JP2001147690A - Performance data generation method, automatic performance device, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate performance data for every desired note group in a method which generates fluctuating performance data to be used for automatic performance or the like. SOLUTION: When a tremolo symbol A is designated, the classification and the number of notes for tremolo performance (8 sixteenth notes) are obtained. Live performance data [reference performance data] B corresponding to the obtained note classification are prepared. Performance data C in a section corresponding to the obtained number of notes are extracted from performance data B. Performance timing information in performance data C is so corrected that the length of the section coincides or approximates the sum total length of notes, and pitch information in performance data C is corrected in accordance with the pitch of notes. As the result, fluctuating performance data D for automatic performance are obtained. It is unnecessary to change reference performance data even if the number of notes, the pitch, or the like are changed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for creating performance data having fluctuation characteristics (glue and fluctuation) suitable for use in automatic performance and the like, an automatic performance device using the created performance data, and an automatic performance device. The present invention relates to a recording medium on which a performance program is recorded, and in particular, to extract performance data of a desired note group from reference performance data based on live performance and the like, and modify the extracted performance data in accordance with the note group to play the desired note group. This makes data creation possible.

[0002]

2. Description of the Related Art Conventionally, there has been known a method of creating performance data for an automatic performance in which performance data according to a musical score is modified based on performance data of fluctuations related to a desired music piece and the degree of fluctuation is changed (see, for example, Japanese Patent Application Laid-Open No. H10-157,027). For example, JP-A-10-2085
No. 6).

[0003]

According to the above-mentioned prior art, the performance data is corrected with reference to the performance data having fluctuation characteristics as the reference performance data. To change the note group to which fluctuations are imparted, it is necessary to prepare new reference performance data, and there is an inconvenience that the processing becomes complicated.
In addition, the same inconvenience occurs when the number of notes or the pitch is changed in a part of the music.

An object of the present invention is to provide a novel performance data creating method capable of easily creating fluctuation performance data for each desired note group.

Another object of the present invention is to provide a novel automatic performance apparatus which can easily impart a fluctuation to a part of a music to be automatically performed.

[0006]

According to a first aspect of the present invention, there is provided a musical performance data creating method comprising the steps of: preparing reference musical performance data having fluctuation characteristics corresponding to the types of notes in a time-series musical note group; Extracting performance data of a performance section corresponding to the number of notes in the reference performance data from the reference performance data;
Correcting according to two notes.

According to the first performance data creation method, reference performance data having fluctuation characteristics corresponding to note types (for example, sixteenth notes) in a time-series note group are prepared based on live performance and the like. Performance data of a performance section corresponding to the number of notes (for example, eight) in a note group is extracted from the reference performance data. Then, the performance data of the performance section related to the extraction is corrected according to at least one note in the note group. As a result, performance data having fluctuation characteristics based on live performances and the like and adapted to the note group can be obtained, and can be used for automatic performances and the like.

In the first performance data creation method, performance timing information (for example, sound generation timing information, mute timing, etc.) in performance data of a performance section is set such that the length of the performance section matches or approximates the section length represented by the note group. Information, etc.), the number of notes in a note group changes,
This can be dealt with by simply changing the number of notes when extracting data, and there is no need to change the reference performance data. Further, when the pitch information in the performance data of the performance section is corrected according to the pitch represented by the note group, even if the pitch in the note group is changed, the pitch information may be corrected according to the changed pitch. The reference performance data need not be changed. Therefore, the processing is simplified.

A second performance data creating method according to the present invention prepares reference performance data of fluctuation characteristics corresponding to note types in a note group representing a repetitive performance of one of a plurality of tones constituting a chord. Extracting the performance data of the performance section corresponding to the number of notes in the note group from the reference performance data, so that the length of the performance section matches or approximates the section length represented by the note group. Correcting the performance timing information in the performance data of the performance section, copying the corrected performance data of the performance section by one less than the number of constituent tones of the chord, Correcting the pitch information in the data according to the pitches of other sounds among the plurality of sounds constituting the chord.

According to the second performance data creation method, the first performance data
The same operation and effect as those of the method for creating performance data can be obtained. In addition, the performance data of the modified performance section is duplicated only for a portion that is one less than the number of tones constituting the chord (for example, 2), and the pitch information in the duplicated performance data is changed to the pitch of the corresponding chord component. , It is possible to create chord-based performance data based on reference performance data having a small number of single-tones.

A third performance data creating method according to the present invention prepares fluctuation-based reference performance data corresponding to note types in a note group representing a repetitive performance of a plurality of sounds having different sounding timings and pitches. Extracting, from the reference performance data, performance data of a performance section corresponding to the number of note sets corresponding to the plurality of notes in the note group,
Modifying the performance data of the performance section in accordance with at least one note in the note group.

According to the third performance data creating method, the first performance data
The same operation and effect as those of the method for creating performance data can be obtained. In addition, fluctuation reference performance data corresponding to note types in a group of notes representing repetitive performances of a plurality of sounds having different sounding timings and pitches are prepared, so that preceding and subsequent sounds are temporally overlapped. Thus, it is possible to create double tone performance data having a natural fluctuation characteristic that is pronounced.

An automatic performance device according to the present invention has a first storage means for storing performance data for automatically playing a musical piece, and a second storage means for storing fluctuation-based reference performance data for each time-series note group. Storage means, designating means for designating a desired time-series note group in the music, and reading out the reference performance data corresponding to the note group designated by the designating means from the second storage means. Means for extracting performance data of a performance section corresponding to the number of notes in the note group specified by the specifying means from the reference performance data read from the second storage means; First correcting means for correcting the extracted performance data of the performance section in accordance with at least one note in the note group specified by the specifying means; and specifying the performance data in the first storage means by the specifying means. Second correction means for correcting performance data relating to the selected note group in accordance with the performance data of the performance section corrected by the first correction means, and the performance data including the performance data corrected by the second correction means. Automatic performance means for automatically performing the music according to the performance data in the first storage means.

According to the automatic performance device of the present invention, when a desired note group is designated by the designation means, the reference performance data corresponding to the designated note group is read out from the second storage means, and the reading related data is read. Performance data of a performance section corresponding to the number of notes in the designated note group is extracted from the reference performance data. The extracted performance data of the performance section is corrected in accordance with the specified note group, and the performance data of the specified note group in the performance data of the first storage means is corrected in accordance with the performance data of the corrected performance section. The automatic performance of the music is performed according to the performance data in the first storage means including the performance data relating to the correction. As a result, in a music piece that is automatically played, a fluctuation portion based on the reference performance data is given to a performance point corresponding to a designated note group.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail by taking tremolo performance as an example. According to the orchestra, the tromolo performance is a repetitive performance of a large number of musical notes in time. Generally, a tremolo symbol is represented by adding a beard H to a note N as illustrated in FIG.
The music is converted into notes for tremolo performance as shown in FIG. The type of notes for tremolo performance is determined according to the number of beards H. If the number of beards is one, it is an eighth note. If it is two, it is a sixteenth note. If it is three, it is a thirty-second note. Become. The number of notes for tremolo performance is a number obtained by dividing the note N into notes of a type corresponding to the number of whiskers H.

