JP2008243345A - Audio device, play list preparing device, and audio playback program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To change a melody close to a desired melody as a track is reproduced, also to reduce variation of the melody. <P>SOLUTION: In an audio device, coordinate points Ptr of respective audio files are arranged in an xy coordinate system based on feature quantity in which tempo of a track and sound quality are indexed. Further in the audio device, a vector VC connecting a head track Pstr and a tail track Petr is divided into a plurality of vectors, also a plurality of areas A1 to A24 and sections S1 to S4 arranged in a line in the direction of the vector VC are prepared. Then, tracks are selected from respective sections so that at the section closer to the tail track Petr out of the sections S1 to S4, more tracks are selected. The audio device plays successively the selected plurality of tracks in the ascending order of section number of the track and area number. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an audio device, a playlist creation device, and an audio playback program, and more particularly to an audio device, a playlist creation device, and an audio playback program that can store a plurality of audio files.

  Recently, audio devices equipped with a large-capacity hard disk capable of storing a large amount of audio files (hereinafter referred to as tracks) have appeared. These audio apparatuses have a mechanism for continuously reproducing a track that meets a user's request from a large number of tracks. For example, in the audio device disclosed in Japanese Patent Application Laid-Open No. 2002-55993 (Patent Document 1), a track and a database in which the genre and tone of the track are classified are stored, and a track of a genre and tone specified in advance is selected. And play it.

  In addition, various techniques have been developed to analyze the tone of each track, such as the tempo, sound quality, and rhythm, and index it as a musical feature, and select a track that meets the user's request based on the indexed feature. Yes. For example, the audio apparatus disclosed in Japanese Patent Application Laid-Open No. 2006-39704 (Patent Document 2) creates a search space in which coordinate points of each track are distributed in a coordinate system having a feature amount as an axis, and displays the search space on a display. Then, when the user designates a predetermined range in the search space, the tracks included in the range are listed in the playlist and reproduced.

By the way, when a track is reproduced continuously at bedtime or when waking up, it is preferable to change the tone of the reproduced track to a desired tone every time the track is reproduced. Further, the change in tune is large at the beginning of playback, and it is preferable that the change in tune gradually decreases as playback is performed. For example, when playing a track continuously at bedtime, even if a track with a dynamic (active) tone is played at first, the track gradually changes to a static (calm) tone as the track is played. Is preferred. Furthermore, it is better to sleep at the beginning of playback when there is a large change from dynamic to static tones, and at the end of playback there is little change in tones and static tracks are played continuously. . In addition, when waking up, a track with a static tone is played at the beginning, and as the track is played, the track gradually changes to a dynamic tone, and the change in tone decreases with the playback, and the end of playback In this case, it is preferable that a track having a dynamic tone is continuously reproduced.
JP 2002-55993 A JP 2006-39704 A Japanese Utility Model Publication No. 6-77082 Japanese Patent Laid-Open No. 10-162025

  An object of the present invention is to provide an audio apparatus capable of reducing the change in tune as the playback progresses as the tune is brought closer to a desired tune as the tracks are played back. That is.

Means for Solving the Problems and Effects of the Invention

  The audio apparatus according to the present invention comprises storage means, arrangement means, area creation means, selection means, and playback means. The storage means stores a plurality of audio files and feature amounts of the audio files. The arrangement unit arranges the coordinate points of the audio file in a coordinate system having a coordinate axis corresponding to the feature amount based on the feature amount. The specifying means specifies a head audio file and a tail audio file. The area creating means divides a vector connecting the coordinate point of the first audio file to the coordinate point of the last audio file into a plurality of areas in the coordinate system, and creates a plurality of areas arranged in a line in the vector direction. The selection means includes one or more specified by the coordinate points included in each area so that the closer to the coordinate point of the last audio file among the plurality of created areas, the larger the number of audio files to be selected. Select an audio file. The reproduction means sequentially reproduces the plurality of selected audio files in an order in which the area including the coordinate point of the audio file is close to the coordinate point of the head audio file. Here, the feature amount is a numerical index of the tune represented by the tempo, sound quality, rhythm, etc. of the audio file.

  The audio device according to the present invention brings the tune closer to that of the last audio file as the audio file is played back. Furthermore, since the number of audio files selected in the area closer to the end audio file is larger, the change in tune is large at the beginning of playback, but the change in tune becomes smaller as playback is performed.

  Preferably, the audio apparatus further includes an area creating unit. The area creating means divides each section into a plurality of areas and creates a plurality of areas arranged in a line in the vector direction. The reproduction means sequentially reproduces the plurality of selected audio files in the order in which the area including the coordinate point of the audio file is close to the coordinate point of the head audio file.

  In this case, the section is composed of a plurality of areas, and the audio files included in the area close to the coordinate point of the first audio file are reproduced in order. Therefore, every time an audio file is played, the tune can be made closer to the end file more accurately, and the change in tune can be gradually reduced.

  Preferably, the storage unit further stores a reproduction time of each of the plurality of audio files. The audio device further includes timer setting means and distribution setting means. The timer setting means sets a timer time. The allocation setting means sets the allocation time of each area so that the total allocation time of each area is equal to or less than the timer time, and the allocation time becomes longer for the area closer to the coordinate point of the end audio file among the plurality of areas. Set. The selection means selects the audio file so that the total reproduction time of the audio file selected in each area is within the distribution time set by the distribution setting means.

  In this case, an audio file can be selected from each area according to the timer time.

  Preferably, the audio apparatus further includes a total reproduction time calculation unit, a difference calculation unit, and an inter-song time calculation unit. The calculation means calculates the total reproduction time of the audio file selected by the selection means as the total reproduction time. The difference calculation means calculates a difference value between the timer time and the total playback time of the selected audio file. The inter-music time calculating means calculates the inter-music time based on the difference value and the number of selected audio files. The reproduction means reproduces the next audio file after the inter-song time has elapsed after completing the reproduction of the audio file.

  In this case, the reproduction of the selected audio file can be completed with the end of the timer time.

  Preferably, when the audio file is being played back, the specifying unit specifies the audio file being played back as the first audio file.

  In this case, even when the audio file is being reproduced, the audio file can be sequentially reproduced so as to approach the tone of the tail audio file from the audio file being reproduced.

