JP5217687B2 - Song editing support apparatus and program - Google Patents

Description

  The present invention relates to a technology that supports editing of a musical composition in which a plurality of musical material is combined.

As a technique related to song editing, there is a technique called audio mosaicing. In this audio mosaicing technology, a relatively short music material of about several bars is collected to construct a database. Then, a desired music material is selected from the database, and the sound wave shapes of the selected music material are connected so as to overlap on the time axis, thereby editing a new music. Note that there are, for example, Patent Document 1 and Non-Patent Documents 1, 2, and 3 as documents relating to this type of technology.
International Publication No. WO2007 / 044389 Pamphlet UPF, Music Surfer Similarity Map, [Search May 13, 2008], Internet <URL: http://musicsurfer.iua.upf.edu/help/html/Music_Surfer-Similarity_Maps.html> Andreas Rauber, PlaySOM, [Search May 13, 2008], Internet <http://www.ifs.tuwien.ac.at/mir/playsom.html> Steinberg, MediaBay [Search May 13, 2008], Internet <http://www.steinberg.net/1419+M52087573ab0.html>

By the way, in the database in the audio mosaicing technology, “3 beats” that constitutes one measure with strong beat → weak beat → weak beat, and “4” that constitutes one measure with strong beat → weak beat → medium weak beat → weak beat. Musical materials of various time signatures such as “Time signature” are collected. Then, the editor who edits the music selects and plays multiple music materials with the same time signature that seem to be compatible with each other, and whether the harmony produced by those playback sounds matches his or her own musical idea. Repeat the process of examining how.
However, in the case of the conventional audio mosaicing technology, it is impossible to synchronize the beats of the playback sounds of a plurality of music materials selected from the database after the playback is started. Therefore, when the user performs the operation of instructing the playback of the music material continuously, the playback sound of what was instructed to play first does not match the strength of the beat of the playback sound of the instruction to play back later. There was a problem that editing did not work.
The present invention has been devised under such a background, and an object of the present invention is to provide a music editing support device and a program that make it easy to edit music by connecting music materials and editing music. To do.

The present invention provides a material selection operation for selecting music material, a detection means for detecting synchronization instruction operation for instructing synchronization of beats of music material selected by the material selection operation, and the detection means detects the material selection operation. Each time a separate reproduction processing task for reproducing the music material selected by the material selection operation is started, and the started reproduction processing task is executed in parallel. When a plurality of playback processing tasks are executed at the time when an operation is detected, execution modes of the plurality of playback processing tasks executed at that time so that the beats of the playback sounds of those playback processing tasks are synchronized A music editing support device is provided that includes a reproduction processing control means for controlling the music.
According to the present invention, when the synchronization instruction operation is performed during execution of a plurality of reproduction processing tasks, the reproduction processing control means is configured so that the beats of the reproduction sounds in the plurality of reproduction processing tasks are synchronized. The execution mode of a plurality of reproduction processing tasks being executed at the time is controlled. Therefore, it is possible to avoid the inconvenience that the user does not have the right and wrong beats of the playback sound of the playback sound that was instructed to play first and the playback sound that was later instructed, and the editing of the music is avoided, making the music editing work easier can do.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a music editing support apparatus according to an embodiment of the present invention. This music editing support device is composed of two beats that make up one measure with strong beats → weak beats, “three beats” that make up one measure with strong beats → weak beats, and strong beats → weak beats → medium. Various patterns pp-k (k = 1, 2 ... M: written as “phrase pattern pp-k”) of phrases with various time signatures such as “4 time signatures” that compose one measure with a weak beat → weak beat The file 21-k (k = 1, 2,... M) is stored in the hard disk 20. Then, the song editing support apparatus displays a screen (referred to as an “editing support screen”) that supports editing of the music that arbitrarily connects the phrase patterns pp-k on the display unit 16, and is selected through the editing support screen. The reproduced phrase pattern pp-k is reproduced. While listening to the reproduced sound of the phrase pattern pp-k by the music editing support device, the user can connect a plurality of phrase patterns pp-k so as to overlap each other on the time axis, or along the time axis. The desired music is edited by connecting them continuously.

  In FIG. 1, a clock 11 generates a clock signal having a frequency f. The CPU 12 is a control center that controls each unit of the music editing support apparatus. The ROM 13 is a read-only memory that stores an IPL (Initial Program Loader) and the like. The RAM 14 provides a work area to the CPU 12. The beat signal generator 15 periodically generates a signal (referred to as a “beat signal”) that determines the beat interval when the music material file 21-k is reproduced under the instruction of the CPU 12. The display unit 16 is a device for displaying various types of information, and includes a display and a drive circuit thereof.

