in step S13. The music selection value is a numerical value used to determine the music to be selected. The music selection value is an index representing the probability that the music is selected as a music selection target. Hereinafter, a specific method for calculating the music selection value will be described.

In order to calculate the music selection value, the music selection processing unit 5 first calculates the listening degree LS (i) and the preference degree TS (i) for each music piece belonging to the selection target area. Specifically, the listening degree LS (i) is calculated by the following equation (4).
LS (i) = wLS1 * PPC (i) + wLS2 * PLT (i) (4)
In Expression (4), wLS1 is a coefficient representing the degree of reflection of the number of reproductions to the listening degree LS, and wLS2 is a coefficient representing the degree of reflection of the last reproduction date to the listening degree LS. PPC (i) and PLT (i) are the same as in equation (1). On the other hand, the preference degree TS (i) is calculated by the following equation (5).
TS (i) = PTP (i) (5)
In formula (5), PTP (i) is the same as formula (2).

The music selection processing unit 5 calculates the listening level LS (i) and the preference level TS (i) for all music pieces belonging to the music selection target area. Next, the music selection processing unit 5 calculates the music selection value ES of each music based on the calculated listening level LS and preference level TS. Specifically, the music selection value ES (i) of the music i is calculated by performing the following equation (6).
ES (i) = wES1 * LS (i) + wES2 * TS (i) (6)
In Equation (6), wES1 is a coefficient representing the degree of reflection of the listening degree LS in music selection, and wES2 is a coefficient representing the degree of reflection of the preference TS in music selection. The music selection processing unit 5 calculates the music selection value of each music piece by calculating Expression (6) for each music piece belonging to the music selection target area.

  After step S14, in step S15, the music selection processing unit 5 uses the music selection values ES to select the music that is the music selection result. Specifically, one piece of music is selected from the music pieces belonging to the music selection target area with a probability proportional to the music selection value ES. As a result, an area having a higher listening level or a higher preference level can be easily selected as a music selection target area. That is, the music that the user often listens to is preferentially selected.

  Note that the music selection processing unit 5 may select music randomly from the music selection target area instead of the processes of steps S14 and S15. Even when music is randomly selected from the music selection target area, music similar to the music that the user often listens to is selected, as in the case of performing the processes of steps S14 and S15.

  Further, in the processing of steps S12 to S15, the processing content may be changed depending on the number of reproduction histories stored in the music database 1. That is, when the number of times of reproduction is small, the music selection target area may be selected at random instead of the processing of steps S12 to S15, or the music in the music selection target area may be selected at random. . This is because when the number of times of reproduction is small, the reproduction history may not sufficiently reflect the user's preference. In this case, the reproduction history may be enriched by preferentially selecting music with a small number of reproductions.

  When the process of step S15 ends, the music selection processing unit 5 determines whether or not a predetermined number of music pieces have been selected in step S16. Here, the predetermined number may be a predetermined number or may be set by the user when the user selects automatic music selection. If it is determined in step S16 that a predetermined number of music pieces have not been selected, the music selection processing unit 5 repeats the processes in steps S12 to S15. Thereafter, the processes in steps S12 to S15 are repeated until a predetermined number of music pieces are selected.

  On the other hand, if it is determined in step S16 that the predetermined number of songs have not been selected, the music selection processing unit 5 performs the process of step S17. That is, in step S <b> 17, the music selection processing unit 5 causes the display unit 7 to display a music list (reproduction music list) obtained as a result of the music selection processing. FIG. 8 is a diagram showing an example of a reproduction song list displayed on the display unit 7. In the reproduction music list shown in FIG. 8, the music selection number indicating the order in the selected music list, the music name of the music, and the performance time of the music are displayed for each music. Note that the reproduction song list may include a song number, artist name, and the like in addition to those shown in FIG. After step S17, the music selection processor 5 ends the automatic music selection process.

  In the automatic music selection process described above, since there is a possibility that all songs in the music selection target area may be selected, it is possible to obtain a variety of music selection results. Furthermore, the music pieces in the music selection target area are music pieces having similar impressions. Therefore, no matter what music is selected in the music selection target area, a music completely different from the user's favorite music type is not selected. That is, it is possible to select music according to the user's preference.

  Note that the user's preference and the music selection priority may not correspond completely. For example, the user may want to listen to a type of music that the user does not normally listen to. In the first embodiment, it is possible to preferentially select a type of music that the user does not normally listen to by changing the settings in step S13 and step S15. Specifically, in step S13, the music selection processing unit 5 may select an area with a probability inversely proportional to the area selection value EA. As a result, an area having a low music priority is easily selected as a music selection target area. That is, it is easy to select a type of music that the user does not normally listen to (a music that has a different impression from the music that the user normally listens to). In addition, the music selection processing unit 5 may select music with a probability inversely proportional to the music selection value ES in step S15. This makes it easier to select music with a lower listening level or a lower preference level. That is, the music is similar to the music that the user often listens to, but the music that the user does not play much is preferentially selected.