In the example of FIG. 1, note N is a half note,
Since the number of the whiskers H is two, the notes for tremolo performance are eight sixteenth notes. As another example, if the note N is a quarter note and the number of the whiskers H is 3, 8
The 32nd notes become tremolo performance notes.

By the way, in the live performance of natural musical instruments, even if the number of tremolo sounds is the same as the number of notes for tremolo performance, the length and volume of each tremolo sound are often not constant. Sounds natural. That is, in a live performance, various non-uniformities and instabilities such as fluctuations in tempo and fluctuations in volume that the player does or do not notice greatly contribute to the freshness and naturalness.

FIG. 2 shows an example of a live tremolo performance.
Therefore, the horizontal axis represents time, and the vertical axis represents pitch. B₁
~ B₇Indicates the beat timing, and T₁, T₂
... T ₈, T₉... are the notes N in FIG.₁, N₂... N₈
Shows a tremolo sound corresponding to. Length of each tremolo sound
Is represented by the length of the black bar. According to FIG.
It can be seen that there is variation in the sound length of the b sound. Figure 2 shows
Is not performed, but the volume of the tremolo sound also varies.
You.

In the automatic performance of music, it is conceivable to create fluctuation performance data for automatic performance based on the live performance in order to realize natural performance, beautiful performance, and vivid performance. FIG. 3 shows an example of performance data created based on the live performance of FIG. With respect to the tremolo _{T 1,} timing data _{TM 11} and the note-on event data _{NE 11} and velocity data _{VL 11}
Timing data TM ₁₂ after a set of on event data along with a set of on-event data is arranged including bets
Set off event data including note-off event data NE ₁₂ is disposed with. Also for each tremolo of T ₂ later, the same way of the on-event data sets and off event data set is arranged. Thus, for example, an off event data relating tremolo _{T 8} set (timing data _{TM 82} and note-off event data NE
After a _82), of the on-event data relating tremolo _{T 9} sets (timing data _{TM 91,} note-on event data _{NE 91} and velocity data _{VL 91)}
Is arranged.

The timing data such as TM ₁₁ and TM ₉₁ indicate the sounding timing of the corresponding tremolo sound. The note-on event data such as NE ₁₁ and NE ₉₁ indicate the pitch of the corresponding tremolo sound by note number. Velocity data such as VL _{11 1} and VL ₉₁ indicate the volume of the corresponding tremolo sound. Timing data such as TM ₁₂ and TM ₈₂
This is for instructing the mute timing of the corresponding tremolo sound. The note-off event data such as NE ₁₂ and NE ₈₂ indicate the pitch of the corresponding tremolo sound by the note number. Each timing data represents a time from a previous event (or a beat if there is no previous event) by, for example, a count value of a tempo clock signal described later.

It is also possible to perform an automatic performance using the performance data as shown in FIG. 3 as automatic performance data. However, it is necessary to prepare new performance data based on the live performance in accordance with the change in the number of notes and the pitch of the tremolo performance notes. Therefore, the performance data as shown in FIG. Is not a good idea. Therefore,
In the present invention, new performance data is created by appropriately processing the performance data as shown in FIG. 3 based on the tremolo performance notes.

Next, a method for creating tremolo performance data according to the present invention will be described in detail with reference to FIGS. 5 shows a process step _S 1 to S _7.

[0023] In step S _1, for example, specify a tremolo symbol as shown in FIG. 4 (A) in the score display screen. In step S _2, it obtains the number and type of notes for tremolo playing as shown in FIG. 1 (B) based on the specified tremolo symbol. In the example of FIG. 4A, the note type is a sixteenth note and the number of notes is eight.

[0024] Next, in step S _3, to prepare a raw performance data corresponding to note the type obtained in step S _2. For this purpose, live performance data based on live performance is stored in advance in the database for each note type, and the live performance data corresponding to the determined note type may be read from the database. As an example, live performance data as shown in FIG. 3 is prepared based on the live performance of FIG.

Next, in step S _4, extracts the performance data of the performance period corresponding to the number of notes obtained in step S ₂ from the live data prepared in step S _3. FIG.
(C) shows an example of the extraction processing at this time. Performance data corresponding to the tremolo sounds T _{1 to} T ₈ is extracted from the live performance data based on the live performance of FIG. 4 (B). FIG. 6 shows the performance data extracted in this way. FIG.
In FIG. 3, the same parts as those in FIG. 3 are denoted by the same reference numerals, and the data and the timing data F in the section from T _{1 to} T ₈
TM is the extracted performance data. Here, the timing data FTM includes the tremolo sound T ₉ as shown in FIG.
(Timing from the time at which the TM ₈₂ is muted). With such T rather than the end timing of the interval of ₁ through T ₈ than match the mute timing interval last tremolo T _8, so that match the tone generation timing of the next tremolo T ₉ of T _8, T ₁ ~
To reflect the section last of silent time to connect the subsequent performance data in the data section of the T ₈ can produce a natural feeling. Also, if the end mark data FMD is added after the timing data FTM, the end can be easily confirmed at the time of data connection.

Next, in step S _5, the total length of the note for the tremolo playing length of the performance section of the extraction (Fig. 4
The timing value in the performance data of the performance section to be extracted is corrected so as to match or approximate the length of the note (A). The timing value corrected at this time is a performance timing value such as a tone generation timing and a mute timing. In the example of FIG. 6, the timing data TM ₁₁ and TM are used.
₁₂ are the values of TM ₈₁ and TM ₉₁ . FIG. 4 (D)
The length of the extracted section as shown in FIG. 4 (C) shows an example in which fixed in step S _5, the section length after modification,
The length is equivalent to two beats.

Next, step S₆Then, tremolo sound T₁
~ T₈The pitch of the note for tremolo performance
And if they do not match, the tremolo performance notes
Note number in performance data related to extraction according to pitch
To correct. The note number corrected at this time is shown in FIG.
In the example, the event data NE₁₁, NE₁₂... NE
₈₁, NE₈₂Note number.

Thereafter, step S₇Now, the velocity value
(Volume level) correction processing. That is, live music
The velocity value in the data is the velocity
Since it does not always match the city value, the sound for tremolo performance
Performance data related to extraction according to the volume of the first note in the note
Correct the velocity value in the data. The volume of the first note is
The tremolo symbol of FIG. 4A is displayed on the staff.
First, the velocity value and tremolo corresponding to the volume of the first note
B sound T₁Difference from velocity value in on-event data of
, The tremolo sound T₁In the on-event data
Velocity value corresponding to the volume of the first note
To be rewritten. And T₁T other than₂~ T ₈Each tremolo
The velocity value in the sound on-event data was determined first.
Correction is performed by addition and subtraction processing according to the difference between the velocity values.