The mouse 17 accepts operations for instructing movement, left click, and right click of the mouse cursor mc within the display area of the display unit 16 and generates a signal corresponding to the operation. A signal generated by the mouse 17 is captured by the CPU 12.
The interface group 18 includes a network interface for performing data communication with other devices via a network, a driver for transmitting and receiving data to and from an external storage medium such as a magnetic disk and a CD-ROM, and the like. It is comprised by.
The sound system 19 includes a D / A converter that converts a digital audio signal, which is a sound wave sample sequence, into an analog audio signal, an amplifier that amplifies the analog audio signal, a speaker that outputs the output signal of the amplifier as sound, and the like. It is configured.

The hard disk 20 stores a music material file 21-k (k = 1, 2,... M), a feature data file 22, a descriptor file 23, a BPM (Beat Per Minute) data file 24, and a control program 25.
The music material file 21-k (k = 1, 2,... M) is a file having sound waveform data. The sound waveform data is data indicating the sound waveform of the phrase pattern pp-k (k = 1, 2,... M) for a certain number of bars (for example, for four bars). This sound waveform data is acquired as follows. First, each phrase pattern pp-k (k = 1, 2,... M) is played at a desired tempo (beats / minute), and the performance sound is recorded. Next, the sound waveform recording the performance sound of each phrase pattern pp-k (k = 1, 2,... M) is sampled at the sampling rate of the same frequency f as that of the clock 11 to generate a sample string. Further, a bit (referred to as “beat point bit”) indicating that it is a beat point is added to a sample at a position corresponding to the beat point in the generated sample sequence (appropriately referred to as “beat point sample”).

The feature data file 22 is a file having feature data C (k, i) (k = 1, 2... M, i = 1, 2... N). Each of these feature data C (k, i) (k = 1, 2,... M, i = 1, 2,... N) is, for example, a phrase pattern pp-k (k) such as timbre, harmony, time signature, etc. = 1, 2,... M). For example, a group of N feature data C (1, i) (i = 1, 2,... N) indicating features of a certain phrase pattern pp-1 is a series of samples that form sound waveform data of the phrase pattern pp-1. Are obtained according to N analysis algorithms for analyzing individual features.
The descriptor file 23 is a file having descriptor data. The descriptor data includes the types of features indicated by the feature data C (k, i) (k = 1, 2... M, i = 1, 2... N) (“feature 1”, “feature 2”... “Feature N”). It is data which shows.

The BPM data file 24 is a file having BPM data Bk (k = 1, 2,... M). The BPM data Bk (k = 1, 2,... M) is data indicating the number of beats per minute that is the rhythm of the phrase pattern pp-k (k = 1, 2,... M). These BPM data Bk (k = 1, 2,... M) are obtained from the sound waveform of the phrase pattern pp-k (k = 1, 2,... M) from the music material file 21-k (k = 1, 2,... M). ) Was obtained at the same time.
The control program 25 executes processing for displaying the editing support screen, switching the display contents of the editing support screen, and reproducing one or more phrase patterns pp-k selected through the editing support screen, on the CPU 12. It is a program to let you.

FIG. 2 is a diagram showing an editing support screen. As shown in FIG. 2, the editing support screen includes a first GUI (Graphical User Interface) area 31 and a second GUI area 32. The first GUI area 31 includes a feature input field 33-j (j = 1, 2), a track bar 35, and a button 36.
The feature input field 33-j (j = 1, 2) is the type of N features (“features” in the feature data C (k, i) (k = 1, 2... M, i = 1, 2... N). 1 ”,“ feature 2 ”...“ Feature N ”) for selecting an arbitrary one. On the right side of the feature input column 33-1, the character “X-axis” is written, and on the right side of the feature input column 33-2, the character “Y-axis” is written.
The track bar 35 is a GUI for designating the playback tempo of the phrase pattern pp-k. A knob 38 is disposed on the axis scale 37 of the track bar 35. The letters “Fastest (120 beats / min)” are on the right of the upper end of the axis scale 37, and the characters “Slowest (60 beats / min)” are on the right of the lower end of the axis scale 37. The letters “Medium (90 beats / min)” are written at the center of the distance.
The button 36 is a GUI for instructing display of the feature distribution screen in the second GUI area 32. The button 36 is marked with “Redraw”. The feature distribution screen includes a coordinate value x (0 ≦ x ≦ 255) indicating a position on the x coordinate axis from the left end to the right end of the second GUI area 32, and a coordinate value y indicating a position on the y coordinate axis from the upper end to the lower end. A diameter corresponding to the tempo (beats / minute) of the phrase pattern pp-k (k = 1, 2... M) at each position in the second GUI area 32 positioned by a pair with (0 ≦ y ≦ 255). This is a screen on which circular marks cf-k (k = 1, 2,... M) are arranged. Details of this feature distribution screen will be described later.