  Furthermore, by changing the combination of the settings in steps S13 and S15, the music selection apparatus can select music by various methods. For example, if an area is selected with a probability proportional to the area selection value EA at step S13, and a song is selected with a probability inversely proportional to the song selection value ES at step S15, then "always listening type" Can be selected with priority. Further, for example, if an area is selected with a probability that is inversely proportional to the area selection value EA in step S13 and a song is selected with a probability that is proportional to the song selection value ES in step S15, then “always listening” It is possible to select songs with priority on “relatively listened songs with no type of music”.

In step S2, the process of selecting a song is a two-stage process: a process of selecting a music selection target area (steps S12 and S13) and a process of selecting a music in the music selection target area (steps S14 and S15). I went to divide. Here, in another embodiment, the total music selection value TES (i) of the music i may be directly calculated using the following equation (7).
TES (i) = EA (BA (i)) * ES (i) (7)
In Expression (7), EA is the same as the area selection value. ES is the same as the music selection value. The music selection processing unit 5 calculates a total music selection value TES for each music in the music database 1. Then, music is selected with a probability corresponding to the total music selection value TES.

  Here, the music area selection value EA is a numerical value considering the music selection priority of other music in the same area in the impression space. That is, the music area selection value EA is a numerical value obtained by changing the music selection priority of the other music according to the positional relationship with the other music in the impression space. Therefore, the music area selection value EA is a value set in association with other music around the music in the impression space. In this sense, the area selection value EA may be called a relevance level. That is, in Expression (7), the total music selection value TES (i) of the music i is based on the relevance level EA (i) of the music i and the music selection value ES (i) obtained from the music selection priority of the music i. It can be said that it is calculated. Thus, the music selection process may select music with a probability corresponding to the degree of relevance of music and the priority of music selection.

The expression (7) selects music with a probability that is directly proportional to both the degree of association EA and the music selection value ES. However, even if the music is selected with a probability that is inversely proportional to one or both, Good. That is, Expression (7) can be expressed as the following Expression (7) ′.
TES (i) = F1 (EA (BA (i))) * F2 (ES (i)) (7) ′
In Expression (7) ′, F1 (x) is a function representing a method of reflecting the area selection value EA with respect to the probability of selecting a song. That is, if F1 (x) = a * x, it means that the area selection value EA is directly proportional to the probability. That is, a song with a larger area selection value EA is selected with a higher probability. Conversely, if F1 (x) = b / x, it means that the area selection value EA is inversely proportional to the probability. F2 (x) is a function representing a method of reflecting the music selection value ES with respect to the probability of music selection.

  In another embodiment, music selection using the impression amount and the music selection priority can be performed without dividing the impression space into areas. Specifically, the music selection processing unit 5 first calculates a set of an impression amount and a music selection value (ES) for each music in the music database 1. Next, music selection values are estimated for all positions in the impression space based on each set by using spline interpolation or the like. Further, one position (reference position) is selected from all positions in the impression space with a probability proportional to (or inversely proportional to) the estimated music selection value. That is, one point in the impression space is selected so that the larger the estimated music selection value is, the higher the probability is selected. Finally, the music that is closest to the coordinates of the selected reference position is determined as the music selection result. The music selection using the impression amount and the music selection priority may be performed by the above method.

  Returning to the description of FIG. 4, if it is determined in step S <b> 1 that automatic music selection processing is not executed (that is, normal music selection processing is executed), normal music selection processing is performed. The normal music selection process is a process of selecting a music piece that is performed according to various types of information (song name, impression amount, etc.) input by the user. Specifically, first, in step S <b> 3, the music selection processing unit 5 receives an input of music selection conditions from the user via the input reception unit 6. Here, in step S <b> 3, an input screen for receiving music selection conditions is displayed on the display unit 7.

  FIG. 9 is a diagram illustrating an example of an input screen for receiving music selection conditions. As shown in FIG. 9, in this embodiment, a text input receiving unit 21 and an impression amount input receiving unit 22 are displayed as the input screen. The text input reception unit 21 displays related information such as a song title and artist name. The user designates music selection conditions by inputting related information. Then, when the user designates the music selection button 23, the music selection conditions are determined. In the following, the music selection performed in accordance with the input of the text input reception unit 21 is referred to as “text music selection”.

  The impression amount input receiving unit 22 displays the impression amount of each song in the song database 1. That is, the impression space is displayed. In Embodiment 1, since the impression amount is two-dimensional data, each piece of music is displayed as one point on the plane. The user designates the music selection condition by designating an arbitrary point on the impression space displayed in the impression amount input reception unit 22. Then, when the user designates the music selection button 24, the music selection conditions are determined. In the following, the music selection performed in accordance with the input of the impression amount input receiving unit 22 is referred to as “impression amount music selection”.