  The playlist creation device according to the present invention creates a playlist for playing back an audio file based on a playlist in which a plurality of audio files are listed. The playlist creation device includes an acquisition unit, an arrangement unit, an identification unit, a zone creation unit, a selection unit, and a playlist creation unit. The acquisition unit acquires feature amounts of a plurality of audio files. The arrangement unit arranges the coordinate points of the audio file in a coordinate system having a coordinate axis corresponding to the feature amount based on the feature amount. The specifying means specifies a head audio file and a tail audio file. The area creating means divides a vector connecting the coordinate point of the first audio file to the coordinate point of the last audio file into a plurality of areas in the coordinate system, and creates a plurality of areas arranged in a line in the vector direction. The selection means includes one or more specified by the coordinate points included in each area so that the closer to the coordinate point of the last audio file among the plurality of created areas, the larger the number of audio files to be selected. Select an audio file. The playlist creation means enumerates a plurality of selected audio files so that the selected audio files are played in the order in which the area including the coordinate point of the audio file is close to the coordinate point of the first audio file. Create a list.

  The playlist creation apparatus according to the present invention creates the above-described playlist. Therefore, when reproducing based on a playlist, the tune approaches the tune of the last audio file as the audio file is played. Furthermore, since the number of audio files to be selected is larger in the area closer to the end audio file, the change in tune gradually becomes smaller as the audio file is played back.

  Preferably, the playlist creation device further includes area creation means. The area creating means divides each section into a plurality of areas and creates a plurality of areas arranged in the vector direction. The playlist creation means adds the selected audio files to the playlist so that the selected audio files are played in the order in which the area including the coordinate point of the audio file is close to the coordinate point of the first audio file. Enumerate.

  In this case, the section is composed of a plurality of areas, and the audio files included in the area close to the coordinate point of the first audio file are reproduced in order. Therefore, every time an audio file is played, the tune can be made closer to the end file more accurately, and the change in tune can be gradually reduced.

  The audio reproduction program according to the present invention is a program that can be executed by a computer that stores a plurality of audio files and feature amounts of the audio file, and causes the computer to function as the above-described audio device.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

[overall structure]
Referring to FIG. 1, audio device 1 includes a central processing unit (CPU) 11, a memory 12, a hard disk drive (HDD) 13, an input unit 14, a D / A converter 15, and a speaker 16. Prepare.

  The input unit 14 receives a command selected by the user. The input unit 14 is, for example, a keyboard, a mouse, an operation key button or the like.

  The HDD 13 stores a plurality of audio files (hereinafter referred to as tracks). The HDD 13 further stores a track table including information on a plurality of tracks. As shown in FIG. 2, the track table includes a track ID, which is a track identifier, a track reproduction time, and a track feature amount. The feature amount is an index of features such as the tempo, sound quality, and rhythm of the track, and can be obtained, for example, by analyzing the audio signal of the track. In the present embodiment, there are two types of feature amounts, a tempo value and a sound quality value, as shown in FIG. Each feature amount is indexed with a maximum value of +100 and a minimum value of −100. The greater the tempo value, the more dynamic the tune is (active tune), and the smaller the tempo value, the tune is the static (calm tune). Further, the smaller the sound quality value, the harder the tone, and the higher the sound quality value, the softer the tone.

  The feature amount of each track is registered in the HDD 13 together with each track. When each track is registered in the HDD 13, the audio apparatus 1 may analyze the audio signal of the track to determine the feature amount, or a predetermined feature amount may be input from the input unit 14.

  The HDD 13 further stores an audio playback program. The audio playback program is loaded into the memory 12 and executed by the CPU 11.

  The CPU 11 selects a track to be reproduced based on the audio reproduction program loaded in the memory 12 and creates a playlist. Therefore, it can be said that the CPU 11 and the memory 12 constitute a playlist creation device. The CPU 11 also reproduces a track based on the playlist.

  The memory 12 includes a ROM (Read Only Memory) and a RAM (Random Access Memory). The D / A converter 15 converts the selected track from a digital signal to an analog signal and outputs it to the speaker 16. The speaker 16 receives an analog signal from the D / A converter 15 and outputs it as sound.

[Operation overview]
The audio device 1 selects a plurality of tracks in advance when executing the audio reproduction process. At this time, the audio device 1 plays back the selected tracks in order of reproduction so that the tune gradually approaches the tune desired by the user and the change in tune between tracks becomes small as the tracks are sequentially reproduced. To decide. Therefore, when the user goes to bed, the audio apparatus 1 changes the music to be reproduced from a song having a dynamic tempo and a hard tone to a song having a static tempo and a soft tone. Can be migrated. Furthermore, in the initial stage of reproduction, the change in the tone of the track that is continuously played is large, but the change in the tone is gradually reduced. Therefore, the user can listen to gradually calmed songs from up-tempo songs. On the other hand, when the user wakes up, the audio device 1 can gradually shift the track to be reproduced from a calm song to an up-tempo song and gradually reduce the change in the tone of the song. The audio playback process will be described below.

[Audio playback processing flow]
The audio apparatus 1 sets the timer time and executes the audio reproduction process. The audio playback process includes a wake-up playback process that sequentially plays a track with a large tempo value and a small sound quality value from a track with a small tempo value and a large sound quality value, and a sound quality with a small tempo value from a track with a large tempo value and a small sound quality value. There is a sleep reproduction process in which tracks having large values are sequentially reproduced. Hereinafter, the sleep reproduction process will be described as an example.

  Referring to FIG. 3, first, the user of audio device 1 instructs sleep reproduction processing using input unit 14. Subsequently, the user sets a timer time Tref. At this time, based on a user operation, the input unit 14 receives an input of a timer time Tref (for example, 90 minutes) (S1).

  Subsequently, the audio apparatus 1 specifies the first track STR of the sleep reproduction process (S2). At this time, the audio apparatus 1 identifies the currently playing track as the first track STR. The track number of the track being reproduced is registered in the memory 12. If the track is not being played back, the audio apparatus 1 prompts the user to select the first track STR. For example, the audio apparatus 1 displays information on selectable tracks on a display (not shown), and causes the user to select the first track STR using the input unit 14. A plurality of head tracks STR corresponding to the sleep playback process are registered in the memory 12 in advance, and when an instruction for the sleep playback process is received, an arbitrary track is selected from the plurality of head tracks STR registered in the memory 12. May be selected.