When the screen shown in FIG. 2 is displayed on the display unit 16, the user can perform a feature selection operation, a track bar adjustment operation, and a button selection operation.
In the feature selection operation, the user moves the mouse cursor mc to the desired feature input field 33-j and left-clicks it. When the CPU 12 detects that the position of the mouse cursor mc when left-clicked is on the feature input field 33-j, the type of feature indicated by the descriptor file 23 stored in the hard disk 20 ("feature 1") , “Feature 2”,... “Feature N”) are displayed side by side. The user moves the mouse cursor mc to one of these features and left-clicks again. When the CPU 12 detects that the position of the mouse cursor mc when left-clicked is on the feature type, the CPU 12 determines the type as the selected type and sets the type name to the feature input field 33-j. To display inside.

  In the track bar adjustment operation, the user moves the mouse cursor mc to the knob 38 of the track bar 35, left-clicks, moves the mouse cursor mc up or down while left-clicked, and moves the mouse cursor mc to the axis. The left click is canceled when the position is moved to a desired position on the scale 37. When the CPU 12 detects that the mouse cursor mc is left-clicked when the mouse cursor mc is on the knob 38 and the mouse cursor mc has moved without releasing the left click, the CPU 12 moves the position of the knob 38 following the mouse cursor mc. .

  In the button selection operation, the user moves the mouse cursor mc to the button 36 and left-clicks. When the CPU 12 detects that the position of the mouse cursor mc when it is left-clicked is on the button 36, it executes a feature distribution display process. The feature distribution display process is a process for displaying the feature distribution screen in the second GUI area 32. This process is executed by the CPU 12 by the operation of the control program 25.

FIG. 3 is a flowchart showing the feature distribution display process.
In FIG. 3, when the CPU 12 detects that the position of the mouse cursor mc when left-clicked is on the button 36, the CPU 12 reads the feature data file 22 and the BPM data file 24 stored in the hard disk 20 to the RAM 14. Later (S100), a file selection dialog screen is displayed in the second GUI area 32 (S110). The file selection dialog screen is a screen in which the file names of the music material files 21-k (k = 1, 2,... M) of the respective phrase patterns pp-k (k = 1, 2,... M) are arranged.

When this file selection dialog screen is displayed, the user can perform a reference material selection operation.
In the reference material selection operation, the user moves the mouse cursor mc to the file name of the music material file 21-k having the desired phrase pattern pp-k on the file selection dialog screen, and left-clicks.
When the CPU 12 detects that the position of the mouse cursor mc when left-clicked is on the file name (S120: Yes), the CPU 12 starts generating the beat signal by the beat signal generator 15 (S130). More specifically, the CPU 12 specifies the tempo (beats / minute) indicated by the position of the knob 38 on the axis scale 37, converts the tempo to the generation time interval (msec / 1 beat) of the beat signal, and generates the tempo. An instruction signal indicating the time interval is output to the beat signal generator 15. When acquiring the instruction signal, the beat signal generator 15 measures the generation time interval indicated by the instruction signal based on the count number of the clock signal generated by the clock 11, and generates a beat signal every time the generation time interval passes. When step S120 in FIG. 3 is “No”, the CPU 12 returns to step S120 without proceeding to step S130.

Next, the CPU 12 determines whether or not a feature type is input in the feature input field 33-1 (S140). When the feature type is input in the feature input field 33-1 (S140: Yes), the CPU 12 sets the feature data C (k, i) (k = 1, 2... M, i = 1, 2. N) of the types input to the feature input field 33-1 by the feature selection operation, the coordinate value x (0 ≦ x ≦ 255) indicating the position on the x coordinate axis from the left end to the right end of the second GUI area 32 (S150).