[0029] As a result, for example, if the volume of the tremolo T ₁ is in the velocity value to that "20" increase (or decrease) the volume of the tremolo T ₂ through T ₈ is also in the velocity value "20" increases ( or decrease) will be, moreover be in the performance data of the section of the T ₂ through T ₈ is maintained the volume variation state of the tremolo during live.

FIG. 7 shows an example of a method for creating long performance data based on short live performance data. In this case, as the live performance data, performance data of a section corresponding to the tremolo sounds T _{1 to} T ₁₂ is prepared as shown in FIG. The performance data prepared at this time is tremolo sound T as data indicating the end timing ED.
_It includes timing data indicating the sound generation timing of the tremolo sound following the _twelve .

[0031] In step S ₄ in FIG. 5, after extracting all data to the end timing ED from live data prepared from tremolo T ₁ of the top as shown in FIG. 7 (B), the head of the tremolo except for data about T ₁ extracts data relating to the required number of tremolo after the tremolo T _12, to connect first extraction data and the second extraction data. In FIG. 7, RH indicates the head position of the second extracted data. The reason why the _first tremolo sound T1 is excluded in the second extraction processing is that the _first note is emphasized in the tremolo performance, which is useful for giving naturalness. In some cases, two or more sounds from the beginning may be excluded in the second and subsequent extraction processes.

FIG. 8 shows a data connection state in the method of FIG. At the end of the first extraction data,
End mark data FMD is arranged following the off event data set after the timing data FTM of tremolo T ₁₂ (data indicating the termination timing ED). Extracting data the second time is to exclude the timing data TM ₂₁ of tremolo T ₂ is intended to include note-on event data NE _{2 1} of tremolo T ₂ as the top data. When connecting the first and second extraction data, it connects the note-on event data NE ₂₁ to the timing data FTM instead of end mark data FMD. After that, if necessary, the same extraction and
Performs connection processing, performing the steps S ₅ and subsequent steps in FIG.

FIG. 9 shows another example of a method for creating long performance data based on short live performance data.
The same parts as those described above are denoted by the same reference numerals, and detailed description will be omitted.

The features of the method of FIG. 9 is a sound generation timing of the tremolo T ₁₂ as shown in FIG. 9 (A) in that the end timing ED. In this case, tremolo T ₁₂ in the first extracted data as shown in FIG. 9 (B) is not pronounced.

FIG. 10 shows a data connection state in the method of FIG. At the end of the first extraction data, note-on tremolo T ₁₂ continues off timing data of the event data set after tremolo T ₁₂ (indicate a tone generation timing) TM2 of tremolo T ₁₁ event data NE2 Is arranged. The second extraction data is the same as in FIG. When connecting the first and second extracted data, the Norton event data NE ₂₁ is used instead of the note-on event data NE2.
Is connected to the timing data TM2. Thereafter, if necessary, the same extraction and connection processing as described above is performed.
Performing Step S ₅ and subsequent steps in FIG.

Next, a description will be given of a method of creating performance data relating to a double tone tremolo. In the case of a multi-tone tremolo, since there can be a plurality of performance modes for one tremolo symbol, it is necessary to determine the performance mode first and prepare live performance data according to the performance mode.

As an example, FIGS. 11A and 12A
Shows the same double tone tremolo symbol. This tremolo symbol may be converted into a chord repetition type tremolo performance note as shown in FIG.
As shown in FIG. 2 (B), the musical note may be played after being converted into a note for tremolo performance of an alternating tone performance type. For example, in the case of a musical instrument such as a violin, which can produce multiple tones and which can be easily repeatedly played, a chord is repeatedly played as shown in FIG. 11B. On the other hand, when it is not possible to produce multiple sounds like a wind instrument, or when it is difficult to repeatedly play multiple sounds like a piano, it is necessary to alternately play the chords as shown in FIG. General.

To create performance data for the tremolo performance of FIG. 11B, the same method as described above with reference to FIG. 5 can be used. In this case, as the live performance data, as shown in FIG. 11B, data based on a 16th note double tone tremolo performance is prepared.

As another method for creating performance data for the tremolo performance shown in FIG. 11B, live performance data based on a repetitive performance of one of a plurality of tones constituting a chord as shown in FIG. Is also available. This method has an advantage that the amount of prepared live performance data can be reduced.

In the step of FIG. 13A, a desired double tone tremolo symbol is designated. As an example, a double tone tremolo symbol having half notes N ₁₁ and N ₁₂ and two whiskers is specified. Then, based on the designated tremolo symbol, the type and number of tremolo performance notes are obtained as shown in FIG. The number of notes is one of upper and lower notes. In the example of FIG. 11B, the note type is a sixteenth note and the number of notes is eight.

Next, in the step of FIG. 13 (B), live performance data corresponding to the previously determined note type is prepared. The live performance data is not based on a chord performance, but based on a repetitive performance of one of a plurality of tones constituting the chord. In the example of FIG. 13 (B), providing a live performance data based on Tremolo playing sixteenth notes corresponding to the note N _11.

Next, in the step of FIG. 13C, the performance data of the performance section corresponding to the previously determined number of notes is extracted from the previously prepared live performance data. The extraction process is performed in the same manner as described above with respect to step S ₄ in FIG. 5, the extraction and connection process in the same manner as towards the live data is described in FIG. 7 or 9 and shorter. In the example of FIG. 13 (C), extracts the performance data of the first sequence corresponding to the tremolo sounds _T 1 through T _8. The performance data includes a tremolo sound T ₈
Of the next tremolo sound.

Next, in the step of FIG.
Correct timing values in the extraction performance data in the same manner as described above with respect to step S ₅ to adjust the length of the playing section of the extraction. In the example of FIG. 13 (D), adjusted to a length corresponding to the length of the performance section including the tremolo T ₁ through T ₈ to 2 beats.

Next, in the step of FIG. 13 (E), the performance data of the performance section that has been corrected
Duplicate only a few minutes. In the example of FIG. 13 (E), the number of chord constituent notes is 2, the performance data of the first sequence corresponding to the tremolo sounds _T 1 through T ₈ copy once tremolo _t 1 ~t ₈ A corresponding second series of performance data is obtained.

Thereafter, the note number (pitch) in the performance data relating to the duplication is corrected in accordance with the pitch of another tone among a plurality of tones constituting the chord. In the example of FIG. 13 (E), it corrects the note number in the performance data corresponding to the tremolo _t 1 ~t ₈ according note _{N 1 2} pitches of Figure 13 (A). Also, when the pitch of the tremolo T ₁ through T ₈ does not follow the pitch of a note N ₁₁ modifies the note number in the performance data corresponding to the tremolo T ₁ through T ₈ in accordance pitch of a note N ₁₁ .

According to the above-mentioned processing, two series of performance data corresponding to the two tones constituting the chord are obtained as shown in FIG. However, in these two series of performance data, the volume is the same and the sounding timing is the same for each of the two sounds that are simultaneously sounded, and it is inevitable that the sense of chord becomes unnatural. Therefore, in the present invention, the correction is made so that the velocity value differs at the sounding timing of the chord between the two series of performance data and the sounding timing value is changed.