  Subsequently, the CPU 11 specifies a track to be reproduced last (hereinafter referred to as an end track ETR) (S3). The CPU 11 selects the last track according to the type of audio playback processing. First, the CPU 11 confirms that the selected reproduction process is a sleep process. The memory 12 stores information related to the selected reproduction process, and the CPU 11 reads out the information from the memory 12 to confirm that the type of the reproduction process is the sleep reproduction process. Subsequently, the CPU 11 selects a plurality of tracks that are candidates for the end track based on the feature amount of each track. Specifically, the CPU 11 selects a predetermined number of tracks in descending order of a sleep feature quantity SFQ (Sleep Feature Quantity) represented by the following formula (A).

Sleep feature value SFQ = sound quality value x-tempo value y (A)
That is, a track with a larger sleep feature value SFQ has a static and soft tone. The CPU 11 selects one of the selected tracks and identifies it as the end track ETR. By such a processing procedure, it is possible to prevent the end track ETR from being always constant.

  Note that the wakeup feature value WFC used to select the last track during the wakeup playback process is expressed by the following equation (B).

Wake-up feature value WFC = −sound quality value x + tempo value y (B)
In this case, the track having a larger wakeup feature value WFC has a dynamic and harder tone.

  In the above description, the end track is determined by the sleep feature value SRQ. However, the user may specify an arbitrary track as the end track using the input unit 14.

  After identifying the end track ETR in step S3, the CPU 11 calculates the remaining reproduction time ΔTstr of the first track STR being reproduced (S4). The CPU 11 calculates the remaining playback time of the first track based on the playback time in the track table and the time when playback up to the present time has ended. After calculating the remaining reproduction time ΔTstr, the CPU 11 compares the remaining reproduction time ΔTstr with the timer time Tref (S5). If the remaining playback time ΔTstr is greater than the timer time Tref (YES in S5), the audio playback process is terminated after the timer time Tref has elapsed (YES in S9). That is, at this time, the reproduction is stopped during the reproduction of the first track STR.

  When the remaining reproduction time ΔTstr is smaller than the timer time Tref (NO in S5), the CPU 11 further determines whether the total value of the remaining reproduction time ΔTstr and the reproduction time Tetr of the end track ETR is larger than the timer time Tref. (S6). The CPU 11 reads the playback time Tetr of the end track ETR from the track table. When the total value of ΔTstr and Tetr is larger than the timer time Tref (YES in S6), the CPU 11 registers only the end track ETR in the playlist PL (S8). As a result, after the reproduction of the first track STR is completed (YES in S10), the CPU 11 starts reproduction of the last track ETR registered in the playlist PL (YES in S11, S12). Since the timer ends during playback of the end track ETR, playback of the end track ETR is stopped during playback.

  On the other hand, as a result of the determination in step S6, when the total value of the remaining reproduction time ΔTstr and the reproduction time Tetr of the end track ETR is smaller than the timer time Tref (NO in S6), the CPU 11 executes sleep reproduction preparation processing. (S7).

  The CPU 11 selects a plurality of tracks listed in the playlist PL so that the total reproduction time is within the timer time Tref by the sleep reproduction preparation process. At this time, the playback order of the tracks is determined so that the tempo value decreases and the sound quality value increases as the timer time advances. Further, the CPU 11 determines the reproduction order of the tracks so that the change amount of the tempo value and the sound quality value is large in the first half of the timer, and the change amount becomes smaller as the second half of the timer is advanced. Hereinafter, the sleep reproduction preparation process will be described.

[Sleep playback preparation process]
Referring to FIG. 4, CPU 11 first executes area creation processing (S701), section creation processing (S702), and selected track number setting processing (S703). By executing these processes, the CPU 11 narrows down to some extent the tracks to be registered in the playlist PL and their enumeration order. Further, the CPU 11 does not change the feature amount of the tracks listed in the playlist PL almost in the order of reproduction, but the feature amount change amount is large in the first half of the playlist PL, and the feature amount is changed in the second half of the playlist. The tracks to be registered and their enumeration order are narrowed down to some extent so that the change amount becomes small.

  The CPU 11 further creates a playlist PL listing a plurality of tracks by a playlist creation process (S704). The area creation process (S701), section creation process (S702), selected track number setting process (S703), and playlist creation process (S704) will be described below.

[Area creation processing]
In the area creation process (S701), the CPU 11 acquires the feature amount of each track from the track table, and distributes a plurality of tracks TR in a coordinate system having xy coordinate axes corresponding to the feature amount. Then, the coordinate system in which the plurality of tracks TR are distributed is divided into a plurality of regions (hereinafter referred to as areas A1 to An) arranged in one row. At this time, the CPU 11 regards a plurality of tracks TR in the same area An as tracks TR having similar tunes. Details of the area creation process will be described below.

  Referring to FIG. 5, the CPU 11 first creates an xy coordinate system in which the x axis corresponds to the sound quality value and the y axis corresponds to the tempo value (S101). The CPU 11 plots the coordinate point Ptr of each track TR stored in the HDD 13 in the xy coordinate system based on the sound quality value and the tempo value in the track table. FIG. 6 shows a state in which the coordinate point Ptr of each track is plotted in the xy coordinate system. The coordinate point Pstr in FIG. 6 is the coordinate point of the head track STR, and the coordinate point Petr is the coordinate point of the end track ETR.

  Subsequently, the CPU 11 obtains the average reproduction time Tave of the tracks stored in the HDD 13 (S102). The CPU 11 sums the reproduction times of all the tracks TR in the track table, and divides the total value by the number of tracks registered in the track table. Thereby, the average reproduction time Tave of the plurality of tracks TR stored in the audio device 1 is determined. The average playback time Tave is stored in the memory 12. Hereinafter, the description will be continued assuming that the average reproduction time Tave is 5 minutes.

  Subsequently, the CPU 11 reads out the timer time Tref and the average reproduction time Tave from the memory 12 (S103). Then, using the timer time Tref and the average reproduction time Tave, the total area number TAN (Total Area Number) is determined by the following equation (1) (S104).

Total area number TAN = timer time Tref / average playback time Tave (1)
Since the total area number TAN is a natural number, the value after the decimal point of the value calculated by Expression (1) is rounded off. Hereinafter, the description will be continued assuming that the calculated TAN = 24.

  As shown in FIG. 7, the CPU 11 draws a vector VC connecting the coordinate point Pstr of the first track STR to the coordinate point Petr of the last track ETR on the xy coordinate system (S105). Subsequently, the CPU 11 equally divides the vector VC by the total area number TAN (S106). Specifically, as shown in FIG. 7, the CPU 11 divides the vector VC into 24 equal parts by dividing points P1 to P23. The CPU 11 stores the coordinate points of the division points P1 to P23 in the memory 12.