In this step S150, the CPU 12 firstly stores the feature data C (k, 1) of the type (for example, feature 1) input from the feature data file 22 of the RAM 14 into the feature input field 33-1 by the feature selection operation. A group (k = 1, 2,... M) is extracted. Next, a phrase pattern pp-k (for example, a phrase pattern pp-1) selected by the reference material selection operation from the group of feature data C (k, 1) (k = 1, 2,... M). ) Feature data C (1, 1) is selected, and the feature data C (1, 1) is converted to “0”, which is the coordinate value x at the left end of the x coordinate axis of the second GUI area 32, and the feature data C (k 1) Each of (k = 2... M) is converted into any one of “1” to “255” which is the coordinate value x to the right of it. Specifically, each of the feature data C (k, 1) (k = 1, 2,... M) including the phrase pattern pp-1 is expressed as a similarity between the feature data C (1, 1). The feature similarity data C-similarity shown is converted, and the feature similarity data C-similarity is converted into a coordinate value x. The conversion from the feature data C (k, 1) (k = 1, 2,... M) to the feature similarity data C-similarity is based on the Euclidean distance between each of them and the feature data C (1, 1). It is good to do. Further, the transformation from the feature similarity data C-similarity to the coordinate value x involves each of the feature similarity data C-similarity and the maximum value C _MAX and the minimum value C _MIN of the feature similarity data C-similarity. Is preferably entered by entering the following equation.
x
= | 255 × {(((C-similarity) -C _MIN ) / (C _MAX -C _MIN )}-255 | (1)
In step S140, when the feature type is not input in the feature input field 33-1 (S140: No), the CPU 12 proceeds to the next step without executing step S150.

Next, the CPU 12 determines whether or not a feature type is input in the feature input field 33-2 (S160). When the feature type is input in the feature input field 33-2 (S160: Yes), the CPU 12 sets the feature data C (k, i) (k = 1, 2... M, i = 1, 2. N) of the type input to the feature input field 33-2 by the feature selection operation, the coordinate value y (0 ≦ y ≦ 255) indicating the position on the y coordinate axis from the upper end to the lower end of the second GUI area 32 (S170).
In this step S170, the CPU 12 firstly stores the feature data C (k, 2) of the type (for example, feature 2) input from the feature data file 22 of the RAM 14 into the feature input field 33-2 by the feature selection operation. A group (k = 1, 2,... M) is extracted. Next, a phrase pattern pp-k (for example, a phrase pattern pp-1) selected by a reference material selection operation from among the feature data C (k, 2) (k = 1, 2,... M) group. ) Feature data C (1,2) is selected, and the feature data C (1,2) is converted to “0”, which is the coordinate value y of the upper end of the y coordinate axis of the second GUI area 32, and the feature data C (k 2) Each of (k = 2... M) is converted into any one of “1” to “255”, which is a coordinate value y below it. Specifically, each of the feature data C (k, 2) (k = 1, 2... M) including the phrase pattern pp-1 is expressed as a similarity between the feature data C (1, 2). The feature similarity data C-similarity shown is converted, and the feature similarity data C-similarity is converted into a coordinate value y. The conversion from the feature data C (k, 2) (k = 1, 2,... M) to the feature similarity data C-similarity is based on the Euclidean distance between each of them and the feature data C (1, 2). It is good to do. Further, the conversion from the feature similarity data C-similarity to the coordinate value y includes each of the feature similarity data C-similarity and the maximum value C _MAX and the minimum value C _MIN of the feature similarity data C-similarity. Is preferably entered by entering the following equation.
y
= | 255 × {(((C-similarity) -C _MIN ) / (C _MAX -C _MIN )}-255 | (2)
In step S160, when the feature type is not input in the feature input field 33-2 (S160: No), the CPU 12 proceeds to the next step without executing step S170.

  The CPU 12 displays a feature distribution screen in the second GUI area 32 (S180). In this step S180, the CPU 12 sets the BPM data B-k (k = 1, 2,... M) in the BPM data file 24 of the RAM 14 in a diameter corresponding to the speed of the tempo (beats / minute) indicated by each. Each is converted into a size value s indicating the size. Next, the CPU 12 positions the mark cf-k (k = 1, 2,... M) having the diameter size indicated by the size value s based on the x coordinate value obtained in step S150 and the y coordinate value obtained in step S170. A screen arranged at a position in the second GUI area 32 is drawn, and the screen is displayed on the second GUI area 32 as a feature distribution screen.

  FIG. 4 is a diagram showing a feature distribution screen displayed when a feature type (feature 1, feature 2) is input to the feature input fields 33-1 and 33-2. The mark cf-k at the upper left corner in this feature distribution screen corresponds to the phrase pattern pp-k selected by the reference material selection operation, and the remaining marks cf-k are those other than those selected by the reference material selection operation. It corresponds to the phrase pattern pp-k. Further, the diameter cf-k of the phrase pattern pp-k having a fast tempo (beats / minute) increases.