In the velocity value correction processing, FIG.
When the chord is composed of the lowest tone ML, the middle tone MT, and the highest tone MH as shown in FIG. 5, and the volumes of the lowest tone, the middle tone, and the highest tone are VML, VMT, and VMH, respectively, the magnitude relation of the volume is VMH> VML. > VMT,
The balance as a chord tends to be better. For example,
When performing volume control by setting the velocity value to a value within the range of 0 to 127, the velocity value of the lowest sound is set to a value of about -5 to -10 for the highest sound, and the velocity value of the intermediate sound is set to-
It is good to set each to about 10-20. Also, instead of always lowering the volume by the determined value,
It is better to change the adjustment at random as it becomes smaller. For example, the velocity value of the lowest sound may be set to -10 with respect to the highest sound, and may be randomly changed within a range of ± 10 over time.

Although the above-described velocity value correction processing can provide a considerably natural non-uniformity, if the volume is controlled such that the root note is large and the third note is small in accordance with the chord element, for example. , Can also give a more natural non-uniformity.

[0049] In the velocity value correction processing for the performance data of two series in FIG. 13 (E), for the performance data of the second sequence corresponding to the tremolo t ₁ ~t ₈ is a first highest pitch to correct the velocity values. That is, note N
After determining the difference between the velocity value of the velocity value and during the on-event data of the tremolo sound t ₁ corresponding to the (top note of the volume in the note for the best sound correspondence of the tremolo playing) the specified volume with respect to _12, tremolo sound t The velocity value in the on-event data ₁ is rewritten to a velocity value corresponding to the volume of the first note. Then, t _{2 to} t other than t ₁
The velocity value in the on-event data of each tremolo sound of No. ₈ is corrected by addition / subtraction processing according to the difference between the velocity values previously obtained.

Next, the velocity value of the performance data of the first series corresponding to the lowest tones T _{1 to} T ₈ is corrected. In this case, the predetermined value than the velocity value corresponding to the volume velocity value corresponding to the (beginning note the volume of notes for tremolo playing lowest note corresponding) is specified with respect to the note N ₁₂ specified volume with respect to the note N ₁₁ (e.g. 10) (if not lower by a predetermined value, lower by a predetermined value). Then, to correct the velocity values in the performance data of the first line as in the case of the performance data of the second sequence using the velocity values corresponding to the specified volume with respect to the note N _11. As a result, the tremolo sound T ₁
Velocity value corresponding to the ~T _8, the tremolo sound _t 1 ~t
_The respective velocity values are lower than the velocity values corresponding to ₈ by a predetermined value (for example, 10).

According to the above-described processing, each velocity value in the first series of performance data differs from the corresponding velocity value in the second series of performance data by a predetermined value, but the first series of performance data is copied. Is the second series of performance data, so that the first and second series of performance data have the same variation in the volume of the tremolo sound. In order to make such a sound volume variation state different, at least one of the first and second performance data can be subjected to the process of randomly correcting the velocity value as described above. .

In the sounding timing value correcting process, the sounding timing value in the first series of performance data is set at random within a predetermined range for each chord sounding timing, and the sounding timing value in the second series of performance data is set. Is randomly set within a predetermined range for each chord sounding timing. In this case, by setting the predetermined range to, for example, a range of ± 5, temporal variation can be expressed. If the amount of time change is too large, it will not be heard as a chord rather than a variation over time, so it is necessary to select an appropriate value. Assuming that the resolution of the quarter note is 480, the time change amount is 5-1 when the tempo is about quarter note = 120.
It is appropriate to set the value to about 0.

When creating performance data for the tremolo performance of FIG. 12B, live performance data of a single tone may be used, but it is better to use live performance data of a double tone. Taking a piano as an example, a single-tone tremolo is a single-tone tremolo, and it is mechanically impossible for the sounding times of the preceding and following sounds to overlap. On the other hand, in the double tone tremolo, the sounding times of the preceding and following sounds can overlap, and it is rather natural to overlap.

FIG. 15 shows a method of creating performance data using live performance data of multiple sounds. In the step of FIG. 15A, the same symbol of the double tone tremolo as in the case of FIG. Then, based on the designated tremolo symbol, the type and number of tremolo performance notes are obtained as shown in FIG. The number of notes is determined by the number of sets, with the upper and lower two notes as one set. In the example of FIG. 12B, the note type is a sixteenth note and the number of notes is four.

Next, in the step of FIG.
Prepare live performance data of multiple sounds corresponding to the note type
You. In the example of FIG. 15B, as shown in FIG.
Two 16th notes with different sounding timing and pitch
Prepare live performance data based on repetitive performances. In this example
Is an alternating sound with different pitches (eg, T₁₁And T₂₁, T
₂₁And T₁₂))
You.

Next, in the step of FIG. 15C, the performance data of the performance section corresponding to the previously determined number of notes is extracted from the previously prepared live performance data. In the example of FIG. 15 (C), it extracts the performance data of the four sets of notes to play interval corresponding (performance data relating tremolo _{T 11 ~T 14, T 2 1} ~T 24). The performance data includes a tremolo sound T
Together it includes timing data indicating the tone generation timing of the next tremolo T ₁₅ of _14, tremolo T
It contains timing data for instructing the sound generation timing of the next tremolo T _{25 24.} Tremolo _{T 24}
The mute timing, is later than the pronunciation timing of the tremolo T _15. This is due to the overlapping of the sounding times.

Next, to adjust the length of the performance section of the extraction to correct the timing values in the extraction performance data in the same manner as described above with respect to step S ₅ of FIG. FIG.
In the example of (C), the length of the performance section including the tremolo sounds T _{11 to} T ₁₄ is set to a length corresponding to two beats, and the tremolo sound T 11
₂₄ silence timing is allowed to be outside the performance section. This helps to give a natural feeling.
After this, if necessary, can be performed pitch and volume of the modification treatment in the same manner as described in step S _6, S ₇ of FIG.

FIG. 16 shows a method of creating long double-tone tremolo performance data based on short double-tone tremolo performance data. Parts similar to those shown in FIG. I do.

In the step of FIG.
The live performance data similar to that described in (B) is converted to a tremolo sound T.
₁₁ ~T _{14, T} 21 ~T prepared by playing time section corresponding to _24. The live data, as well as including the timing data for instructing the sound generation timing of the next tremolo T ₁₅ of tremolo T _14, includes timing data indicating sound generation timing of the next tremolo T ₂₅ of tremolo T ₂₄ In. EN indicates the end timing of the performance section corresponding to two beats.