  Further, as shown in FIG. 8, the CPU 11 creates dividing line functions f0 (x) to f24 (x) passing through Pstr, the dividing points P1 to P23 and Petr in the coordinate system (S106). Here, the dividing line functions f0 (x) to f24 (x) are parallel to each other. That is, the distance between adjacent dividing line functions is equal. The CPU 11 registers the dividing line functions f0 (x) to f24 (x) in the memory 12.

  After creating the dividing line functions f1 (x) to f23 (x), the CPU 11 creates a plurality of areas A1 to A24. The CPU 11 sets a region between adjacent dividing line functions fn-1 (x) to fn (x) as one area An (S107). For example, referring to FIG. 8, the area between functions f0 (x) and f1 (x) is area A1, and the area between functions f1 (x) and f2 (x) is area A2. Thus, the created areas A1 to A24 divide the vector VC into a plurality and are arranged in a line in the vector direction.

  The CPU 11 registers the relationship between the function fn (x) and the area An in the area table shown in FIG. Referring to FIG. 9, for example, f0 (x) and f1 (x) are registered as functions corresponding to area A1.

  After creating the plurality of areas A1 to A24, the CPU 11 identifies the area An to which the coordinate point Ptr of each track TR belongs (S108). The CPU 11 can specify the area An to which each track TR belongs based on the functions f0 (x) to f24 (x) and the coordinate point Ptr of each track TR. For example, the area to which the track TR10 corresponding to the coordinate point Ptr10 in FIG. 10 belongs can be specified by the following method. First, the CPU 11 calculates the values of the functions f0 (x) to f24 (x) at the x coordinate = x10 of the coordinate point Ptr10. Then, the calculated value is compared with the y coordinate = y10 of the coordinate point Ptr10. As shown in FIG. 10, y10 is located between f4 (x10) and f5 (x10). As a result, the CPU 11 determines that the coordinate point Ptr10 is located between the function f4 (x) and the function f5 (x), and refers to the area table to identify the area to which the track TR10 belongs as A4.

  With the above-described method, the CPU 11 specifies an area to which all tracks stored in the HDD 13 belong. Then, it is registered in the area table shown in FIG.

  With the above operation, the CPU 11 divides the vector VC into a plurality of areas based on the head track STR and the tail track ETR, and creates a plurality of areas A1 to A24 arranged in a line in the vector direction. Then, the track TR belonging to each of the areas A1 to A24 is specified. The CPU 11 considers that the tracks TR in the same area have similar tunes.

  In the above description, the function f (x) is a curved line, but may be a straight line. Further, as shown in FIG. 11, the widths of the areas A1 to A24 may be limited to a predetermined range.

[Section creation processing]
After completing the area creation processing, the CPU 11 executes section creation processing (S702).

  In the present embodiment, a section (area) is a concept larger than an area, and each section includes a plurality of areas. In short, in the plurality of areas, each section is divided into a plurality of sections and arranged in a line in the vector direction.

  In the section creation process, the CPU 11 assigns the areas A1 to A24 to a predetermined number of sections S1 to Sn. As a result, the CPU 11 divides the vector sc connecting the coordinate point Pstr of the leading track and the coordinate point Petr of the trailing track into a plurality of sections, and creates a plurality of sections arranged in a line in the vector direction. The section creation process will be described below.

  Referring to FIG. 12, CPU 11 first acquires the number of sections SN (S201). The section number SN is a predetermined number (natural number) set in advance, and is stored in the HDD 13 or the memory 12. The number of sections SN may be set by the user using the input unit 14 or may be set in the memory 12 in advance. Hereinafter, the description will be continued assuming that the set section number SN is “4”. The CPU 11 further reads the total area number TAN from the memory 12.

  The CPU 11 calculates the number of areas (hereinafter referred to as area allocation number) AN allocated to each section based on the following equation (2) using the read section number SN and total area number TAN (S202). .

Area allocation number AN = total area number TAN / section number SN (2)
In the above example, since TAN = 24 and SN = 4, the area allocation number AN = 6. The area allocation number AN is a natural number obtained by rounding off the decimal point of the value calculated by the equation (2).

  After calculating the area allocation number AN, the CPU 11 allocates the areas A1 to A24 to the sections S1 to S4 (S203). In the above example, since the area allocation number AN is 6, 6 areas are allocated to each section. Specifically, as shown in FIG. 13, the CPU 11 includes areas A1 to A6 as section S1, areas A7 through A12 as section S2, areas A13 through A18 as section S3, and areas A19 through A24 as section S4. assign.

  The CPU 11 registers information regarding the area allocated to each section in the section table shown in FIG. In the section table, a section ID that is an identifier of each section and an area ID of an area included in each section are registered in association with each other. Sections S1 to S4 are created by the above operation.

  In step S202, when the area allocation number AN is determined by rounding off as a result of the calculation in equation (2), the allocation area number allocated to each section is determined as follows. The area allocation number AN determined in step S202 is sequentially allocated in order from the section S1. Then, the area allocation number ANe determined by the following equation (3) is allocated to the final section Sn.

Area allocation number ANe = total area number TAN−area allocation number AN × (SN−1) (3)
For example, when the number of sections SN = 4, the total number of areas TAN = 25, and the area allocation number AN = 6 determined in step S202, the area allocation number ANe of the final section S4 is 7 according to the equation (3).

[Selected track count setting process]
After completing the section creation processing, the CPU 11 executes selection track number setting processing (S703). In the selected track number setting process, the CPU 11 sets the number of tracks to be listed in the playlist for each section. This makes it possible to adjust the change in tune of a plurality of tracks that are continuously played back. Specifically, at the initial stage of sleep reproduction, the change in the tune of the continuously reproduced track can be increased, and the change in the tune can be reduced toward the end of the sleep reproduction. As a result, it is possible to suppress a change in the tune between tracks played at the end of the sleep timer, and to increase the proportion of the static soft music suitable for sleep during the timer time.

  Referring to FIG. 15, first, CPU 11 distributes timer Tref set in step S1 to each section (S301). At this time, the CPU 11 distributes the timer Tref so that the distribution time Tsn of each section Sn satisfies the following formula (4).

Tsn> Tsn-1 (4)
In short, the CPU 11 sets the distribution time Ts1 of the section S1 as shortest as possible, and sequentially sets the distribution time Tsn as the n of the section Sn increases. Then, the allocation time Ts4 of the final section S4 is set to be the longest. Hereinafter, an example of a method for setting the distribution time Tsn will be described.