  When the feature distribution screen is displayed in the second GUI area 32, the user can perform a material selection operation, a synchronization instruction operation, and a stop instruction operation. The material selection operation is an operation of selecting a desired mark cf-k on the feature distribution screen and reproducing the phrase pattern pp-k corresponding to the mark cf-k. The synchronization instruction operation is an operation of synchronizing the beats of the phrase patterns pp-k when a plurality of phrase patterns pp-k are being reproduced. The stop instruction operation is an operation for stopping the reproduction of the phrase pattern pp-k being reproduced.

  As shown in FIG. 5, in the material selection operation, the user moves the mouse cursor mc to a desired mark cf-k and left-clicks. The user moves the mouse cursor mc to a certain mark cf-k and left-clicks, and then moves the mouse cursor mc to another mark cf-k and left-clicks again to continue the material selection operation. Can be done. When the CPU 12 detects that the position of the mouse cursor mc when it is left-clicked is on the mark cf-k, the CPU 12 displays a circle surrounding the mark cf-k, and starts and starts a playback processing task. Execute the replay processing task. In addition, when the CPU 12 detects that the position of the mouse cursor mc when the next left-click is performed after starting a certain playback processing task, the CPU 12 changes another playback processing task. After the startup, the plurality of playback processing tasks are executed in parallel. The reproduction processing task is a task for executing loop reproduction processing, and is executed by the CPU 12 by the operation of the control program 25.

In the loop playback process in one playback process task, the CPU 12 reads the corresponding music material file 21-k into an area secured in the RAM (referred to as “file storage area”). Then, every time the beat signal generator 15 generates a beat signal, the CPU 12 reads the sample while shifting the sample reading position to the beat point sample in the sound waveform data of the music material file 21-k. Each of the samples after the beat point sample at the position is output to the sound system 19 one by one in synchronization with the clock signal, and when the beat signal is generated after the readout position is shifted to the last beat point sample, the position is first set. The process of returning to the above is repeated according to the following two rules.
a. If the next beat signal is generated before the end of the output of the sample from one beat point sample to the previous one, the sample after the next beat point sample is output without outputting the unoutput sample. Move on to sample output.
b. If a clock signal is generated between the time when a sample from a certain beat point sample is output before the next beat point sample and the time when the next beat signal is generated, the value of “0” is synchronized with the clock signal. Output a sample.

Here, as described above, the sound waveform data is obtained from the sound waveform of the recorded sound by playing each phrase pattern pp-k (k = 1, 2,... M) at a desired tempo.
Therefore, when the track bar 35 designates the tempo at the same speed as when recording the phrase pattern pp-k to be looped, each beat in the sound waveform data is designated as shown in FIG. Since the time required to output samples from a point sample to each previous beat point sample (for example, “r” is the number of samples) is equal to the time interval at which the beat signal is generated, the above rule is applied. a and b are not applicable.

  On the other hand, when a tempo faster than that recorded at the time of recording of the phrase pattern pp-k to be loop-played is designated by the track bar 35, as shown in FIG. The time interval for generating beat signals is shorter than the time required for r clock signals required for sample output. In this case, the above-mentioned rule a is applied. That is, when a beat signal is generated while outputting r samples starting from a certain beat point sample, the CPU 12 outputs the next beat point sample without outputting the unoutputted sample at that time. Move to the output of r samples at the head. As a result, it is possible to obtain an audible feeling as if the phrase pattern pp-k is skip-played for each beat point.

  When the track bar 35 designates a tempo that is slower than that at the time of recording the phrase pattern pp-k for loop playback, as shown in FIG. 6C, r samples of the sound waveform data are recorded. The time interval for generating beat signals is longer than the time required for r clocks for output. In this case, the above-mentioned rule b is applied. That is, the CPU 12 outputs a sample of “0” from the end of outputting r samples starting from a certain beat point sample until the next beat signal is generated. As a result, it is possible to obtain an audible feeling as if the phrase pattern pp-k is being reproduced while being divided every beat.

By continuously performing the material selection operation, the user performs a synchronization instruction operation in a state where a plurality of phrase patterns pp-k having the same beat including those selected by the reference material selection operation are reproduced in a loop. For the phrase pattern pp-k having the same time signature, for example, “feature” is input to one of the feature input fields 33-1 and 33-2, the feature distribution screen is displayed, and the mark cf-k ( It is preferable to search with reference to the arrangement of k = 1, 2,.
As shown in FIG. 7, in the synchronization instruction operation, the user moves the mouse cursor mc to a position without the mark cf-k in the feature distribution screen and left-clicks.
When the CPU 12 detects that the position of the mouse cursor mc when left-clicked is a position without the mark cf-k in the feature distribution screen, the CPU 12 waits for generation of the next beat signal by the beat signal generator 15. . When the next beat signal is generated, the read position of the sample in each music material file 21-k in each file storage area of the RAM 14 is shifted to the first beat point sample, and after the first beat point sample. Start outputting each sample again. Therefore, for example, as shown in FIG. 8, when three phrase patterns pp-1, pp-2, and pp-3 of four beats are reproduced with a time difference of one beat per measure, the synchronization instruction operation is performed. Once this is done, all the timings of strong beat → weak beat → medium weak beat → weak beat in each measure of the three phrase patterns pp-1, pp-2, pp-3 are reproduced.

