JPH07199930A - Automatic player - Google Patents

Abstract

PURPOSE:To provide the automatic player which precisely controls the emission timing of musical sounds in real time. CONSTITUTION:Performance information of plural tracks are stored in an automatic performance memory 2. A CPU 6 reads out performance information from this performance information storage means and calculates the sound emission timing so that sound emission timings of respective performance information are converged to a reference beat timing. After the sound emission timing is calculated, performance information corresponding to this sound emission timing is reproduced through a tone generator 3 when the time corresponding to this sound emission timing elapses.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic performance device for automatically performing performance based on performance information.

[0002]

2. Description of the Related Art Conventionally, some automatic performance devices such as sequencers have a function called "quantize". As shown in FIG. 11, this function adjusts the sounding timing of the input performance information to the just timing of the clock. By the way, the quantize function in a typical automatic performance device automatically edits the performance information from the beginning to the end of the song, and then plays the song, so you can use the quantize function in real time while listening to the song. I couldn't adjust the size.

On the other hand, according to Japanese Patent Application Laid-Open No. 3-228591, the quantizing is realized in real time by setting the clock cycle for reproducing the performance information to be short near the beat timing and long otherwise. The technology is disclosed. That is, the sounding timing of the original performance information is shown on the horizontal axis of FIG. 12, but by controlling the clock cycle as described above, each sounding timing is controlled at the timing shown on the vertical axis of FIG. To do. This makes it possible to adjust the quantize strength in real time while listening to music. In addition, a performance method called "swing" has been conventionally known in jazz performance. As shown in FIG. 6, this is to keep the performance timing for a strong beat at the original timing, while slightly delaying the performance timing for a weak beat.

[0004]

By the way, the performance information is generally composed of a plurality of tracks, and by reproducing this, it is possible to perform an ensemble with a wide variety of musical tones. At that time, you may want to quantize only some tracks. However, according to the technique disclosed in Japanese Unexamined Patent Publication No. 3-282591, the clock period itself is changed, so that it is impossible to quantize only some tracks. Further, although there is an automatic performance device that automatically performs the same swing on all of the plurality of tracks, it is not possible to independently perform the swing on each of the plurality of tracks. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an automatic performance device capable of precisely controlling the sounding timing of a musical sound in real time.

[0005]

In order to solve the above-mentioned problems, in the structure according to claim 1, performance information storage means for storing performance information of a plurality of tracks, each of which includes musical tone generation timing data, and The reading means for reading the performance information from the performance information storage means and the tone generation timing of the performance information read by the reading means are compared with the corresponding tone generation timing data and converged to the beat timing which is the reference of the corresponding track. As described above, the calculation means for calculating the sounding timing and the reproducing means for reproducing the performance information corresponding to the sounding timing when the time corresponding to the sounding timing elapses after the sounding timing is calculated. It is characterized by doing.

Further, in the structure according to claim 2,
Performance information storage means for storing performance information of a plurality of tracks, each of which includes musical sound generation timing data, reading means for reading the performance information from the performance information storage means, and musical sound generation timing of the performance information read by the reading means. When the data belongs to a predetermined time range of a predetermined track, a calculating means for calculating a sounding timing so that the reproduction time is delayed, and after the sounding timing is calculated, when a time corresponding to this sounding timing elapses, It is characterized by comprising a reproducing means for reproducing the performance information corresponding to the sounding timing.

[0007]

In the structure according to the first aspect, when the reading means reads the performance information from the performance information storage means, the calculating means calculates the sound generation timing so that the sound generation timing converges to the beat timing. When the time corresponding to this sounding timing elapses, the reproducing means reproduces the performance information corresponding to this sounding timing, so that the musical sound is quantized in real time.

Further, in the structure according to claim 2,
When the reading means reads the performance information from the performance information storage means, if the musical tone generation timing data of the performance information belongs to a predetermined time range of a predetermined track, the arithmetic means sets the sounding timing so that the reproduction time is delayed. Calculate
Then, when the time corresponding to this sounding timing elapses, the reproducing means reproduces the performance information corresponding to this sounding timing, so that the musical sound is swung in real time.