  Referring to FIG. 16, a linear function y = g (x) is set in an allocation time coordinate system in which the x-axis indicates an area number. At this time, the slope of y = g (x) is positive. The CPU 11 calculates the linear function g (x) so that the integral value of the linear function g (x) from x = 0 to x = 24 (that is, the end area number) is equal to the timer time Tref set in step S1. Set.

ここで、ＡＮｎは、セクションＳｎでの末尾エリア番号である。たとえば、セクションＳ４の配分時間Ｔｓ４（ｎ＝４）を算出する場合、ＡＮ４＝２４であり、ＡＮ３＝１８である。なお、セクションＳ１の場合、ＡＮｎ−１は「０」とする。 After setting the linear function g (x), the allocation time Tsn of each section Sn (n = 1 to 4) is calculated based on the following equation (5).

Here, ANn is the end area number in section Sn. For example, when calculating the allocation time Ts4 (n = 4) of the section S4, AN4 = 24 and AN3 = 18. In the case of section S1, ANn−1 is “0”.

  The allocation time Tsn calculated by the above method increases as n increases. Thereby, the allocation time Tsn satisfies the formula (4). The calculated distribution time Tsn is stored in the memory 12.

  In the above description, g (x) is a linear function, but may be a quadratic function. Further, as shown in FIG. 17, the ratio of each distribution time Tsn to the timer Tref may be set in advance and stored in the memory 12. Hereinafter, in step S301, it is assumed that the distribution time Ts1 is set to 10 minutes, Ts2 is set to 20 minutes, Ts3 is set to 35 minutes, and Ts4 is set to 55 minutes.

  The CPU 11 sets the number of tracks to be selected for each section in order to create the playlist PL (S302 to S307). In S <b> 302, the CPU 11 sets the section counter n = 1 set in the memory 12. Subsequently, the CPU 11 reads the allocation time Ts1 of the section Sn (here, S1) from the memory 12 (S303). Based on the following equation (6), the number of tracks selected in the section Sn (hereinafter referred to as the number of selected tracks) TRNsn is determined.

Number of selected tracks TRNsn = allocation time Tsn / average playback time Tave (6)
In the above example, TRNs1 is 2 (= 10/5). Note that the value of the number of selected tracks TRNsn calculated by Expression (6) is rounded off to the nearest decimal point.

  The determined number of selected tracks TRNs1 is registered in the playlist condition table shown in FIG. 18 (S305). Referring to FIG. 18, the selected track number TRNsn is registered in association with the section ID.

  Subsequently, the CPU 11 determines whether or not the selected track number TRNs1 registered in step S305 is smaller than the area number ANs1 of the section S1 registered in the section table (FIG. 14) (S306). Since the selected track number TRNs1 (= 2) is smaller than the area number ANs1 (= 6) (YES in S306), the CPU 11 has the same number as the selected track number TRNs1 among the areas A1 to A6 allocated to the section S1. Select the area. That is, the CPU 11 selects two areas from the areas A1 to A6 (S307). For example, the CPU 11 selects areas A2 and A5. The selected areas A2 and A5 are registered in the playlist condition table. Hereinafter, an area registered in the playlist condition table is referred to as a registered area. When the CPU 11 selects a track listed in the playlist PL from the section S1, the CPU 11 selects a track from each of the registration areas A2 and A5.

  The CPU 11 determines whether or not the section counter n is equal to the set section number SN (here, 4) (S308). If the section counter n is different, the section counter n = 1 is incremented and n = 2 is set (S309). The process returns to step S303. Thereby, the CPU 11 calculates the selected track numbers TRNs1 to TRNs4 for all the sections S1 to S4 and registers them in the playlist condition table.

  Since the selected track number TRNs3 is 7 and the selected track number TRNs4 is 11, both are larger than the area numbers ANs3 (= 6) and ANs4 (= 6) (NO in S306). In this case, the CPU 11 does not execute step S307, but instead registers all the areas A13 to A18 and A19 to A24 belonging to the sections S3 and S4 (S310).

  As will be described later, when the playlist is constructed, the CPU 11 selects at least one track from each of all the areas A13 to A24 allocated to the sections S3 and S4. Since the selected track number TRNs2 is 4 (S304), the CPU 11 selects four areas (A7, A8, A10, A11 in FIG. 18) as registration areas of the section S2 and registers them in the playlist condition table. To do.

  With the above operation, the CPU 11 sets the number of tracks selected from the sections S1 to S4 by the selected track number setting process. At this time, the CPU 11 minimizes the number of selected tracks in the section S1, and increases the number of selected tracks as the section number increases. Then, the number of selected tracks is maximized in section S4. As a result, it is possible to increase the ratio of playing back tracks that are close in tone to the last track, and to suppress changes in tone as sleep playback progresses.

[Playlist creation process]
In the playlist creation process, the CPU 11 creates a playlist PL for sleep reproduction. At this time, the tracks are arranged in the playlist PL so that the tempo changes from dynamic to static and the sound quality changes from hardware to software as the playlist PL progresses from the beginning to the end. Furthermore, the tracks are arranged so that the change in the tune of the track continuously reproduced at the beginning of the play list is large and the change in the tune is reduced toward the end of the play list.

  Referring to FIG. 19, CPU 11 first sets section counter n in memory 12 to “1” (S401). Also, the playlist cumulative playback time PPT stored in the memory 12 is reset to “0” (S402).

  Subsequently, the CPU 11 refers to the playlist condition table of FIG. 18 and selects the area with the largest area number from the registered areas of the section Sn (here, S1) (S403). CPU11 selects area A5 as a registration area with the largest area number. Hereinafter, the area selected in steps S403 and S420 to S422 is referred to as a selection area.

  The CPU 11 resets the section cumulative playback time SPT stored in the memory 12 to “0” (S404), and then selects a track of the number of selected tracks (here, 2) from the section S1.

  The CPU 11 refers to the area table of FIG. 9 and determines whether or not there is a track in the selected area A5 (S405). If no track exists in the selection area A5 (NO in S405), the process proceeds to step S424. On the other hand, when there is a track in the selection area A5 (YES in S405), the CPU 11 selects one track from the selection area A5 (S406). Hereinafter, the track selected in step S406 is referred to as a selected track. Here, the description is continued assuming that the CPU 11 has selected the track TR5 from the selection area A5.

  Subsequently, the CPU 11 reads the reproduction time of the selected track TR5 from the track table, and calculates the section cumulative reproduction time SPT based on the following equation (7).