The user performs a synchronization instruction operation and listens to the reproduced sound output from the sound system 19 thereafter. Then, when the music that is the reproduced sound does not match the own musical idea, after performing the stop instruction operation, the material selection operation and the synchronization instruction operation described above are performed again.
As shown in FIG. 9, in the stop instruction operation, the user moves the mouse cursor mc to the mark cf-k surrounded by a circle, that is, the mark cf-k instructed to reproduce by the material selection operation, and left again. click.
When the CPU 12 detects that the position of the mouse cursor mc when left-clicked is on the mark cf-k of the phrase pattern pp-k being reproduced, the phrase pattern pp corresponding to the mark cf-k -K Stops playback.

In the present embodiment described above, the button selection operation for selecting the button 36 after the user performs a type input operation for inputting a desired feature type in the feature input field 33-j (j = 1, 2). When the reference material selection operation is performed on the subsequent file selection dialog screen, the CPU 12 displays a feature distribution screen. This feature distribution screen includes a mark cf-k in the upper left corner indicating the characteristics of the phrase pattern pp-k selected by the reference material selection operation and a plurality of marks cf-k indicating the characteristics of other phrase patterns pp-k. It is arranged. Then, the CPU 12 sets the positions of the marks cf-k other than those in the upper left corner on the feature distribution screen between the phrase patterns pp-k selected by the reference material selection operation and the other phrase patterns pp-k. The feature similarity data C-similarity indicating the similarity is determined by the converted coordinate values x and y. Therefore, the user can use the selection of the reference material selection operation on the basis of the distance between the mark cf-k in the upper left corner of the feature distribution screen and the other mark cf-k. The phrase pattern pp-k can be searched.
Further, in this embodiment, whenever a user performs a material selection operation for instructing reproduction of the phrase pattern pp-k corresponding to the mark cf-k on the feature distribution screen, the phrase pattern instructed to be reproduced by the operation. The pp-k loop playback is started. Furthermore, when the user performs a synchronization instruction operation when a plurality of phrase patterns pp-k are being played back in a loop, the loop playback is executed again in response to the generation of the next beat signal. Therefore, when the user selects and reproduces a plurality of phrase patterns pp-k, the timings of strong beats, weak beats, medium weak beats, etc. that form the rhythm of those phrase patterns pp-k are aligned. Therefore, the inconvenience of not being able to grasp the tune of the music connected so as to overlap them is avoided.

Although one embodiment of the present invention has been described above, the present invention may have other embodiments. For example, it is as follows.
(1) In the loop reproduction process of the above embodiment, the CPU 12 generates a beat signal from the time required to output r samples from each beat point sample in the sound waveform data to before each next beat point sample. If the time interval is longer or shorter, the r samples are subjected to known resampling processing or time stretching processing to obtain a number of sample sequences corresponding to the time interval of occurrence of the beat signal. The columns may be output one by one in synchronization with the clock signal. In the resampling process, the CPU 12 obtains a sample having a sound waveform obtained by expanding and contracting the sound waveform indicated by the original r samples to the same length as the time interval of generation of the beat signal by interpolation. On the other hand, in the time stretch process, the CPU 12 refers to the primary differential value of the original r samples, etc., and identifies a section with a stable amplitude in the sound waveform indicated by the r samples. If the time interval of generation of the beat signal is longer, samples for one cycle are cut out from the section, and duplicates of the cut out samples are inserted as necessary. Also, if the time interval for generating the beat signal is shorter, samples for the necessary period are deleted from that interval.
(2) In the above embodiment, beat point bits are added to the beat point samples in the sound waveform data of the music material file 21-k (k = 1, 2,... M). However, the beat data indicating how many beats the sound waveform data of the file may be described in the header of the music material file 21-k (k = 1, 2,... M). In the loop reproduction process in this case, the CPU 12 obtains the number of samples corresponding to one beat by dividing the number of samples from the beginning to the end of the sound waveform data by the number of beats indicated by the beat data, and the sound waveform data It is preferable to treat each sample separated by that number as a beat point sample. In addition, in the header of the music material file 21-k (k = 1, 2,... M), tempo data indicating the tempo (beats / minute) when the sound of the sound waveform data of the file is recorded is described. Also good. In the loop reproduction process in this case, the CPU 12 cuts out a sample string corresponding to the first one minute from the sound waveform data, and divides the number of samples in the cut out sample string by the tempo (beats / minute) indicated by the tempo data. Thus, the number of samples corresponding to one beat may be obtained, and each sample separated by that number in the sound waveform data may be handled as a beat point sample. Also, without describing the beat number data or tempo data in the header, the sample sequence forming the sound waveform data is divided into a predetermined number, and the sample located at the divided boundary is handled as a beat point sample. Good.
(3) In the above embodiment, the CPU 12 uses the feature similarity data C-similarity as the cosine angle between the feature data C (k, i) (k = 1, 2... M, i = 1, 2... N). You may obtain | require based on the value which shows other similar indices.