In the step of FIG. 16B, the tremolo sounds T ₁₁ , T ₂₁ to T _{21 of the first} set are prepared from the prepared live performance data.
After extracting the performance data of the last set of tremolo sounds T ₁₄ and T ₂₄ (including timing data indicating the sounding timing of the tremolo sounds T ₁₅ and T ₂₅ ), the performance data relating to the first set of tremolo sounds T ₁₁ and T _{21 is} extracted. performance data (tremolo _T 15 of the second set of tremolo _{T 12, T 22} ~ end sets of tremolo _{T 14, T 24} except, T
₍ Including timing data designating ₂₅ sounding timings). In FIG. 16, RH indicates the head position of the second extracted data. The reason why the first set of tremolo sounds T ₁₁ and T ₂₁ are excluded in the second extraction processing is that the first set of notes is emphasized in the multi-tone tremolo performance, which is useful for giving naturalness.

The first extracted data and the second extracted data are connected as shown in FIG. That is, the connecting set of tremolo T _{1 2} ON event data to the set of off event data tremolo T _14, off event data tremolo T ₂₄ to the set of on-event data tremolo T ₁₂ set connect to connect the set of on-event data tremolo T ₂₂ to set off event data tremolo T _24. Here, the timing data tremolo T _12, tremolo T
Using the timing data indicating the tone generation timing TM15 of _15, as the timing data tremolo _{T 22,} using the timing data indicating the tone generation timing TM25 of tremolo _{T 25.} This way,
A natural feeling can be obtained by reflecting the silence time after the performance of the last set of notes.

After that, if necessary, the same extraction and connection processing as described above is performed, and the processing after the section length adjustment described in FIG. 15C is performed.

FIG. 18 exemplifies the conversion of various double tone tremolo symbols into notes for double tone tremolo performance.

FIGS. 18 (A) to 18 (C) show examples of tremolo performances by repeating single and double tones. FIG.
In the example of (C), the combination of repetitions is explicitly shown on the score, but in the examples of FIGS. 18A and 18B, the combination of repetitions is not explicitly shown in the score. Unless explicitly stated, the combination of repetitions is free, but the function of the finger follows the normal playing style.

In the example of FIG. 18 (A), when playing with the right hand, it is general to divide into 2: 1 (double tone + single tone). When playing with the left hand, 1: 2 (single tone + single tone) In general, it is divided into multiple sounds. That is, the thumb becomes a single note,
It is customary for the other fingers to have multiple sounds. Since the thumb is stronger than the other fingers, it is considered that such a combination naturally occurs when the balance is achieved. Therefore,
When it is possible to determine whether a specific tremolo symbol is played with the right or left hand by looking at the entire score, the division mode is determined according to the determination result.

FIGS. 18D and 18E show two tremolo performance notes based on the same double tone tremolo symbol. The example of FIG. 18D is divided into 1: 3 (single sound + multiple sounds), and the example of FIG. 18E is 2: 2 (single sound + multiple sounds).
(Multiple sounds). Assuming right-handed playing,
Generally, the performance is as shown in FIG. But,
It is not without reason that the performance as shown in FIG. 18E is performed aiming at a certain effect. Therefore, it is only necessary to determine in advance which division mode to use.

It may be unclear whether a particular tremolo is played with the right hand or the left hand just by looking at the entire score. In such a case, it may be a situation where one of the hands is used, so that the division form may be determined assuming that the performance is performed with the right hand.

After the division mode is determined as described above, a live performance is performed in accordance with the tremolo performance note corresponding to the determined division mode, and live performance data is prepared. And
Using the prepared live performance data, performance data with fluctuation characteristics is created as described with reference to FIGS.

FIG. 19 shows a circuit configuration of an electronic musical instrument provided with an automatic performance device according to an embodiment of the present invention. In this electronic musical instrument, tone generation, automatic performance, and the like are controlled by a small computer such as a personal computer.

The bus 10 has a CPU (central processing unit) 1
2, ROM (Read Only Memory) 14, RAM
(Random access memory) 16, detection circuit 18,
20, a display control circuit 22, a timer 24, an interface 26, a storage device 28, a sound source device 30, a DSP (digital signal processor) 32, a sound system 34, and the like.

The CPU 12 executes various processes for tone generation, automatic performance, and the like according to the program stored in the ROM 14. These processes are described in FIG.
24 will be described later.

The RAM 16 includes a large number of storage areas used as flags, registers, and the like in various processes performed by the CPU 12. The storage areas related to the embodiment of the present invention include a run / stop flag RUN and a fluctuation addition flag UA. , Fluctuation enhancement flag UB, music performance data storage units MA and MB, reference performance data storage unit RA, extracted performance data storage unit RB, and the like.

The detection circuit 18 detects key operation information from the keyboard 36. The detection circuit 20 includes an operator group 38
To detect operation information of various operators. The operator group 38 includes various operators provided on the panel surface, a keyboard capable of inputting characters and numerals, and a mouse as a pointing device. As operators related to the embodiment of the present invention, a fluctuation addition switch, a fluctuation enhancement switch, a start / stop switch, and the like are provided. In addition, mouse and keyboard, music selection operation,
Used for tremolo symbol designation operation.

The display control circuit 22 controls the display operation of the display device 40 to perform various displays (for example, musical score display).
Is what makes it possible.

The timer 24 receives the given tempo data T
A clock pulse is generated as a tempo clock signal TCL at a cycle corresponding to D. The tempo clock signal TCL is supplied to the CPU 12 as an interrupt command.
The CPU 12 starts the interrupt processing of FIG. 24 every time a clock pulse as the tempo clock signal TCL is generated. By this interrupt processing, an automatic performance is performed based on the music performance data in the storage unit MA or MB.

The interface 26 is a MIDI (Musi)
It is provided to perform information communication such as musical tone control information and performance information with an external device 42 such as a MIDI device.

The storage device 28 includes HD (hard disk), FD (floppy disk), CD (compact disk), DVD (digital multipurpose disk), M
One or more types of storage media such as O (magneto-optical disk) can be attached and detached. When a desired storage medium is mounted on the storage device 28, data can be transferred from the storage medium to the RAM 16. Also, if the attached storage medium is HD
If it is a writable device such as a FD or FD, the data in the RAM 16 can be transferred to a storage medium. In this embodiment, performance data of a desired music piece can be selected and read out from a storage medium storing performance data of a large number of music pieces, and an automatic performance can be performed.

As the program storage means, the ROM 14
Instead of the storage medium of the storage device 28 (HD, CD, D
VD, MO, etc.). In this case, the program stored in the storage medium is transferred from the storage device 28 to the RAM.
Transfer to 16. Then, the CPU 12 is operated according to the program stored in the RAM 16. By doing so, it is possible to easily add a program, upgrade a version, and the like. By connecting the storage device 28 to a communication network, music performance data and reference performance data to be edited can be obtained, and the program can be easily updated.

The tone generator 30 has a large number (for example, 64) of tone generating channels. When there is a sounding request based on a key depression operation on the keyboard 36 and / or data reading from the storage unit MA or MB, the CPU 12 generates a sounding instruction signal, a note number (pitch information), and velocity corresponding to each sounding request. The value (volume information) is assigned to one of the idle tone generation channels. Therefore, a tone signal having a pitch corresponding to the note number is generated from the assigned tone generating channel at a volume corresponding to the velocity value. Further, when there is a mute request based on a key release operation on the keyboard 36 and / or data reading from the storage unit MA or MB, the CPU 12 generates a musical tone signal corresponding to a note number associated with each mute request. A mute command signal is supplied to the generation channel to start attenuating the tone signal being generated.