SPT = SPT + Selected track playback time (7)
Here, the section cumulative reproduction time SPT calculated by the equation (7) is 4 minutes 30 seconds. The CPU 11 determines whether or not the section cumulative reproduction time SPT is longer than the distribution time Ts1 (= 10 minutes) of the section S1 stored in the memory 12 (S408). Here, since SPT is smaller than Ts1 (NO in S408), the CPU 11 registers the selected track TR5 in the playlist preparation table shown in FIG. 20 (S415). Further, the CPU 11 registers the area to which the selected track TR5 belongs (that is, the selected area A5) in the playlist preparation table in association with the track ID.

  The CPU 11 further calculates the sleep feature amount SFQ of the selected track TR5 (S416). The CPU 11 reads the sound quality value and tempo value of the track TR5 from the track table, and obtains the sleep feature value SFQ based on the formula (A). The obtained sleep feature value SFQ (= −181) is registered in the playlist preparation table. As will be described later, the CPU 11 determines the order of reproducing the tracks TR (that is, the order listed in the playlist) based on the belonging area in the playlist preparation table or the sleep feature value SFQ.

  Subsequently, the CPU 11 changes the selected flag of the area A5 stored in the memory 12 from “0” to “1” (S424). When the selected flag is “1”, this indicates that the area has already been selected. The memory 12 stores selected flags for the areas A1 to A24. At the start of the playlist creation process, all selected flags are reset to “0”.

  Subsequently, the CPU 11 refers to the playlist preparation table and determines whether or not there is an area with the selected flag being 0 among the registration areas of the section S1 (S417). Here, the selected flag in the registration area A2 is 0 (YES in S418). Therefore, the CPU 11 selects a registration area with the largest area number among unselected areas (S421), and proceeds to step S405. Here, the CPU 11 selects area A2.

  Returning to step S405 again, the CPU 11 determines whether or not there is a track in the selection area A2 (S405), and if there is a track, selects any one track (S406). Here, the description will be continued assuming that the CPU 11 selects the track TR4 from the registration area A2.

  The CPU 11 reads the playback time (3 minutes 50 seconds) of the track TR4 from the track table, and sets the section cumulative playback time SPT to 8 minutes 20 seconds (= 4 minutes 30 seconds + 3 minutes 50 seconds) based on the equation (7). . As a result of the determination in step S408, since SPT is smaller than Ts1 (NO in S408), the CPU 11 registers the track TR4 and the registration area A2 in the playlist preparation table (S415), and sets the sleep feature amount SFQ of the track TR4. Calculate and register (S416).

  In step S417, the CPU 11 determines that the selected flags of all the registration areas A2 and A5 in the section S1 are 1 (NO in S417). At this time, the CPU 11 refers to the section table to determine whether or not an area (hereinafter referred to as an unregistered area) other than the registered areas (A2 and A5) registered in the playlist condition table exists in the section S1. Judgment is made (S418). Since there are unregistered areas A1, A3, A4 and A6 in addition to the registered areas A2 and A5 in section S1 (YES in S418), the CPU 11 selects one area from the unregistered areas ( S420), the process proceeds to step S405. Here, the description will be continued assuming that the unregistered area A3 is selected.

  When the CPU 11 selects the track TR1000 in the selection area A3 in step S406, the CPU 11 reads the reproduction time (4 minutes 30 seconds) of the track TR1000, and executes step S407. As a result, the CPU 11 determines that the section cumulative reproduction time SPT exceeds the distribution time Ts1 of the section S1 (YES in S408). Therefore, the CPU 11 proceeds to step S409 without registering the selected track TR1000 and area A3 in the playlist preparation table.

  In step S409, the CPU 11 calculates a surplus time RT based on the following equation (10) (S409).

Surplus time RT = allocation time Tsn-section cumulative playback time SPT (10)
The CPU 11 determines whether or not the calculated surplus time RT is longer than a predetermined time ΔTmin (for example, 2 minutes) (S410). If the surplus time RT is smaller than ΔTmin (NO in S410), the process proceeds to step S411. On the other hand, when the surplus time RT is larger than ΔTmin (YES in S410), the CPU 11 executes a surplus time adjustment process (S430). The surplus time adjustment process is a process of searching for tracks within the section S1 whose playback time is within the surplus time RT. Hereinafter, the surplus time adjustment process will be described.

  Referring to FIG. 21, CPU 11 searches for a track whose playback time is within surplus time RT in the currently selected area (S501). The CPU 11 refers to the area table, identifies a track belonging to the selected area (here, area A3), and refers to the track table to check the reproduction time. As a result of the search in step S501, if a track whose playback time is within the surplus time RT is found (YES in S501), the process proceeds to step S502. On the other hand, when a track whose reproduction time is within the surplus time RT is not found (NO in S501), the CPU 11 determines whether there is an unselected area in the section S1 (S507). If there is no unselected area (NO in S507), the process proceeds to S509.

  If the result of determination in step S507 is that there is an unselected area (YES in S507), the CPU 11 searches for a track whose playback time is shorter than the surplus time RT in the unselected area (S508). If a track whose playback time is shorter than the surplus time RT is found (YES in S508), the process proceeds to step S502. On the other hand, when a track whose playback time is shorter than the surplus time RT is not found (NO in S508), the CPU 11 determines that the playback time is longer than the surplus time RT in the registration areas A2 and A5 of the section S1 in the playlist condition table. The short track is searched (S509). If a track whose playback time is shorter than the surplus time RT is found, the process proceeds to step S502. If no track is found, the surplus time adjustment process is terminated.

  The CPU 11 registers the track ID of the selected track and the area ID to which the track belongs in the playlist preparation table (S502), and calculates and registers the sleep feature value SFQ of the track (S503). After registration, the CPU 11 calculates the section cumulative playback time SPT (S504), and proceeds to step S409.

  With the above operation, the CPU 11 searches for a track whose reproduction time is shorter than the surplus time RT among the tracks in the section S1. Thereby, it is possible to suppress the occurrence of excessive surplus time for the allocation time Tsn allocated to each section Sn. Further, the CPU 11 selects a track as much as possible from an area excluding the area where the track has already been selected. As a result, it is possible to suppress selection of tracks having similar tunes as much as possible.

  Returning to FIG. 19, when the surplus time RT becomes equal to or shorter than the predetermined time ΔTmin (NO in S410), the CPU 11 calculates the playlist cumulative reproduction time PPT based on the following equation (11) (S411).