(4) In the above embodiment, instead of the mouse 17, another pointing device such as a trackball is provided, and the CPU 12 detects that various operations have been performed based on a signal output by the pointing. Good.
(5) The layout of the first GUI area 31 on the editing support screen in the above embodiment may be arbitrarily changed. For example, a software keyboard may be displayed instead of the pull-down menu, and the feature type may be input to the feature input field 33-j (j = 1, 2) using this software keyboard. Also, a button labeled “Synchronize” is arranged in the first GUI area 31, and when the CPU 12 detects that the position of the mouse cursor mc on the button when left-clicked, the loop playback process is performed. You may make it perform again.
(6) In the above embodiment, when the CPU 12 detects that the synchronization instruction operation has been performed, the CPU 12 shifts the sample reading position in each of the music material files 21-k to the first beat point sample. However, when the CPU 12 detects that the synchronization instruction operation has been performed, the CPU 12 may shift the reading position to the first beat point sample of the next measure in each of the music material files 21-k. This embodiment can be realized by adding a bit indicating the first beat of a measure only to the “strong beat” of the beat point samples in the sound waveform data of the music material file 21-k. . In short, when a plurality of playback processing tasks are executed when a synchronization instruction operation is detected, a plurality of playback processes executed at that time so that the beats of the playback sounds of those playback processing tasks are synchronized. It is only necessary to control the execution mode of the task.

(7) In the above embodiment, the CUP 12 has drawn a feature distribution screen in which the mark cf-k of the phrase pattern pp-k selected by the reference material selection operation is arranged at the upper left corner in the second GUI area 32. However, a feature distribution screen in which the mark cf-k of the phrase pattern pp-k selected by the reference material selection operation is arranged at the center in the second GUI area 32 may be drawn. In this case, the CPU 12 is similar to the feature data C (k, i) (k = 1, 2... M, i = 1, 2... N) of the phrase pattern pp-k selected by the reference material selection operation. It is preferable to obtain positive or negative feature similarity data C-similarity indicating the position of the remaining mark cf-k based on the feature similarity data C-similarity. The CUP 12 has three types of feature data C (k, i) (k = 1, 2... M, i = 1, 2... N) for each phrase pattern pp-k (k = 1, 2... M). A feature distribution screen may be drawn in which marks cf-k (k = 1, 2,... M) are respectively arranged in a three-dimensional coordinate system.
(8) In the above embodiment, each phrase pattern pp-k is played at a certain standard tempo (for example, 90 beats / minute), and each performance sound is recorded, and the music material file 21- of the same tempo is recorded. k (k = 1, 2,... M) may be stored in the hard disk 20.
(9) In the above embodiment, the CPU 12 determines the diameter sizes of the marks cf-k (k = 1, 2,... M) of each phrase pattern pp-k (k = 1, 2,... M) on the feature distribution screen. The size was set according to the tempo (beats / minute). However, the diameter size of the mark cf-k (k = 1, 2,... M) of each phrase pattern pp-k (k = 1, 2,... M) is set to their time signature (2 time, 3 time, 4 time, etc.). You may make it the magnitude | size according to.

(10) In the above embodiment, the CPU 12 sets the phrase pattern pp-k (k = 1, 2,... M) so that the colors of the marks cf-k indicating the same tempo phrase pattern pp-k are the same. The display color of the mark cf-k (k = 1, 2,... M) may be changed according to the tempo of the phrase pattern pp-k (k = 1, 2,... M).
(11) The control program 25 in the above embodiment may be downloaded from a server device on the WWW (World Wide Web) to a personal computer, a PDA (Personal Data Assistance), a mobile phone terminal, or the like. Further, such a program may be distributed after being stored in a storage medium.