[0009]

EXAMPLES A. Configuration of Embodiments The configuration of an electronic musical instrument according to an embodiment of the present invention will be described below with reference to FIG. In the figure, 1 is an operator group,
A keyboard operated by the performer, a tone color operator, a display device, and the like are provided. Operation information for these various operators is output via the bus 9, and the information supplied via the bus 9 is displayed on the display device. indicate. In addition, in the operator group 1, (a) a quantize value setting switch that specifies the setting of the quantize value (the length of the beat timing for quantizing) and (b) the quantize strength (the sounding timing converges to the beat timing). (Quantity) setting, quantize strength setting switch, (c) swing ratio (swing strength) setting swing ratio setting switch, (d) memory switch for specifying performance information storage, (e) A RUN switch for instructing start / end of automatic performance is provided. Among the above-mentioned switches, the memory switch is a lock type switch, and the other switches are unlock type switches.

An automatic performance memory 2 stores performance information used for automatic performance. A tone generator 3 synthesizes a tone signal based on the performance information supplied via the bus 9. The synthesized tone signal is sounded through the sound system 4. 6 is a CPU,
The other components are controlled based on the control program stored in the program memory 7. 5 is a timer, C
A timer interrupt is generated for the PU 6 at every predetermined time.
A working memory 8 is used for various processes in the CPU 6.

B. Operation of the embodiment B-1. Overall Operation of the Embodiment Next, the operation of this embodiment will be described. First, when the power of the electronic musical instrument of this embodiment is turned on, the main routine shown in FIGS. 3 and 4 is started. When the process is started in these figures, first, in step SP1, a predetermined initialization is performed. Next, the processing of steps SP2 to SP19 is repeatedly executed, but steps SP2, 5,
At 8, 11, and 13, the operating states of predetermined switches (quantize value setting switch, quantizing strength setting switch, swing rate setting switch, memory switch, and RUN switch) in the operator group 1 are determined.
That is, if there is an ON event of the corresponding switch when these steps are executed, it is determined as “YES” in these steps and the corresponding processing is executed. On the other hand, if there is no on event, it is determined to be "NO" and the process is not executed. Note that other processing that should be performed regardless of the operation state of the operator group 1 is step SP.
It is executed at 19. Hereinafter, each process when there is an on event of each switch will be described.

Processing for an ON event of the memory switch
In the present embodiment, the performance information stored in the automatic performance memory 2 is quantized or swung. However, as a prerequisite, it is necessary to store the performance information to be processed in the automatic performance memory 2. . This performance information may be input from the outside by a MIDI signal or the like, or may be input by the performance of a keyboard provided in the operator group 1. In the latter case, the performer turns on the memory switch provided in the operator group 1. When the memory switch is set to the ON state, it is determined as "YES" when step SP11 is executed, the process proceeds to step SP12, and the performance information memory process is performed. That is, in step SP12, new performance events (events such as keyboard and tone setting operators) in the operator group 1 are detected at any time, and performance information is generated based on these events and stored in the automatic performance memory 2. To be done. When the performer finishes inputting the performance information, he turns off the memory switch. As a result, if step SP11 is subsequently executed, it is determined to be "NO", and the process proceeds to step SP13.

On / off of the quantize value setting switch
The player performing the process can properly reproduce the performance information stored in the automatic performance memory 2 (details will be described later).
To quantize a musical sound, press the quantize value setting switch. When an ON event occurs in the quantize value setting switch, the process goes to step SP2.
When the procedure proceeds to step S, it is determined to be “YES”, and the processing is step S.
Proceed to P3. In step SP3, the player specifies the track number for which the quantize value is to be set, and the specified value is stored in the variable TRK. In the present embodiment, it is possible to provide a maximum of "16" tracks in the performance information, and each track is "0" to "1".
It corresponds to the track number "5".