  In step S3, an input of related information or an input for designating an impression amount is received from the user. In the subsequent step S4, the music selection processing unit 5 selects one or a plurality of songs from each song in the song database 1 in accordance with the song selection condition input in step S3. In other words, in the case of the text music selection, music having related information that matches the specified music selection conditions is selected. In the case of the impression amount music selection, the music piece having the closest distance from the designated position is selected from the music pieces in the music database 1. In any music selection, as in step S17, the related information of each selected music is displayed on the display unit 7 as a reproduction music list. In the normal music selection process, the music selection processing unit 5 may select a plurality of music pieces for one music selection condition, or accept a plurality of music selection conditions.

  The reproduction music list obtained in the above step S2 or S4 is output from the music selection processing unit 5 to the control unit 4. In step S5, the control unit 4 initializes a counter N representing the music selection number in the reproduction music list, that is, N = 1. At this time, the control unit 4 detects the list length L of the reproduction song list, that is, the number of songs included in the reproduction song list.

  After step S5, in step S6, the control unit 4 causes the reproduction unit 3 to reproduce the music. Specifically, the playback unit 3 is notified of the music number of the music number of the music selection number corresponding to the value of the counter N. The reproduction unit 3 reads out the music data of the music corresponding to the music number from the music database 1 and plays it back.

  After step S6, in step S7, the control unit 4 receives a reproduction operation performed by the user during the reproduction of the music reproduced in step S6 via the input reception unit 6. Specifically, the reproduction operation is an operation for skipping the music being reproduced. When a playback operation is received during playback of a music piece, the control unit 4 stops the music piece being played back. Further, the update unit 2 is instructed to update the music selection priority of the music. Specifically, the music number of the music and the playback time of the music are passed to the update unit 2. When the reproduction of the music is completed to the end without performing the reproduction operation, the control unit 4 determines that the situation is the same as the situation that “the skip button was pressed at the end of the reproduction of the music”, and the music of the music The number and the reproduction time of the music piece (in this case, the reproduction time and the performance time are equal) are passed to the update unit 2.

In step S8, the update unit 2 performs an update process. The update process is a process of updating the contents of the music database 1 using information related to the update received from the control unit 4. FIG. 10 is a flowchart showing the detailed processing flow of step S7 shown in FIG. In the update process, first, in step S21, the update unit 2 determines the music of the music based on the reproduction time of the music reproduced in the immediately preceding step S6 and the performance time of the music stored in the music database 1. Completion rate is calculated. The completion rate PR of the music is calculated by the following equation (8), where PT is the reproduction time and ST is the performance time of the music.
PR = PT / ST (8)
From Equation (8), the completion rate when the reproduction operation is not performed and the music is reproduced to the end is PR = 1.0.

  In step S22, the update unit 2 determines whether or not the completion rate PR is smaller than 0.4. If the completion rate PR is less than 0.4, the update unit 2 performs the process of step S23. That is, in step S23, the preference score of the music reproduced in the immediately preceding step S6 is decreased by one. That is, the preference score of the music stored in the music database 1 is reduced by 1 and updated. At this time, if the preference score becomes smaller than 0 as a result of the subtraction, the preference score is set to 0. After step S22, the update unit 2 ends the update process.

  On the other hand, when the completion rate PR is 0.4 or more, the update unit 2 performs the process of step S24. That is, in step S24, the update unit 2 updates the reproduction history of the music reproduced in the immediately preceding step S6. Specifically, the number of reproductions of the music stored in the music database 1 is updated by one, and the last reproduction date is updated to today's date.

  In step S25 following step S24, the update unit 2 determines whether or not the completion rate PR is greater than 0.8. If it is determined in step S25 that the completion rate PR is greater than 0.8, the update unit 2 increases the preference score by 1 in step S26. That is, the preference score of the music stored in the music database 1 is updated by one. After step S26, the update unit 2 ends the update process. On the other hand, if it is determined in step S25 that the completion rate PR is 0.8 or less, the updating unit 2 ends the updating process.

  In the updating process of step S8 described above, the number of reproductions and the last reproduction date are updated when the completion rate PR is larger than 0.8, that is, when the user listens to the music for a long time. Therefore, if the user performs a skip operation immediately after the music is played, the playback history is not updated. That is, if the user performs a skip operation immediately after the music is played, the music is not considered to be the user's preference, so the music selection device does not update the playback history. This makes it possible to create a playback history that more accurately reflects user preferences. If the playback history is not updated for all skipped songs, the playback history is not updated even if only the ending portion is skipped, and as a result, the user's preferences may not be correctly reflected. On the other hand, in the update process, even if the music has not been played to the end, the music that has been played to some extent is determined as the music that the user likes, and the playback history is updated. Accordingly, it is possible to update the reproduction history that correctly reflects the user's preference.