The DSP 32 can apply various effects such as timbre, volume and sound to the digital tone signal from the tone generator 30. The effects to be applied are the panel operators in the operator group 38. Is instructed by the CPU 12 based on the above operation.

The sound system 34 includes a digital / analog converter, an output amplifier, a speaker and the like.
The tone signal from the DSP 32 is converted into sound.

FIGS. 20 and 21 show the flow of the processing of the main routine, which is started when the power is turned on or the like. In step 50, an initial setting process is performed to initialize various flags, registers, and the like.

Next, in step 52, it is determined whether or not there is a music selection operation. On the screen of the display 40, the titles of selectable songs are displayed. The user can select a desired song from the displayed songs by operating a mouse or the like.
If the determination result of step 52 is affirmative (Y), the process proceeds to step 54.

In step 54, the storage device 28 (or R
The performance data of the selected music piece is read from the OM 14) and stored in the storage units MA and MB. In step 56, the musical score is displayed on the screen of the display 40 based on the performance data in the storage unit MB. Note that the storage device 28 (or the ROM
Each piece of music performance data 14) is assumed to have the same musical score and has no fluctuation.

When the result of the determination at step 52 is negative (N) or when the processing at step 56 is completed, it is determined at step 58 whether or not the fluctuation addition switch (SW) has been operated. If the result of this determination is affirmative (Y), the contents of the flag UA are inverted at step 60 (1 if 0, 1
If so, set it to 0).

When the result of the determination in step 58 is negative (N) or when the process in step 60 is completed, it is determined in step 62 whether or not the operation of the fluctuation enhancement switch (SW) has been performed. If the result of this determination is affirmative (Y), the contents of the flag UB are inverted at step 64.

If the decision result in the step 62 is negative (N) or the process in the step 64 is completed, it is decided in a step 66 whether or not the start / stop switch (SW) is operated. If the result of this determination is affirmative (Y), the contents of the flag RUN are inverted at step 68.

When the result of the determination at step 66 is negative (N) or when the processing at step 68 is completed, it is determined at step 70 whether the value of the flag RUN has changed from 0 to 1. If the result of this determination is affirmative (Y), a read position is set at the head position in the storage units MA and MB in step 72. This is a preparation process for reading performance data.

When the result of the determination at step 70 is negative (N) or when the processing at step 72 is completed, it is determined at step 74 whether the value of RUN has changed from 1 to 0. If the result of this determination is negative (Y), step 7
At 6, the automatic performance sound is silenced. That is, the tone generator 30 starts to attenuate all tone signals that are being generated as automatic performance tone signals.

If the decision result in the step 74 is negative (N) or the process in the step 76 is completed, it is decided in a step 78 whether or not a tremolo symbol designation operation is performed.
The user can designate a tremolo symbol for which fluctuation is desired to be added in the musical score displayed on the screen of the display unit 40 by mouse operation or the like. If the determination result of step 78 is affirmative (Y), the process proceeds to step 80.

In step 80, it is determined whether the flag UA is 1 (whether the mode is the fluctuation addition mode). If the result of this determination is affirmative (Y), a subroutine of the fluctuation adding process is executed in step 82. See Figure 2 for this subroutine.
2 will be described later.

When the result of the determination at step 78 is negative (N), when the result of the determination at step 80 is negative (N), or when the processing at step 82 has been performed, the flag UA is determined at step 84. , UB are both 1. If the determination result is affirmative (Y), the performance data in the storage unit MB is randomly corrected within a predetermined range in step 86. For example, the tone generation timing value and the velocity value in the performance data are each randomly corrected within a predetermined range.
As a result, the performance data to which the fluctuation has been added in the fluctuation adding process in step 82 has the fluctuation further enhanced.

When the result of the determination in step 84 is negative (Y) or when the processing in step 86 is completed, other processing is performed in step 88. Other processes include a tone generation process based on a key operation on the keyboard 36 and a setting process of various tone parameters (for example, tone, effect, automatic performance tempo, etc.) based on an operation of a panel operator.

After step 88, it is determined in step 90 whether there is a termination instruction such as power-off. If the result of this determination is negative (Y), the process returns to step 52, and step 52
The subsequent processing is repeated in the same manner as described above. If the result of the determination in step 90 is affirmative (Y), the processing ends.

FIG. 22 shows the flow of the fluctuation adding process. In step 100, step 78 in FIG.
Based on the tremolo symbols according to specifications in the in the same manner as described in step S ₂ in FIG. 5 obtains the type and number of notes for tremolo playing.

Next, at step 102, the live performance data corresponding to the previously determined note type is read from the database and stored in the storage unit RA as the reference performance data. As the database, a storage device 28 (or ROM 14) is used, in which live performance data is stored in advance for each note type. As described above with reference to FIGS.
If there are a plurality of performance modes for one tremolo symbol, a plurality of performance modes (tremolo performance notes) are displayed on the screen of the display unit 40, and when the tremolo symbol is designated by the user, the user selects a desired performance mode. To be specified. Then, in step 102, the live performance data corresponding to the performance form designated by the user is read out and stored in the storage unit RA. In this way, the user's preferences can be further reflected on the created performance data.

Next, at step 104, the performance data of the performance section corresponding to the previously determined number of notes is extracted from the performance data of the storage unit RA and stored in the storage unit RB. Extraction process at this time can be carried out in the same manner as described in step S ₄ in FIG.

[0098] Next, in step 106, the correction step S ₅ to S ₇ in in the same manner as described performance timing value in the performance data in the storage unit RB according note for tremolo playing Figure 5, note number, the velocity value . As a result, in the storage unit RB, the generation of the fluctuation performance data based on the live performance is completed.

Thereafter, at step 108, the storage unit RB
The performance data in the storage unit MB is corrected according to the performance data of the storage unit MB. The correction of the performance data is performed, for example, as shown in FIG.

That is, the storage section MB stores performance data of a selected tune, and the performance data corresponding to the designated tremolo symbol is obtained from the set of the tremolo tone T _N1 on-event data and the tremolo tone T _{N1. It is} assumed that the data is up to a set of _NM off-event data. On the other hand, in the storage unit RB, as performance data corresponding to the tremolo symbol of the specified data from the set of off event data tremolo T _n1 to set off event data tremolo T _nm, tremolo T _nm , And timing data FTM indicating the timing of sounding the next tremolo sound, and data FMD representing the end mark. Tremolo sound T
The timing data of _N1 indicates the time from the mute timing of the immediately preceding tremolo sound to the beginning of the beat, and the tremolo sound T
_{Assuming that} the timing data of _n1 represents the time from the start, the timing value of T _{N1 and} the timing value of T _n1 are added to the new timing value of T _N1 . In addition, from the note-on event data of T _N1 , T
_The data up to the note-off event data of _NM is T _n1
Is rewritten according to the data from the note-on event data of _{Tnm to} the note-off event data of Tnm. further,
Rewritten in accordance with the timing data FTM the timing data of the next tremolo _{T NN} of T _NM. This is to reflect the silent time at the end of the section in the performance data in the storage unit MB. When the performance data has been corrected as described above, the process returns to the routines of FIGS.