Next, when the processing proceeds to step SP4, the performer sets a quantize value. The quantize value set here is any of those shown in the left column of Table 1 below. In step SP4, the clock values in the right column of Table 1 below are variable according to the set quantize value.
It is assigned to QVCL (TRK). That is, when the number of clocks of the "quarter note" is set to "96", the clock value is set according to the note length related to the quantize value.

[0015]

[Table 1]

On the quantize strength setting switch
Processing for Vent Next, when an on event occurs in the quantize strength setting switch, when the next step SP5 is executed, "Y
ES ”, the process proceeds to step SP6. At step SP6, the player specifies the track number for which the quantizing strength is to be set, and the specified value is stored in the variable TRK. Next, when the processing advances to step SP7, the quantizer strength QSTR (TRK) is set within the range of "0" to "1.0" by the performer.
FIG. 5 shows that the sounding timing of the performance information converges to the beat timing according to the quantizing strength QSTR (TRK). FIG. 10A shows beat timing when the quantize value is “quarter note” (note that the eighth note in the figure indicates sounding timing). FIG. 11C shows an example of the sounding timing of the performance information before quantizing.

Although the details will be described later, if the quantizing strength QSTR (TRK) is set to "0" here, no quantizing is performed at all. Also, quantize strength QS
When TR (TRK) is set to "1.0", the sounding timing of each performance information coincides with the closest beat timing, as shown in FIG. Also, quantize strength QST
When R (TRK) is set to "0.5", the sounding timing of each performance information converges within the range of "beat timing ± 0.25T (where T is the cycle of beat timing)". In this way, when the desired quantizing strength QSTR (TRK) is set, the process proceeds to step SP8 and step SP8.
The loop of 2 to SP19 is repeated.

On event of the swing rate setting switch
When performing a swing on the performance information, the performer presses the swing rate setting switch in the operator group 1. When the on event of the swing rate setting switch is detected, when the process proceeds to step SP8 next, it is determined as "YES", and the process proceeds to step SP9. In step SP9, the player specifies the track number for which the swing rate is to be set, and the specified value is stored in the variable TRK. Next, when the processing proceeds to step SP10,
The swing rate SFLVL (TRK) is set within the range of "0.5" to "0.75" by the performer.

Here, the swing rate SFLVL (TRK) is "0.
In the case of "5", it is equal to the case where no swing is performed, the value becomes large and the sounding timing in the weak beat is delayed, and in the case of "0.75", the sounding timing in the weak beat is equivalent to half the beat cycle. Will be delayed. In addition,
The details of the processing will be described later. In this way, when the desired swing rate SFLVL (TRK) is set, the process proceeds to step SP11 and the loop of steps SP2 to SP19 is repeated.

For the ON event of the RUN switch
The processing performer presses the RUN switch in the operator group 1 when starting or stopping the automatic performance. When the ON event of the RUN switch is detected, when the process proceeds to step SP13 next time, it is determined as "YES", and the process proceeds to step SP14. In step SP14, the value of the flag RUN is inverted. Here, flag R
UN is a flag that takes a value of "1" or "0",
"1" indicates that automatic performance is in progress,
When it is "0", it indicates that the automatic performance is stopped. The flag RUN is set to "0" in the initial setting of step SP1. Next, the process is step SP
Proceeding to 15, it is determined whether the flag RUN is "1".

When an RUN switch ON event occurs during execution of the automatic performance, the flag RUN is changed from "1" to "0" in step SP14, and thus "NO" is determined in step SP15. It In this case, the process proceeds to step SP18, and the automatic performance sound mute process is performed. That is, the note-off signal is supplied to each channel of the tone generator 3.