  Note that the update process does not have to be executed when a music piece to be reproduced next to the currently reproduced music piece is displayed on the display unit 7. This is because when the music to be reproduced next is displayed, the user may skip the music currently being reproduced in order to listen to the music displayed next. In such a case, since the relevance between the skip operation and the preference of the currently played music is weak, the update process may not be performed. As described above, the content of the update process may be changed according to the display content of the display unit 7.

  Returning to the description of FIG. 4 again, in step S9, the control unit 4 determines whether or not all the music pieces included in the reproduction music list have been reproduced. Specifically, it is determined whether or not the reproduction music list length L and the counter N representing the current music selection number match. If the playback song list length L matches the current song selection number N, it is determined that all the songs included in the playback song list have been played. If it is determined in step S9 that all the songs included in the playback song list have been played, the control unit 4 performs the process of step S1. Thereafter, the processes of steps S1 to S10 are repeated. On the other hand, if it is determined in step S9 that at least one of the songs included in the reproduction song list has not been reproduced, the control unit 4 performs the process of step S10. That is, in step S10, the control unit 4 increases the value of the counter N by 1. Further, after step S10, the process of step S6 is performed. Thereafter, the control unit 4 repeats the processes of steps S6 to S10 until all the music pieces included in the reproduction music list are reproduced. Above, description of the flowchart shown in FIG. 4 is complete | finished.

  The music selection apparatus shown in FIG. 1 can be realized by causing a general computer system to execute a predetermined program. FIG. 11 is a block diagram showing a configuration when the music selection apparatus shown in FIG. 1 is realized by a computer system. In FIG. 11, the computer system includes a computer 30 that is a music selection device and a recording medium 37 that can be read by the computer 30. The computer 30 includes an arithmetic processing device 31, a music playback device 32, a disk drive device 33, a display device 34, an input device 35, and a storage device 36.

  The recording medium 37 stores a program for causing the computer to function as a music selection device. As the recording medium 37, any type of recording medium readable by a computer such as a flexible disk or an optical disk can be used. The disk drive device 33 reads the program from the recording medium 37. The arithmetic processing unit 31 includes a CPU, a memory, and the like, and implements the functions of the control unit 4, the music selection processing unit 5, and the update unit 2 illustrated in FIG. 1 by executing the program. The music database 1 shown in FIG. 1 is realized by the storage device 36. Note that the program may be supplied to the computer by a recording medium, or may be supplied to the computer by data distribution via the Internet or the like. By installing the program on an arbitrary computer, the computer can function as a music selection device.

  The process in the music selection apparatus may be executed by a hardware device built in or connected to the computer, or may be executed by software on a part of the process on the software. .

  In the present embodiment, the playback operation is described as a skip operation. However, the playback operation may include a cue operation for returning to the beginning of the music or a stop operation for stopping the music playback. . In the case of “accepting a reproduction operation” in step S7 shown in FIG. 4, if a cueing operation is performed, the process returns to step S6 without performing the process of step S10, and the music is reproduced from the beginning. . In the case of “stop operation”, it is only necessary to return to step S1 without determining the condition in step S9. Even in the case of a cueing operation or a stop operation, the update process in step S8 is performed, so the update process for the music database 1 is performed with the same contents as the skip operation.

  Moreover, you may enable it for a user to input a preference degree directly during reproduction | regeneration of a music. For example, the user may be able to perform an operation of increasing / decreasing the preference score by 1 while playing music.

  In the first embodiment, the update process is performed when either the automatic music selection process or the normal music selection process is performed. However, in other embodiments, it is not always necessary to do so. For example, in the automatic music selection mode, since the user does not select music by his / her own intention, performing the update process on the music selection result of the automatic music selection process obtains a music selection priority that accurately reflects the user's preference. Is not limited. Therefore, when the automatic music selection process is performed, a part or all of the update process may be omitted.

  In the first embodiment, the last reproduction date is used as the reproduction history. However, in other embodiments, the reproduction date (reproduction time) may be used as the reproduction history. FIG. 12 is a diagram illustrating an example of data stored in the music database in the modification of the first embodiment. In FIG. 12, not only the last reproduction date but also all reproduction dates so far are stored in the music database 1. The music database may store data as shown in FIG. When the music database shown in FIG. 12 is used, for example, among the reproduction dates stored in the music database, a reproduction date (reproduction date indicating June to August) that matches the current season (for example, summer). The music selection process is performed using only. Thereby, the music selection result according to the season can be obtained. Further, by storing the reproduction date information in detail in units of seconds, for example, it is possible to obtain a music selection result corresponding to the time zone.