FIG. 24 shows an interrupt process for automatic performance. In step 100, it is determined whether the flag RUN is 1. If the result of this determination is negative (N), the processing described below is unnecessary, and the routine returns to the routines of FIGS.

The determination result of step 100 is positive (Y)
If so, the process proceeds to step 112, where it is determined whether the flag UA is 1 (whether the mode is the fluctuation addition mode). If the result of this determination is negative (N), it is determined at step 114 whether or not it is a reproduction timing. This determination is made based on the timing value indicated by the timing data of the read position in the storage unit MA and the tempo counter (RAM1) that counts the tempo clock signal TCL.
6) is checked to see if it matches the count value. When the determination result of step 114 is negative (N), the process returns to the routine of FIGS.

The determination result of step 114 is positive (Y)
In step 116, the event data paired with the timing data determined to be the reproduction timing is read from the storage unit MA in step 116. The event data read out at this time is note-on event data and velocity data, note-off event data, or other event (for example, a tone or volume change event) data.

Next, at step 118, it is determined whether or not the read out is note event data. If the result of this determination is affirmative (Y), it is determined in step 120 whether the data is note-on event data. If the result of this determination is affirmative (Y), sound generation processing is performed in step 122. That is,
The note number in the note-on event data for reading, the velocity value in the velocity data, and the tone generation command signal are assigned to one of the idle tone generation channels to generate a tone signal. Then, the process returns to the routine of FIGS.

The result of the determination in step 120 is negative (N)
If so, it means that a note-off event has occurred, and a mute process is performed in step 124. That is, according to the same note number as the note number in the note-off event data relating to the reading, a mute command signal is supplied to the tone generation channel for which the tone signal is being generated, and the tone signal is attenuated. Then, the process returns to the routine of FIGS.

The determination result of step 118 is negative (N)
Is satisfied, a process corresponding to the read data is performed in step 126. For example, if the read data is data instructing to change the volume corresponding to the total volume, the volume is changed according to the data. When the event data is read as described above, the read position is changed to the position of the timing data in the next event data.

According to the processing of steps 114 to 126 described above, an automatic performance without fluctuation can be performed based on the performance data in the storage unit MA.

The determination result of step 112 is positive (Y)
(In the fluctuation addition mode), it is determined in step 128 whether or not it is a reproduction timing. This determination can be made by checking whether the timing value indicated by the timing data of the read position in the storage unit MB matches the count value of the above-described tempo counter. When the determination result of step 128 is negative (N),
Return to the routine of FIGS.

The judgment result of step 128 is positive (Y)
In step 130, the event data paired with the timing data determined to be the reproduction timing in the storage unit MB in step 130 is read. The event data read at this time is of the same type as described in step 116.

Next, at step 132, it is determined whether or not the read out is the note event data. If the result of this determination is affirmative (Y), it is determined in step 120 whether the data is note-on event data. The result of this determination is positive (Y)
If so, the tone generation process is performed in the same manner as described above in step 122 to generate a tone signal.

The result of the determination in step 120 is negative (N).
If so, the mute processing is performed in the same manner as described above in step 124, and the attenuation of the tone signal is started.

The determination result of step 132 is negative (N)
If so, the process moves to step 126, and the process according to the read data is performed in the same manner as described above.

Steps 128 to 132, 120 described above
According to the processes of Nos. To 126, the tremolo section is performed based on the performance data of the storage section MB (that is, the performance data having the fluctuation performance data created based on the live performance in the tremolo section as described in FIG. 23). Automatic performance can be performed with emphasis on the fluctuation of the sound. In addition, when the mode is set to the fluctuation enhancement mode, the fluctuation characteristics of the tremolo portion are further enhanced, and the fluctuation characteristics are given to portions other than the tremolo portion.

The present invention is not limited to the above embodiment, but can be implemented in various modified forms. For example, the following changes are possible.

(1) The performance data of the storage unit MB is transferred to the storage medium of the storage device 28, and at another occasion, the present electronic musical instrument or another electronic musical instrument performs an automatic performance based on the performance data of the storage medium. You may do so.

(2) The arithmetic processing for setting the volume is not limited to the addition / subtraction processing, but may be a multiplication / division processing. However, in one process, it is better not to use the addition / subtraction process and the multiplication / division process together.

(3) The performance data is not limited to tremolo performance data, and performance data of note groups such as arpeggios, trills, and glissandos may be created.

(4) The format of the performance data is not limited to the “event + relative time” method in which the event occurrence time is represented by a relative time from the immediately preceding event. "Event +"
"Absolute time" method, "Pitch (rest) + note length" method, which expresses the contents of the music with note pitch and note length and rest and rest length, secures a storage area for each minimum time unit of event occurrence However, an arbitrary method such as a method of storing an event in a storage area corresponding to an event occurrence time can be used.

[0119]

As described above, according to the present invention, the reference performance data of the fluctuation characteristic corresponding to the note type in the time-series note group is prepared, and the performance corresponding to the number of notes in the note group is prepared. Extract the performance data of the section from the reference performance data,
Since the performance data of the performance section related to the extraction is modified in accordance with at least one note in the note group, it is possible to easily create fluctuation performance data for each desired note group, and to perform the same kind of note Does not need to change the reference performance data, and has the effect of simplifying the processing.

Further, the performance timing information and the pitch information in the performance data of the extracted performance section are modified according to the section length and the pitch of the note group, respectively, so that the number of notes and the pitch in the note group are changed. Even if it is changed, there is no need to change the reference performance data, and an effect that processing is simplified can be obtained.

In addition, a desired note group is designated, and performance data corresponding to the designated note group is created by the method of the present invention, and a part of the performance data for automatic performance is corrected according to the created performance data. As a result, it is possible to easily add fluctuation to a part of the music that is automatically played, and to easily change the note group to which the fluctuation is applied. An effect that can be enjoyed easily can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of conversion of a tremolo symbol to a tremolo playing note.

FIG. 2 is a diagram showing an example of a live tremolo performance.

FIG. 3 is a diagram showing an example of performance data created based on the live performance of FIG. 2;

FIG. 4 is a diagram showing an example of a tremolo performance data creating method according to the present invention.

FIG. 5 is a flowchart illustrating the method of FIG. 4;

FIG. 6 is a diagram showing an example of performance data created by the methods of FIGS.

FIG. 7 is a diagram showing an example of a method for creating long performance data based on short live performance data.

FIG. 8 is a diagram showing a data connection state in the method of FIG. 7;

FIG. 9 is a diagram showing another example of a method for creating long performance data based on short live performance data.