On the other hand, when the automatic performance is not performed, the RU
When an ON event occurs in the N switch, step SP
Since the flag RUN is changed from "0" to "1" in 14, the determination is "YES" in step SP15, and the process proceeds to step SP16. Step SP
In 16, the various variables used in the timer interrupt routine (FIGS. 7 and 8) are initialized. The details will be described later. By the way, in this embodiment, pointer variables are set for the respective tracks, and the read address in the automatic performance memory 2 is determined by the pointer variables. Next, the process is step SP
Proceeding to 17, the pointers of all tracks are set to the head address of each track. When the above processing is completed, the processing proceeds to step SP19, and steps SP2 to SP2.
The loop of SP19 is repeated.

B-2. Timer Interrupt Processing In the present embodiment, a timer interrupt is generated in the CPU 6 every "1/96" of the quarter note length, whereby the automatic performance is performed. The details will be described with reference to FIGS. 7 and 8. When the timer interrupt occurs, the process proceeds to step SPA1 in FIG. 7 and it is determined whether the flag RUN is "1". If the flag RUN is "0", it means that the automatic performance is not being performed, so that the process returns to the main routine (FIGS. 3 and 4) without performing any process. If the flag RUN is "1", it is determined to be "YES" and the process proceeds to step SPA2. In step SPA2,
The variable TR is set to "0". The variable TR indicates the track number to be processed and is referred to as "track number TR" below.

Next, when the processing proceeds to step SPA3,
It is determined whether or not the variable PTM (TR) (PTM (0) at this point) is "0" or less. By the way, step SPA3
Is executed only when the flag RUN is previously set to "1" in step SP14 (see FIG. 4).
When the flag RUN becomes "1", step SP1
In 6, the variable PTM (i) (where i = 0 to 15) is set to “0”. Therefore, the first step S
When PA3 is executed, it is determined to be "YES" and the process proceeds to step SPA4.

In step SPA4, the data designated by the pointer of the track number TR is read from the automatic performance memory 2. Here, first in step SP17
Since the pointer is set to the head for each track number when the above is executed, the first performance information of the track number "0" is read from the automatic performance memory 2 here. Next, when the process proceeds to step SPA5, the pointer is incremented to point to the next performance information.

Now, the format of the performance information in this embodiment will be described with reference to FIG. The performance information is roughly classified into pronunciation data 20, time data 30, and end data 40. The pronunciation data 20 is composed of an identification code 21, a key code 22 and a pronunciation time 23. The identification code 21 indicates that this performance information (sounding data 20) is sounding data, the key code 22 indicates a pitch to be sounded, and the sounding time 23 corresponds to this pitch. It indicates the time when pronunciation should be made.

Further, the time data 30 is the identification code 31.
And duration 32. The identification code 31 indicates that the performance information (time data 30) is time data, and the duration 32 specifies the time from the last sounding event to the next sounding event. Also, the end data 40
Indicates that the performance information of the corresponding track has ended. In the following processing, the processing branches depending on the type of performance information, so the operation will be described separately for each case.

When the processing for reading the pronunciation data proceeds to step SPA6, it is determined whether or not the performance information read in step SPA4 is end data. Assuming that the read performance information is pronunciation data, it is determined to be "NO" here, and the process proceeds to step SPA7. In step SPA7, it is determined whether or not the read performance information is time data. Also in this case, it is determined as "NO", and the processing is step SPA.
Proceed to 16. In step SPA16, the tone generation processing subroutine shown in FIG. 9 is called.

When the processing is started in FIG. 9, the content of the key code 22 is changed to the variable K in step SPA51.
Assigned to C. Next, in step SPA52, the contents of the tone generation time 23 are substituted into the variable GT (TR).
Next, when the processing advances to step SPA53, the key-on signal, the variable KC (key code) and the track number TR are supplied to the tone generator 3. In the tone generator 3, tone signals are synthesized based on these data, and the synthesized tone signals are sounded through the sound system 4.

When the above processing is completed, the processing returns to step SPA3 of the timer interrupt routine (FIGS. 7 and 8).
Here, since the variable PTM (TR) is still "0", the process proceeds to step SPA4 to read the next performance information. Then, the pointer relating to the track number TR is incremented via step SPA5. In the example shown in FIG. 2, the time data 30 is provided immediately after the sound generation data 20, but in the case of producing a chord, the time data may be provided after a plurality of sound generation data. In this case, the sounding process (step SPA16) is repeated for these plural sounding data.