(Embodiment 2)
Next, a music selection apparatus according to Embodiment 2 of the present invention will be described. The music selection apparatus according to Embodiment 2 performs the music selection process by using additional information in addition to the impression amount and the music selection priority. The configuration of the music selection device is the same as that shown in FIG. Hereinafter, differences from the first embodiment will be mainly described.

  FIG. 13 is a diagram showing an example of data stored in the music database in the second embodiment. The music database shown in FIG. 13 is obtained by adding additional information to the music database shown in FIG. Here, it is assumed that the additional information is, for example, information on “applicable season”. The suitable season is information indicating which season the music is suitable for. For example, the music database shown in FIG. 13 indicates that the music having the music number “1” is suitable for the spring season. Note that the additional information may not be automatically assigned, or may not be assigned to all music pieces. The additional information may be input by the user himself at the time of registration or after registration of the music, or may be additionally input from the outside.

Hereinafter, the music selection process using the additional information will be described. In the second embodiment, the music selection processing unit 5 uses the music selection reference AS (i) based on the additional information when calculating the music selection value ES (i) described in the first embodiment. Here, AS (i) is expressed as the following formula (9).
AS (i) = 3 (matching season of music i = “summer”)
AS (i) = 1 (matching season of music i = “spring” or “autumn”)
AS (i) = 0 (matching season of music i = “winter”) (9)
As shown in Expression (9), AS (i) is set in accordance with the degree of adaptation between the adaptation season of the music and the current season. Equation (9) is used when the current season is summer, and when the current season is other than summer, equations other than equation (9) are used.

When there are a plurality of types of additional information, the music selection criterion AS (i) based on the additional information is calculated according to the equation (10).
AS (i) = ΣwA (k) * AS (i, k) (k is an item of additional information) (10)
In Expression (10), wA (k) is a coefficient representing the degree of reflection of each item on the selection criterion AS (i).

In the second embodiment, the music selection value ES is calculated using the following equation (11) instead of the equation (6).
ES (i) = w1 * (wES1 * LS (i) + wES2 * TS (i)) + w2 * AS (i) (11)
As shown in Expression (11), ES (i) in the second embodiment is added by multiplying each of ES (i) in the first embodiment and the music selection criterion AS (i) based on the additional information with weights. Obtained by. That is, in Expression (11), w1 is a coefficient representing the degree of reflection of ES (i) in the first embodiment, and w2 is a coefficient representing the degree of reflection of AS in music selection. In step S15, the music selection processing unit 5 calculates the music selection value ES using the equation (11).

  When the number of reproduction histories is small, that is, when the number of reproductions is too small, there is a possibility that the reproduction histories so far do not sufficiently reflect the user's preference. If music selection using the playback history is performed in such a state, the user's actual preference may not be accurately reflected. Therefore, when the reproduction history is small, music selection using AS (i) may be performed. Alternatively, the coefficient w1 in the equation (11) may be reduced and the coefficient w2 may be increased. As a result, the music suitable for the current situation is selected, so that the reproduction history can be enhanced while performing the music selection with high user satisfaction.

(Embodiment 3)
Next, a music selection apparatus according to Embodiment 3 of the present invention will be described. The music selection apparatus according to Embodiment 3 performs a setting process for allowing the user to input a preference level for each area at an arbitrary timing. In this way, the third embodiment enables music selection processing that reflects user preferences even when the number of playbacks is small. The configuration of the music selection device is the same as that shown in FIG. Hereinafter, differences from the first embodiment will be mainly described.

  FIG. 14 is a diagram showing an example of data stored in the music database in the third embodiment. In the music database 1 in the third embodiment, the area preference is set for each music. The area preference level indicates the user's preference level for each area. In the third embodiment, the music selection process is performed using the music selection priority including the area preference level. Moreover, the area preference degree of the music which belongs to the same area becomes the same value. In the third embodiment, it is assumed that the area into which the impression space is divided is fixedly determined.

  FIG. 15 is a flowchart showing the flow of setting processing in the third embodiment. The setting process is executed at an arbitrary timing. For example, the setting process is executed when the music selection device is activated for the first time or when the execution is instructed by the user. First, in step S31, the music selection processing unit 5 selects music from each area of the impression space. In step S31, in order to calculate the area preference level for many areas, it is preferable to select a plurality of music pieces whose mutual distance in the impression space is larger than a predetermined distance. The music selection processing unit 5 selects music for an area that has not yet been selected in the loop of steps S31 to S33.