PPT = PPT + SPTsn (11)
SPTsn is the section cumulative playback time of section sn. Here, the section cumulative playback time of section S1 is substituted. The calculated pre-list cumulative playback time PPT is registered in the memory 12.

  The CPU 11 increments the section counter n and sets n = 2 (S412). Then, it is determined whether or not the section counter has exceeded the section number SN (S413). Since the section counter (= 2) does not exceed the number of sections SN (= 4) (NO in S413), the CPU 11 reads the allocation time Ts2 of the section S2, and based on the following equation (12), the allocation time Ts2 Is readjusted (S414).

Allocation time Tsn = Tsn + RT (12)
In short, a value obtained by adding the surplus time RT to the distribution time Ts2 currently stored in the memory 12 is set as a new distribution time Ts2. The calculated distribution time Ts2 is stored in the memory 12. Thereby, it can suppress that excessive surplus time generate | occur | produces after playlist creation.

  After executing the above operation, the CPU 11 returns to step S403 and selects at least four tracks (the number of selected tracks in the section S2) from the section S2.

  By repeating the above operation, the CPU 11 selects a track to be registered in the playlist from the sections S1 to S4. As shown in FIG. 18, in sections S3 and S4, the number of selected tracks TRNsn exceeds the number of registered areas. Therefore, after selecting one track in all areas, the CPU 11 selects a new track again in any area. For example, when the track of section S3 is selected and one track is selected from each of the registration areas A13 to A18, the CPU 11 determines in step S418 that the area where the selected flag is 0 in section S3. It is determined that there is not (NO in S418). In this case, the CPU 11 resets the selected flag of the areas A13 to A18 stored in the memory 12 to 0 (S419), and sets the area A18 having the largest area number as the selected area (S422). Then, returning to S405, the track is selected again for the area A18.

  With the above operation, two tracks are selected from the area A18 in the section S3. By the same processing, two tracks are selected from the areas A20 to A24 in the section S4. Although not shown, the CPU 11 determines whether or not the selected track has already been registered in the playlist preparation table when the result of step S408 is NO so that the same track is not selected. If it is not registered, the process proceeds to step S415. If it is already registered, the process returns to step S406, and a track is selected again.

  The end track belongs to the area A24 in the section S4. However, the CPU 11 may or may not select the end track. Further, when a track in the area A24 is selected, it may be set in advance so that the last track is selected.

  After the selection of tracks in all sections S1 to S4 is completed (YES in S413), the CPU 11 creates the playlist PL shown in FIG. 22 based on the playlist preparation table (S423). At this time, the CPU 11 registers the track TR in the playlist PL in ascending order of the number of the belonging area registered in the playlist preparation table. That is, the tracks are registered in the playlist in the order in which the area to which the coordinate point in the xy coordinate system belongs is closer to the coordinate point Pstr of the first track STR. In other words, the section to which the coordinate point of the track belongs is registered in the playlist in the order from the closest coordinate point of the first audio file. Thus, the CPU 11 lists tracks on the playlist PL so that the tempo changes from dynamic to static and the sound quality changes from hardware to software. Furthermore, the change in the tune of a plurality of tracks listed on the playlist PL is large at the beginning of the playlist, and the change in the tune is reduced toward the end of the playlist. Therefore, in the track in the early stage of sleep reproduction, the tune changes greatly every time the next track is played, but the change in tune decreases as the play list reproduction progresses. Therefore, at the end of sleep reproduction, the music can be reproduced quietly by software, and the change in the tone of each song can be reduced.

  In step S423, the CPU 11 may register the tracks TR in the playlist PL in ascending order of the sleep feature amount SFQ. In this case, even when a plurality of tracks are selected in the same area, the tone of music can be accurately and sequentially changed from dynamic to static and from hardware to software based on the sleep feature value SFQ. it can.

  At the end of the playlist reproduction process, the CPU 11 calculates the time between songs (S425). The CPU 11 reads the playlist cumulative reproduction time PPT and the timer time Tref registered in the memory 12. The CPU 11 further specifies the number of tracks registered in the playlist PL. The CPU 11 specifies the number of tracks by the maximum value of the reproduction number on the playlist PL. The CPU 11 calculates the inter-song time based on the following equation (13) using the playlist cumulative playback time PPT, the timer Tref, and the number of tracks.

Time between songs = (Tref−PPT) / (number of tracks−1) (13)
Thereby, CPU11 can allocate the surplus time to the time between music. For this reason, when the timer time Tref elapses, the reproduction of the music can be completed, and the occurrence of excessive time can be suppressed. After executing the above operation, the CPU 11 completes the playlist creation process.

[Operation after playlist creation processing]
Returning to FIG. 3 again, when the reproduction of the first track is completed (YES in S10), the CPU 11 starts reproduction based on the playlist (YES in S11 and S12). At this time, the audio apparatus 1 can sequentially reproduce the tracks so that the tempo changes from dynamic to static and the sound quality changes from hardware to software. Furthermore, at the initial stage of sleep reproduction, as the track is reproduced, the change in the tune is large, and as the reproduction proceeds, the change in the tune is reduced. For this reason, as the end of sleep playback approaches, tracks with similar tunes are played, and a song suitable for the user's psychology of wanting to fall asleep can be played.

  Note that when the reproduction of the track is completed, the CPU 11 starts the reproduction of the next track after the time between songs calculated in step S425 has elapsed. Thereby, the reproduction of the track can be just completed at the end of the timer time, and there is no remainder in the timer time.

  As described above, in the present embodiment, the feature amount is two, that is, the sound quality value and the tempo value. However, the feature amount may be one, or may be three or more. Alternatively, only one feature amount (for example, tempo value) among a plurality of feature amounts may be selected to create a one-dimensional coordinate system, and sections and areas may be created by the above-described method. Further, sections and areas may be created by the above-described method using a multidimensional coordinate system in which coordinate axes corresponding to three or more feature quantities are set.

  In the above embodiment, sections and areas are created. However, only sections may be created, and tracks may be selected in units of sections.

  In this embodiment, the section is created after the area is created first, but a section may be created first, and then a plurality of areas may be created to divide each section.

  In the present embodiment, the audio apparatus 1 creates the playlist PL and reproduces the music based on the playlist PL, but the creation of the playlist may be performed by another apparatus instead of the reproduction apparatus. For example, in the case of a network AV system in which a server that stores a plurality of tracks, a renderer that reproduces music, and a server and a controller that controls the renderer are connected to the network, the controller is characterized by each track stored in the server. The amount may be acquired, a playlist may be created by the above-described method, and playback based on the playlist may be instructed to the renderer.