Processing when time data is read If the performance information read in step SPA4 is time data, the process proceeds to step SPA8 via steps SPA6 and SPA7, and the contents of duration 32 are substituted into variable TM. It Next, when the process proceeds to step SPA9, it is determined whether or not the clock value QVCL (TR) is "0" or more. That is, when quantizing is performed, the clock value QVCL (TR) becomes a value of "1" or more (see Table 1). Therefore, in step SPA9, it is determined whether or not quantizing is performed. Here, each process according to the determination content of step SPA9 will be described.

(A) Processing When Quantization Is Not Executed When quantization is not executed, it is determined to be “NO” in step SPA9, and the processing proceeds to step SPA17. In step SPA17, the content of the variable TM (that is, the content of the duration 32) is added to the variable PTM (TR). As a result, the content of the variable PTM (TR) becomes "1" or more. Therefore, the next process is step S.
When the process proceeds to PA3, the determination is "NO" here, and therefore the process proceeds to step SPA13, and the count subroutine shown in FIG. 10 is called.

When the process is started in FIG. 10, the variable PTM (TR) is decremented by "1" in step SPA61. Therefore, the content of the variable PTM (TR) is
It becomes "TM-1". Next, when the processing advances to step SPA62, it is determined whether or not the variable GT (TR) is "0" or less. The variable GT (TR) is substituted with the content of the sound generation time 23 when the step SPA52 was previously executed. Therefore, since the variable GT (TR) is "1" or more at this point, it is determined to be "YES", and the process proceeds to step SPA64. In step SPA64, the content of variable GT (TR) is decremented by "1", and the process returns to the timer interrupt routine (FIGS. 7 and 8).

In step SPA14, the track number TR is incremented by "1". Since the track number TR is "0", it is set to "1" here. Next, when the process proceeds to step SPA3 via step SPA15, the same process as described above is performed on the track with the track number “1”. In this way, the track number is transmitted through step SPA14.
TR is incremented and track number "1" ~
The process for each track of "15" is performed. In step SPA15, the track number TR is "1.
6 "is determined, and if" YES "is determined here, the process returns to the main routine (FIGS. 3 and 4).

After that, a timer interrupt is generated at every predetermined time, but as described above, steps SPA17 and SPA1.
Track number TR =
For "0", the variable PTM (TR) is set to "TM-1". Therefore, when the track number TR = "0" and step SPA3 is executed, it is determined to be "NO",
Through the step SPA13, the variable PTM (TR) is changed to "TM-
2 ”. In this way, the variable PTM (TR) is decremented by "1" each time a timer interrupt occurs.
Then, the timer interrupt is repeated "TM" times, and the variable PTM
When (TR) becomes "0", it is determined to be "YES" in step SPA3, and the next performance information is read out via step SPA4. That is, the variable PTM (TR) is
This corresponds to the time (timer interrupt count) until the next performance information is read for the track number TR. Therefore, hereinafter, the variable PTM (TR) is called "residual duration" and the variable TM is called "duration".

(B) Processing When Performing Quantization On the other hand, when performing quantization, a “YES” determination is made in step SPA9, and the processing proceeds to step SPA10. In step SPA10, the variable QCNT1 (TR)
The duration TM is added to. Here, in the variable QCNT1 (i) (where i = 0 to 15), the step S
At P16 (see Fig. 4), each clock value QVCL (i)
Initially set to half the value of. As an example, if the quantize value is "quarter note", the clock value QVCL (TR)
Is 96, the variable QCNT1 (TR) is initialized to "48". If the duration TM is "24" (= 96/4) corresponding to "sixteenth note", the variable QCNT1 (TR) is set in step SPA10.
"72" is substituted into.