  In step S32, an input from the user is accepted. That is, the user inputs an evaluation of likes or dislikes about the music selected in step S31. FIG. 16 is a diagram illustrating an example of a button provided in the input receiving unit 6 for performing an input in step S31. As shown in FIG. 16, the input receiving unit 6 includes a button 41 indicating “like” and a button 42 indicating “dislike”. The user evaluates each music piece by pressing one of the buttons 41 and 42. In the third embodiment, the user is allowed to input the evaluation of the music with two values of “like” or “dislike”, but the user may be allowed to input a numerical value. Further, in step S32, the control unit 4 preferably causes the playback unit 3 to play back the music to be evaluated.

  In step S33, the music selection processing unit 5 determines whether or not input has been performed for all areas to be evaluated. Here, “all areas to be evaluated” does not necessarily mean all areas in the impression space. For example, the evaluation may be performed only for an area where the number of times of reproduction of music is small. When input is performed for all areas to be evaluated, the music selection processing unit 5 performs the process of step S34. On the other hand, if there is no input for at least one area to be evaluated, the process of step S31 is performed. Thereafter, the music selection processing unit 5 repeats the processes of steps S31 to S33 until input is performed for all areas to be evaluated.

In step S34, the music selection processing unit 5 calculates the area preference level. Specifically, the area preference MEA (i) of the area to which the music i belongs is calculated by using the following equation (12).
MEA (i) = 1 (when evaluated as “like”)
MEA (i) = 0.1 (when evaluated as “dislike”) (12)

In the third embodiment, the music selection processing unit 5 performs music selection using the area preference degree. That is, the music selection processing unit 5 calculates the area selection value EA using the following equation (13) instead of the equation (3).
EA (j) = Σ (wEA1 * LA (i) + wEA2 * TA (i) + wEA3 * MEA (i)) (13)
In Equation (13), wEA3 is a coefficient representing the degree to which the area preference level MEA is reflected in area selection. The result of evaluation performed by the user in the setting process is reflected in the area selection by Expression (13).

  In the third embodiment, only one piece of music is selected for each area, but a plurality of music pieces may be selected for one area. At this time, the area selection value may be calculated as an average value of the area preference degrees calculated for the plurality of music pieces.

(Embodiment 4)
Next, a music distribution system according to Embodiment 4 of the present invention will be described. FIG. 17 is a diagram showing a music distribution system according to Embodiment 4 of the present invention. The music distribution system includes a server 50, a terminal 60, and a network 70. Connected to the network 70 are a server 50 that distributes music data and a terminal 60 that receives the distributed music data. In addition, a plurality of terminals having functions similar to those of the terminal 60 are connected to the network 70.

  In FIG. 17, the server 50 includes a music database 51, an update unit 52, a control unit 53, a music selection processing unit 54, and a communication unit 55. In addition, each component except the communication part 55 among each component which the server 50 has has the same function as the component of the same name shown in FIG. The communication unit 55 transmits / receives data to / from each terminal via the network 70. Further, the terminal 60 includes a reproduction unit 61, an input reception unit 62, a display unit 63, and a communication unit 64. In addition, each component except the communication part 64 among each component which the terminal 60 has has the same function as the component of the same name shown in FIG. The communication unit 64 transmits and receives data to and from the server 50 via the network 70. As described above, the music distribution system shown in FIG. 17 has a configuration in which the components of the music selection device shown in FIG. 1 are divided into the server side and the terminal side. The operation of the music distribution system is the same as that of the music selection apparatus shown in FIG. 1 except that communication between the server 50 and the terminal 60 is performed via the network 70. With the configuration shown in FIG. 17, it is possible to download and play music data from the server 50 even when the user terminal 60 does not have the music database 1.

  According to the fourth embodiment, when the user instructs automatic music selection on the terminal 60 side, automatic music selection processing is performed on the server 50 side. As a result, the reproduction music list is transmitted from the server 50 to the terminal 60. Further, music data included in the reproduction music list is transmitted from the server 50 to the terminal 60. That is, the communication unit 55 of the server 50 transmits the music data to the terminal 60. The communication unit 64 of the terminal 60 receives music data transmitted from the communication unit 55 of the server 50. The received music data is played back by the playback unit 61. The data of each song may be transmitted in a batch for the songs included in the playback song list, or every time music playback is started (if a skip operation is performed, the skip operation is performed). It may be done one by one).

  The music selection apparatus according to the present invention can be used as an audio player, a jukebox, or the like.