  While the embodiments of the present invention have been described above, the above-described embodiments are merely examples for carrying out the present invention. Therefore, the present invention is not limited to the above-described embodiment, and can be implemented by appropriately modifying the above-described embodiment without departing from the spirit thereof.

It is a functional block diagram which shows the structure of the audio apparatus by embodiment of this invention. It is a figure which shows the data structure of the track table memorize | stored in the hard disk shown in FIG. FIG. 6 is an operation flowchart of sleep reproduction processing executed by the audio device shown in FIG. 1. It is a flowchart which shows the detail of step S7 in FIG. It is a flowchart which shows the detail of step S701 in FIG. It is a schematic diagram for demonstrating step S101 in FIG. It is a schematic diagram for demonstrating step S105 in FIG. FIG. 10 is another schematic diagram for explaining step S105 in FIG. 5, which is different from FIG. It is a figure which shows the data structure of the area table in which data are registered by step S107 and S108 in FIG. It is a schematic diagram for demonstrating step S108 in FIG. FIG. 9 is another schematic diagram for explaining step S105 in FIG. 5, which is different from FIG. 7 and FIG. It is a flowchart which shows the detail of step S702 in FIG. It is a schematic diagram for demonstrating step S203 in FIG. It is a figure which shows the data structure of the section table in which data are registered by step S203 in FIG. It is a flowchart which shows the detail of step S703 in FIG. It is a schematic diagram for demonstrating step S301 in FIG. It is a figure which shows the data structure of the time allocation table which can be used by step S301 in FIG. FIG. 16 is a diagram illustrating a data structure of a playlist condition table in which data is registered in steps S305 and S307 in FIG. It is a flowchart which shows the detail of step S704 in FIG. It is a figure which shows the data structure of the play list preparation table in which data are registered by step S415 in FIG. It is a flowchart which shows the detail of step S430 in FIG. It is a figure which shows the data structure of the play list produced by step S423 in FIG.

Explanation of symbols

1 Audio device 11 CPU
12 Memory 13 HDD
14 Input section

Claims (8)

Storage means for storing a plurality of audio files and feature amounts of the audio files;
Arrangement means for arranging coordinate points of the audio file in a coordinate system having coordinate axes corresponding to the feature amount based on the feature amount;
A specifying means for specifying the first audio file and the last audio file;
In the coordinate system, an area creating means for dividing a vector connecting the coordinate point of the head audio file to the coordinate point of the tail audio file into a plurality of areas and creating a plurality of areas arranged in a line in the vector direction;
Among the plurality of created areas, one or a plurality of audio specified by the coordinate points included in each area so that the area closer to the coordinate point of the end audio file has a larger number of selected audio files. A selection means for selecting a file;
An audio apparatus comprising: playback means for sequentially playing the selected plurality of audio files in an order in which an area including the coordinate point of the audio file is close to the coordinate point of the head audio file. The audio device of claim 1, further comprising:
An area creating unit that divides each area into a plurality of areas and creates a plurality of areas arranged in a line in the vector direction,
The audio apparatus is characterized in that the reproduction means sequentially reproduces the selected plurality of audio files in an order in which an area including the coordinate point of the audio file is close to the coordinate point of the head audio file. The audio device according to claim 2,
The storage means further stores a playback time of each of the plurality of audio files,
The audio device further includes
Timer setting means for setting the timer time;
The allocation time of each area is set such that the total of the allocation time of each area is equal to or less than the timer time and the allocation time is longer in the area closer to the coordinate point of the end audio file among the plurality of areas. Distribution setting means for setting,
The audio device is characterized in that the selection means selects the audio file so that the total reproduction time of the audio files selected in the respective areas is within the distribution time set by the distribution setting means. . The audio device according to claim 3, further comprising:
Total playback time calculating means for calculating the total playback time of the audio file selected by the selection means;
Difference calculating means for calculating a difference value between the timer time and the total reproduction time;
Based on the difference value and the selected number of audio files, inter-song time calculating means for calculating a time between songs,
The audio device is characterized in that, after the reproduction of the audio file is completed, the reproduction means reproduces the next audio file after the time between the songs has elapsed. The audio device according to claim 1,
The audio device is characterized in that, when an audio file is being played back, the specifying means specifies the audio file being played back as the head audio file. A playlist creation device that creates the playlist for playing an audio file based on a playlist in which a plurality of audio files are listed,
Acquisition means for acquiring feature quantities of a plurality of audio files;
Arrangement means for arranging coordinate points of the audio file in a coordinate system having coordinate axes corresponding to the feature amount based on the feature amount;
A specifying means for specifying the first audio file and the last audio file;
In the coordinate system, an area creating means for dividing a vector connecting the coordinate point of the head audio file to the coordinate point of the tail audio file into a plurality of areas and creating a plurality of areas arranged in a line in the vector direction;
Among the plurality of created areas, one or a plurality of audio specified by the coordinate points included in each area so that the area closer to the coordinate point of the end audio file has a larger number of selected audio files. A selection means for selecting a file;
A playlist that lists the selected audio files so that the selected audio files are played back in the order that the area including the coordinate points of the audio file is closest to the coordinate points of the first audio file. A playlist creation device comprising playlist creation means for creating. The playlist creation device according to claim 6, further comprising:
An area creating unit that divides each section into a plurality and creates a plurality of areas arranged in the vector direction,
The playlist creating means is configured to reproduce the selected plurality of audio files so that an area including the coordinate point of the audio file is played back in an order close to the coordinate point of the first audio file. A playlist creating apparatus, wherein files are listed in the playlist. An audio playback program executable by a computer that stores a plurality of audio files and feature quantities of the audio files,
Arranging coordinate points of the audio file in a coordinate system having a coordinate axis corresponding to the feature amount based on the feature amount;
Identifying the first and last audio files;
Dividing the vector connecting the coordinate point of the leading audio file to the coordinate point of the trailing audio file in the coordinate system into a plurality, and creating a plurality of areas arranged in a line in the vector direction;
Among the plurality of created areas, one or a plurality of audio specified by the coordinate points included in each area so that the area closer to the coordinate point of the end audio file has a larger number of selected audio files. Selecting a file;
An audio playback program that causes the computer to execute the step of sequentially playing the selected plurality of audio files in an order in which an area including the coordinate point of the audio file is close to the coordinate point of the head audio file. .

Images