In step SPA10, the residual duration PTM (TR) is set to "TM × (1-QSTR (TR))".
Is added. As mentioned above, the residual duration PT
Since M (i) (where i = 0 to 15) is initially set to "0" in step SP16, at this point, the residual duration PTM (TR) is set to "TM x (1-QSTR (TR)
) ”Will be substituted. If the quantizing strength QSTR (TR) is "0", the processing is the same as that in step SPA17 described above. In this step SPA10, the quantizing strength Q
The value of residual duration PTM (TR) is changed according to the value of STR (TR). Example above (clock value QVCL (TR) = 9
6, duration TM = 24), and assuming that the quantizing strength QSTR (TR) is "0.5", the residual duration PTM (TR) is set to "12".

Further, in step SPA10, the quantizing strength QSTR (TR) is added to the variable QCNT2 (TR). In the variable QCNT2 (i) (where i = 0 to 15), the multiplication result of the clock value QVCL (i) and the quantizing strength QSTR (i) is substituted in step SP16.
Example above (clock value QVCL (TR) = 96, duration TM = 24, quantize strength QSTR (TR) = 0.5)
In, variables before step SPA10 are executed
The initial value of QCNT2 (TR) is “48” (= 96 × 0.5), which is the variable QCN after step SPA10 is executed.
T2 (TR) becomes “60” (= 48 + 24 × 0.5).

Next, when the processing advances to step SPA11, the variable QCNT1 (TR) is divided by the clock value QVCL (TR) at an integer level, the quotient is substituted into the variable QTT, and the remainder is substituted into the variable SRP. To be done. In the above example, the clock value QVCL (TR) is "96" and the variable
Since QCNT1 (TR) was "72", the variable QTT becomes "0" and the variable SRP becomes "72". Next, when the process proceeds to step SPA12, the variable QTT is "0".
It is determined whether or not it is greater than. According to the above example, the variable
Since QTT is "0", it is determined to be "NO", and the processing advances to step SPA13 via step SPA3.

After that, when a timer interrupt occurs, the process proceeds to step SPA13 via step SPA3 for the track number TR, and the residual duration is calculated.
The count operation is executed while PTM (TR) is decremented. After the residual duration PTM (TR) becomes "0", when the pronunciation data is read in step SPA4, the process proceeds to step SPA16 via steps SPA6 and SPA6 to produce the pronunciation process for the newly read pronunciation data. Is done. Example above (clock value QVCL (T
R) = 96, duration TM = 24, quantizing strength QSTR (TR) = 0.5), duration T
Residual duration even though M is "24"
PTM (TR) was set to "12". Explaining with reference to FIGS. 5 (c) and 5 (d), the pronunciation in which the duration is set to be pronounced at time t ₂ (time after "24" clocks have passed from time t ₀ ) in the original performance information. The data is time-quantized t by being subjected to such quantization.
It is sounded at ₁ (time after "12" clocks have passed from time t ₀ ). That is, the tone generation timing of such a musical tone converges toward time t ₀ which is the beat timing.

After the tone generation processing in step SPA16 is performed, the processing proceeds to step SPA8 via steps SPA3 to SPA7, and the next duration TM is read from the automatic performance memory 2. According to the example in FIG. 5 (c), a time t ₆ next note timing has elapsed from time t ₂ "48" clocks. That is, the duration TM read is "48". Next, when the process proceeds to step SPA10 via step SPA9, the variable QCNT1
Duration TM is added to (TR). Variable QCNT1
Since (TR) was previously set to "72" in step SPA18, it is set to "120" here. Further, the residual duration PTM (TR) has been set to "0" because the step SPA14 has been executed a plurality of times so far, and thus is set to "24" (0 + 48 × 0.5) this time. Also, the variable QCNT2 (TR) is set in the previous step S.
Since it was set to "60" when PA10 was executed,
This time, it is set to “84” (= 60 + 48 × 0.5).