The block diagram which shows the music selection apparatus which concerns on Embodiment 1. FIG. The figure which shows an example of the data stored in the music database shown in FIG. The figure which shows the example of the impression space in Embodiment 1 The flowchart which shows the flow of a process of the music selection apparatus which concerns on Embodiment 1. FIG. The figure which shows an example of the button provided in the input reception part 6 for selecting a music selection mode. The flowchart which shows the detailed process of step S2 of FIG. The figure which shows an example of a mode that impression space was divided into a plurality of areas The figure which shows an example of the reproduction | regeneration music list displayed on the display part 7 The figure which shows an example of the input screen for accepting music selection conditions The flowchart which shows the flow of the detailed process of step S7 shown in FIG. The block diagram which shows the structure at the time of implement | achieving the music selection apparatus shown in FIG. 1 with a computer system. The figure which shows an example of the data stored in the music database in the modification of Embodiment 1 The figure which shows an example of the data stored in the music database in Embodiment 2 The figure which shows an example of the data stored in the music database in Embodiment 3 Flowchart showing the flow of setting processing in the third embodiment The figure which shows an example of the button provided in the input reception part 6 for performing input of step S31 The figure which shows the music distribution system which concerns on Embodiment 4 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Music database 2 Update part 3 Playback part 4 Control part 5 Music selection process part 6 Input reception part 7 Display part

Claims (18)