It is a block diagram which shows the structure of the music edit assistance apparatus which is embodiment of this invention. It is a figure which shows the edit assistance screen displayed on the display part of the music edit assistance apparatus of FIG. It is a flowchart which shows the feature distribution display process which the music edit assistance apparatus of FIG. 1 performs. It is a figure which shows the feature distribution screen displayed on the display part of the music edit assistance apparatus of FIG. It is a figure which shows material selection operation. It is a figure which shows the mode of the loop reproduction | regeneration processing of the sound waveform data of a 4-beat phrase pattern. It is a figure which shows synchronous instruction | indication operation. It is a figure which shows the process which a music edit assistance apparatus performs when synchronous instruction | indication operation is performed. It is a figure which shows stop instruction | indication operation.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 ... Clock, 12 ... CPU, 13 ... ROM, 14 ... RAM, 15 ... Beat signal generation part, 16 ... Display part, 17 ... Mouse, 18 ... Interface group, 19 ... Sound system, 20 ... Hard disk, 21 ... Music material File 22 characteristic data file 23 descriptor file 24 BPM data file 25 control program

Claims (6)

Detection means for detecting a material selection operation for selecting music material, and a synchronization instruction operation for instructing synchronization of beats of the music material selected by the material selection operation;
Each time the detection means detects the material selection operation, a means for starting an individual reproduction processing task for reproducing the music material selected by the material selection operation and executing the started reproduction processing task in parallel. In the case where a plurality of playback processing tasks are executed at the time when the detection means detects the synchronization instruction operation, the playback processing tasks are started from the first beat point of each music material or the next measure. A music editing support apparatus, comprising: a playback processing control unit that controls to start output of playback sound . Display means;
Storage means for storing a plurality of types of feature data of each music material;
Of the feature data stored in the storage unit, the first type of feature data is converted into an x coordinate value indicating a position on the x coordinate axis in the display area of the display unit, and the feature data stored in the storage unit Of the data, the second type of feature data is converted into a y-coordinate value indicating the position on the y-coordinate axis in the display area of the display means, and the x-coordinate value and the y-coordinate value obtained by converting the feature data of each music material Display control means for causing the display means to display a screen in which a mark indicating a feature of each music material is arranged at a position positioned by a pair with
The detection means includes
The music editing support apparatus according to claim 1, wherein an operation for selecting a mark on the screen is detected as the material selection operation. The detection means includes
The music editing support apparatus according to claim 2, wherein a feature selection operation for selecting the first type and the second type is detected. The detection means includes
Detecting a reference material selection operation for selecting one of the music materials as a reference music material;
The display control means includes
The feature data of the first type of each of the music material is converted into feature similarity data indicating the similarity of the feature with the reference music material, and the feature similarity data is displayed on the display means. A feature similarity that converts the second type feature data of each of the music materials to the reference music material and the similarity of the features while converting the x-coordinate value indicating the position on the x-coordinate axis in the region 4. The music editing support according to claim 2, wherein the feature similarity data is converted into a y coordinate value indicating a position on a y coordinate axis in a display area of the display means. apparatus. Beat signal generating means for periodically generating a beat signal,
The storage means
A sound wave sample sequence of music material, storing sound wave data which is data obtained by adding a beat point bit to a sample at a position corresponding to the beat point of the sound wave,
The reproduction control means includes
Each time the beat signal generating means generates a beat signal, each sample after the sample that has become the read position is shifted to the back sample to which the beat point bit is added in the sound waveform data. Is output one by one, and when the beat signal is generated after the read position is shifted to the last sample with the beat point bit added, the read position is returned to the first sample with the beat point bit added. When a task is executed as the playback processing task and the playback unit is executing a plurality of playback processing tasks at the time when the detection unit detects the synchronization instruction operation, the plurality of sound waves that are targets of the playback processing tasks Shifting each readout position to the first sample with added beat bit in the shape data 5. The music editing support apparatus according to any one of 4 above. On the computer,
Detection means for detecting a material selection operation for selecting music material, and a synchronization instruction operation for instructing synchronization of beats of the music material selected by the material selection operation;
Each time the detection means detects the material selection operation, a means for starting an individual reproduction processing task for reproducing the music material selected by the material selection operation and executing the started reproduction processing task in parallel. In the case where a plurality of playback processing tasks are executed at the time when the detection means detects the synchronization instruction operation, the playback processing tasks are started from the first beat point of each music material or the next measure. A reproduction processing control means for controlling to start output of reproduced sound .

Images