Next, when the processing advances to step SPA11, the variable QCNT1 (TR) is divided by the clock value QVCL (TR), the quotient is substituted for the variable QTT, and the remainder is substituted for the variable SRP.
In the above example, the variable QCNT1 (TR) is "120", clock value Q
The variable QTT is "1" because VCL (TR) is "96".
And the variable SRP is set to "24". Next, when the process proceeds to step SPA12, the variable QTT is "1", so that "YES" is determined and the process proceeds to step SPA18. In this way, step SPA12
In the case of "YES", the beat of the original performance information changes at least once from the strong beat to the weak beat or from the weak beat to the strong beat.

In step SPA18, the variable QCN
The variable SRP (“24” in the above example) is assigned to T1 (TR). Next, when the process proceeds to step SPA24, the variable Q
It is determined whether TT is odd. The variable QTT indicates the number of changes from strong beats to weak beats or from weak beats to strong beats since the last time this step was performed. Therefore,
The fact that the variable QTT is not an odd number (that is, an even number) means that it has the same strong or weak beat as when the processing of this step was performed last time. In this case, the processing proceeds to step SPA23. On the other hand, if the variable QTT is an odd number, it means that the beat is a strong or weak beat different from the time when the process of this step was performed last time. In this case, the process proceeds to step SPA19.

When the processing advances to step SPA19, the content of the flag FSFL (TR) is inverted. The flag FSFL
(i) (however, i = 0 to 15) is set to "0" in step SP16. Therefore, this time the flag
FSFL (TR) is set to "1". Here, step S
The condition for executing PA19 is that "YES" is determined in step SPA24, that is, "beat strong / weak beat is switched". Therefore, the flag
When FSFL (TR) is “0”, the next pronunciation is a strong beat,
In the case of "1", it is understood that the next pronunciation is a weak beat.

Next, when the processing advances to step SPA20, it is judged whether or not the flag FSFL (TR) is "1", that is, whether or not the performance information is for a weak beat. In the above example, it is determined to be "YES", and the processing is step SPA2.
Go to 1. In step SPA21, the variable SFLTM (TR) is added to the residual duration PTM (TR). By the way, the variable SFLTM (i) (however, i = 0 to 15)
At step SP16, “QVCL (i) × 2 × (SFLV
L (i) -0.5) ”is the default setting. In the above example, the clock value QVCL (TR) was "96", but in step SP10 (see FIG. 3), the swing rate SFL
Assuming that VL (TR) is set to "0.5", the delay clock number SFLTM (TR) is initialized to "0".

Therefore, in the above example, step SP
Substantial processing is not performed in A21, and the processing advances to step SPA23. In step SPA23,
Clock value QV with respect to residual duration PTM (TR)
The product of CL (TR) and the quantizing strength QSTR (TR) is added. That is, the residual duration PTM (TR) is
It is set to “72” (24 + 96 × 0.5). When the above process ends, the process proceeds to step SPA3.
Return to. FIG. 5 (c), when explained with reference to (d), pronunciation duration is set to be pronounced at time t ₆ ( "48" clocks elapses after the time from the time t ₂₎ in the original performance data The data will be sounded at time t ₇ (time after “72” clocks have passed from time t ₁ ) by being subjected to such quantization. That is, the tone generation timing of such a musical sound converges toward time t ₁₀ , which is the closest beat timing.

Processing for Swing In the above example, the swing rate SFLVL (TR) is "0.
5 ", so steps SPA20-22
Was not substantially processed in step S,
In P10, when the swing ratio SFLVL (i) (where i = 0 to 15) of more than “0.5” and less than “0.75” is set, the swing effect is applied to the performance information here. It The details will be described below.

First, when the swing rate SFLVL (TR) is set and then step SP16 is executed, the variable SFLTM is changed.
(i) (however, i = 0 to 15) means “QVCL (i) × 2 × (SFLV
L (i) -0.5) ”is set. For example, swing rate SFLVL
Assuming that (TR) is set to "0.6", "QVCL
(i) × 2 × (SFLVL (i) -0.5) ”is“ 19.2 ”(96 × 2
X (0.6-0.5)). However, step SP1
In 6, the fractional part of the delay clock count SFLTM (TR) is rounded off, resulting in the delay clock count SFLTM (TR).
Will be set to "19".