A music selection device that selects music to be played and presents it to the user,
A music data storage unit for storing music data;
An impression amount storage unit that stores an impression amount obtained by quantifying a listener's impression of a song for each song in which data is stored in the song data storage unit;
A music selection priority storage unit that stores, for each of the music pieces, a music selection priority that represents at least one of the music playback history and the music preference;
A music selection processing unit that performs music selection processing for selecting a music to be played from each of the music pieces, using a distance between the music pieces in an impression space having each component of the impression amount as a coordinate axis and the music selection priority;
The music selection processing unit
An area determination unit that determines one of the areas obtained by dividing the impression space into a plurality of areas based on the music selection priority of each music;
As for Music impression amount belongs to the determined area by the area determination unit, and a music selection execution unit for performing music selection processing using the music selection priority, comfortable music song selection system. A music selection device that selects music to be played and presents it to the user,
A music data storage unit for storing music data;
An impression amount storage unit that stores an impression amount obtained by quantifying a listener's impression of a song for each song in which data is stored in the song data storage unit;
A music selection priority storage unit that stores, for each of the music pieces, a music selection priority that represents at least one of the music playback history and the music preference;
A music selection processing unit that performs music selection processing for selecting a music to be played from each of the music pieces, using a distance between the music pieces in an impression space having each component of the impression amount as a coordinate axis and the music selection priority;
The music selection processing unit
A degree-of-association calculating unit that calculates a degree of relevance obtained by changing the music selection priority of the other music according to the positional relationship with the other music in the impression space;
The probability corresponding to the music selection priority and relevance of each song, and a music selection unit for selecting the music to be reproduced, comfortable music song selection system. A music selection device that selects music to be played and presents it to the user,
A music data storage unit for storing music data;
An impression amount storage unit that stores an impression amount obtained by quantifying a listener's impression of a song for each song in which data is stored in the song data storage unit;
A music selection priority storage unit that stores, for each of the music pieces, a music selection priority that represents at least one of the music playback history and the music preference;
A music selection processing unit that performs music selection processing for selecting a music to be played from each of the music pieces, using a distance between the music pieces in an impression space having each component of the impression amount as a coordinate axis and the music selection priority;
The music selection processing unit
A music selection priority calculating unit that calculates music selection priority at an arbitrary position in the impression space based on the music selection priority of each music;
A reference position determination unit that determines a reference position in the impression space with a probability according to the music selection priority calculated by the music selection priority calculation unit;
Wherein the impression space based on a positional relationship between the reference position and the position of each song, and a music determination unit that determines music to be reproduced, comfortable music song selection system. The music song selection apparatus according to any one of claims 1 to 3 , further comprising a reproduction unit that reproduces the music selected by the music selection processing unit using the data stored in the music data storage unit. The music selection priority includes, as information indicating the reproduction history, at least one of the number of times the music is reproduced and the date and time when the music data was last reproduced,
A music designation accepting unit that accepts input for designating music to be played from the user;
5. The music piece according to claim 4 , further comprising: a reproduction history update unit that updates the reproduction history stored in the music selection priority storage unit according to the input when an input is received by the music designation reception unit. Music selection device. The music selection priority includes at least information indicating a preference level,
A preference setting accepting unit that accepts an input from a user for setting a preference for the music played by the playing unit;
5. The music piece according to claim 4 , further comprising: a preference level update unit that updates a preference level stored in the music selection priority storage unit based on the input when an input is received by the preference level setting reception unit. Music selection device. A test music selection unit that selects a plurality of music pieces each having a distance greater than a predetermined distance in the impression space as a test music piece from among the music pieces stored in the music data storage unit; ,
The music playback device according to claim 6 , wherein the playback unit plays back the test music in response to selection of the test music by the test music selection unit. A stop accepting unit that accepts an input to stop the reproduction of the music by the playing unit from a user;
Depending on the ratio of the portion reproduced by the reproduction unit for the entire song, further comprising a music selection priority update unit for updating the music selection priority stored in the music selection priority storage unit, according to claim 4 Song selection device. A music selection method used in a music selection device that selects and presents music to be played to the user,
A music data acquisition step for acquiring music data;
An impression amount acquisition step of acquiring an impression amount obtained by quantifying the impression of a listener with respect to a song for each song for which data has been acquired in the song data acquisition step;
A music selection priority acquisition step for acquiring a music selection priority representing at least one of a music reproduction history and a music preference level for each music;
A music selection process for selecting a music to be played from each of the music, using a distance between the music in the impression space with each component of the impression amount as a coordinate axis and the music selection priority,
The music selection processing step includes
A region determining step for determining one of the regions obtained by dividing the impression space into a plurality of regions based on the music selection priority of each music;
The music impression amount belongs as an object within the determined region in the reference position determining step, and a music selection execution step of performing music selection processing using the music selection priority, comfortable music song selection method. A music selection method used in a music selection device that selects and presents music to be played to the user,
A music data acquisition step for acquiring music data;
An impression amount acquisition step of acquiring an impression amount obtained by quantifying the impression of a listener with respect to a song for each song for which data has been acquired in the song data acquisition step;
A music selection priority acquisition step for acquiring a music selection priority representing at least one of a music reproduction history and a music preference level for each music;
A music selection process step of selecting a music to be played from each of the music using a distance between the music in the impression space having each component of the impression amount as a coordinate axis and the music selection priority;
The music selection processing step includes
A relevance level calculating step for calculating a relevance degree obtained by changing the music selection priority of the other music according to the positional relationship with the other music in the impression space;
The probability corresponding to the music selection priority and relevance of each song, and a music selection step of selecting the music to be reproduced, easy music song selection method. A music selection method used in a music selection device that selects and presents music to be played to the user,
A music data acquisition step for acquiring music data;
An impression amount acquisition step of acquiring an impression amount obtained by quantifying the impression of a listener with respect to a song for each song for which data has been acquired in the song data acquisition step;
A music selection priority acquisition step for acquiring a music selection priority representing at least one of a music reproduction history and a music preference level for each music;
A music selection process for selecting a music to be played from each of the music, using a distance between the music in the impression space with each component of the impression amount as a coordinate axis and the music selection priority,
The music selection processing step includes
A music selection priority calculating step for calculating a music selection priority at an arbitrary position in the impression space based on the music selection priority of each music;
A reference position determining step for determining a reference position in the impression space with a probability corresponding to the music selection priority calculated in the music selection priority calculating step;
Wherein the impression space based on a positional relationship between the reference position and the position of each song, and a music determining step of determining the music to be reproduced, easy music song selection method. The music selection method according to any one of claims 9 to 11 , further comprising a reproduction step of reproducing the music selected in the music selection processing step using the data acquired in the music data acquisition step. The music selection priority includes, as information indicating the reproduction history, at least one of the number of times the music is reproduced and the date and time when the music data was last reproduced,
A music designation accepting step for accepting an input for designating a music to be played from the user;
The music selection according to claim 12 , further comprising a reproduction history update step of updating the reproduction history acquired in the music selection priority acquisition step based on the input when an input is received by the music specification reception step. Method. The music selection priority includes at least information indicating a preference level,
A preference setting accepting step for accepting an input from a user for setting a preference for the music reproduced by the reproduction step;
The music composition selection according to claim 12 , further comprising: a preference degree update step of updating the preference degree acquired in the music selection priority acquisition step based on the input when an input is accepted by the preference degree setting acceptance step. Method. A test music selection step of selecting a plurality of music pieces whose mutual distance in the impression space is larger than a predetermined distance as a test music piece from among the music pieces for which data has been acquired in the music data acquisition step;
15. The music selection method according to claim 14 , wherein the reproduction step reproduces the test music in response to selection of the test music in the test music selection step. A stop accepting step for accepting an input to stop the reproduction of the music by the reproduction step from the user;
Depending on the ratio of the portion reproduced by said reproducing step for the entire song, further comprising a music selection priority updating step of updating the music selection priority acquired in the music selection priority acquisition step, according to claim 12 Music selection method. The computer of the music selection device according to any one of claims 1 to 3, wherein the music to be reproduced is selected and presented to the user.
Music data storage means for storing music data;
Impression amount storage means for storing an impression amount obtained by quantifying the impression of a listener for a song for each song for which data is stored in the song data storage means;
Music selection priority storage means for storing a music selection priority representing at least one of the music reproduction history and music preference for each music;
A program that functions as music selection processing means for performing music selection processing for selecting music to be played from each of the music pieces using the distance between the music pieces in the impression space with each component of the impression amount as a coordinate axis and the music selection priority. . A computer-readable recording medium on which the program according to claim 17 is recorded.

Images