FIG. 10 is a diagram showing a data connection state in the method of FIG. 9;

FIG. 11 is a diagram showing an example of conversion of a multi-tone tremolo symbol to a tremolo playing note.

FIG. 12 is a diagram showing another example of conversion of a double tone tremolo symbol into a tremolo playing note.

FIG. 13 is a diagram showing an example of a method for creating tremolo performance data of multiple sounds according to the present invention.

FIG. 14 is a diagram for explaining volume setting of chord constituent sounds in the method of FIG. 13;

FIG. 15 is a diagram showing another example of the method for creating tremolo performance data of multiple sounds according to the present invention.

FIG. 16 is a diagram showing an example of a method of creating long double-tone tremolo performance data based on short double-tone tremolo live performance data.

FIG. 17 is a diagram showing a data connection state in the method of FIG. 16;

FIG. 18 is a diagram exemplifying conversion of various multi-tone tremolo symbols into multi-note tremolo playing notes;

FIG. 19 is a block diagram showing a circuit configuration of an electronic musical instrument provided with the automatic performance device according to one embodiment of the present invention.

FIG. 20 is a flowchart showing a part of a main routine.

FIG. 21 is a flowchart showing the rest of the main routine.

FIG. 22 is a flowchart illustrating a subroutine for adding fluctuation.

FIG. 23 is a diagram showing an example of performance data correction.

FIG. 24 is a flowchart showing an interrupt process.

[Explanation of symbols]

10: bus, 12: CPU, 14: ROM, 16: RA
M, 18, 20: detection circuit, 22: display control circuit, 2
4: timer, 26: interface, 28: storage device, 30: sound source device, 32: DSP, 34: sound system, 36: keyboard, 38: operator group, 40: display device,
42: external device, RUN: running / stop flag, UA: fluctuation addition flag, UB: fluctuation enhancement flag, MA, M
B: music performance data storage, RA: reference performance data storage, RB: extracted performance data storage.

Claims (15)

[Claims] A step of preparing reference performance data of fluctuation characteristics corresponding to a note type in a time-series note group; and performing the performance data of a performance section corresponding to the number of notes in the note group with the reference performance data. And modifying the performance data of the performance section according to at least one note in the note group. 2. The performance timing information in the performance data of the performance section is corrected so that the length of the performance section matches or approximates the length of the section represented by the note group. How to create the described performance data. 3. The performance data creating method according to claim 1, wherein the correcting step corrects pitch information in performance data of the performance section according to a pitch represented by the note group. 4. The performance data according to claim 1, wherein in the extracting step, sounding timing information relating to a note following a last note of the performance section is extracted from the reference performance data and included in the performance data of the performance section. Performance data creation method described in any of them. 5. The preparing step prepares the reference performance data corresponding to a predetermined number of notes smaller than the number of notes in the note group, and the extracting step includes the first note in the reference performance data. 5. Performance data creation according to any one of claims 1 to 4, wherein the performance data of the performance section is extracted by extracting data of a required number of notes excluding the leading note after extracting data of a plurality of notes. Method. 6. A step of preparing reference performance data of fluctuation characteristics corresponding to a note type in a note group representing a repetitive performance of one of a plurality of notes constituting a chord; Extracting performance data of a corresponding performance section from the reference performance data; and performing timing in the performance data of the performance section so that the length of the performance section matches or approximates the section length represented by the note group. Modifying the information; replicating the modified performance data of the performance section by one less than the number of constituent notes of the chord; and constructing the chord with the pitch information in the performance data relating to the replication. Correcting according to the pitch of another sound of the plurality of sounds to be played. 7. The method according to claim 1, further comprising the step of modifying the volume information in the performance data of the modified performance section and the volume information in the duplicated performance data so as to differ at the sounding timing of the chord. 6. The performance data creation method described in 6. 8. The method according to claim 1, further comprising the step of modifying the sounding timing information in the performance data of the modified performance section and the sounding timing information in the duplicated performance data so as to differ in the sounding timing of the chord. The performance data creation method according to claim 6. 9. A step of preparing reference performance data of fluctuation characteristics corresponding to a note type in a note group representing a repetitive performance of a plurality of sounds having different sounding timings and pitches, and the plurality of notes in the note group. Extracting performance data of a performance section corresponding to the number of corresponding note sets from the reference performance data; and modifying the performance data of the performance section in accordance with at least one note in the note group. Data creation method. 10. The performance timing information in the performance data of the performance section is corrected so that the length of the performance section matches or approximates the length of the section represented by the note group. How to create the described performance data. 11. The performance data creating method according to claim 10, wherein, in the correcting step, silence timing information relating to the last note of the performance section allows to indicate a silence timing outside the performance section. 12. The preparing step includes preparing the reference performance data corresponding to a predetermined number of note sets smaller than the number of note sets corresponding to the plurality of notes in the note group, and extracting the reference performance data. By extracting performance data relating to a plurality of sets of notes including the first set of notes in the reference performance data and then extracting performance data relating to a required number of notes excluding the first set of notes, the performance of the performance section is extracted. The performance data creating method according to any one of claims 9 to 11, wherein data is extracted. 13. A first storage means for storing performance data for automatically playing music, a second storage means for storing fluctuation performance reference performance data for each time-series note group, Designating means for designating a desired time-series note group in the program; reading means for reading reference performance data corresponding to the note group designated by the designating means from the second storage means; Extracting means for extracting, from the reference performance data read from the storage means, performance data of a performance section corresponding to the number of notes in the note group specified by the specifying means; and extracting the performance data of the performance section extracted by the extraction means. First correcting means for correcting performance data in accordance with at least one note in the note group specified by the specifying means; performance relating to the note group specified by the specifying means in the performance data of the first storage means; Second correction means for correcting the performance data according to the performance data of the performance section corrected by the first correction means; and the first storage means including the performance data corrected by the second correction means. And an automatic performance means for automatically performing the music in accordance with the performance data. 14. The musical instrument further comprising: an instruction unit for designating a desired performance symbol in the music, and a conversion unit for converting the performance symbol specified by the instruction unit into a plurality of time-series note groups. 14. The automatic performance device according to claim 13, wherein the means specifies a desired note group among the plurality of time-series note groups. 15. A first storage means for storing performance data for automatically playing music, a second storage means for storing fluctuation performance reference performance data for each time-series note group, And a designation means for designating a desired time-series note group in the automatic performance device, wherein the reference performance data corresponding to the note group designated by the designation means is stored in the first And extracting from the reference performance data read from the second storage means performance data of a performance section corresponding to the number of notes in the note group specified by the specification means. And at least one of the musical note groups specified by the specifying means, the performance data of the performance section extracted in the extracting step.
A first correction step of correcting according to two notes; and a performance data of the musical note group specified by the specifying means in the performance data of the first storage means, the performance data of the performance section corrected in the first correction step. Recording a program comprising: a second correction step of correcting according to performance data; and a step of automatically performing the music in accordance with the performance data of the first storage means including the performance data corrected in the second correction step. Recording medium.