Next, a timer interrupt routine (FIGS. 7 and 8)
When the process proceeds to step SPA20 in step F1, flag F
Based on SFL (TR), it is determined whether the next pronunciation is a strong beat or a weak beat. By the way, since the first beat of a musical composition is a strong beat, the first change in beat is a change from a strong beat to a weak beat. Therefore, step SPA2
In 0, it is determined to be “YES”, and the processing is step SP.
Go to A21. In step SPA21, with respect to the residual duration PTM (TR), the delay clock number SFLT
M (TR) is added. As a result, the sounding timing of the weak beat is delayed by a time corresponding to the delay clock number SFLTM (TR).

Next, when the beat changes from a weak beat to a strong beat, it is judged "YES" again in step SPA12, and the flag FSFL (T
R) is set to "0". Therefore, the process is step S
Proceed to step SPA22 via PA20. In step SPA22, the delay clock number SFLTM (TR) is subtracted from the residual duration PTM (TR). That is, the delay clock number SFLTM (TR) was added to the residual duration PTM (TR) when step SPA21 was executed earlier, but this time it is subtracted, so the musical tone is generated at the original timing. It turns out that In this way, by the processing of steps SPA20 to 22, in the strong beat, the original sounding timing is obtained, and in the weak beat, the timing delayed by the delay clock number SFLTM (TR),
Pronunciation processing is performed.

As described above, in this embodiment, it is possible to appropriately apply the "quantize" or "swing" effect to the musical sound, and by using both of them together, a more interesting automatic performance can be performed. . By the way, it is common for a performer to set parameters such as quantize strength and swing rate at the beginning to perform an automatic performance, but during the performance, these parameters are unsuitable and the performance becomes unnatural. It may be possible. In such a case, various parameters may be changed by appropriately operating the corresponding operators in the operator group 1. When a parameter is changed, the next time a timer interrupt occurs, various processing is performed based on the changed parameter, so the performer can quickly judge whether the changed parameter is right or wrong. Is. Furthermore, in the present embodiment, various parameters relating to "quantize" and "swing" can be set for each track of the performance information, so that extremely precise tone control can be performed.

C. Modifications The present invention is not limited to the embodiments described above, and various modifications can be made, for example, as follows. In the above embodiment, duration type performance information (information consisting of sound data, time data, etc. as shown in FIG. 2) was used, but event type performance information (note-on data,
Information including time data, note-off data, etc.) may be used. In the above embodiment, the processing of step SPA23 was performed after the processing of steps SPA20-22. That is, when the "quantize" and the "swing" are used together, the "quantize" is performed after the "swing". However, it is also possible to execute step SPA23 immediately after step SPA19, perform “quantize”, and then perform “swing”.

[0053]

As described above, the automatic musical instrument according to claim 1 can quantize the musical sound in real time, and the automatic musical instrument according to claim 2 can swing the musical sound in real time. It is possible to apply. Further, in any case, the calculating means calculates the sounding timing according to the track, so that the sounding timing of the musical sound can be precisely controlled in real time.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an example.

FIG. 2 is a diagram showing a format of performance information in one embodiment.

FIG. 3 is a flowchart of a main routine of one embodiment.

FIG. 4 is a flowchart of a main routine of one embodiment.

FIG. 5 is a time chart of an example.

FIG. 6 is an explanatory diagram of a swing effect.

FIG. 7 is a flowchart of a timer interrupt routine according to an embodiment.

FIG. 8 is a flowchart of a timer interrupt routine according to an embodiment.

FIG. 9 is a flowchart of a pronunciation subroutine of one embodiment.

FIG. 10 is a flowchart of a count subroutine of one embodiment.

FIG. 11 is an explanatory diagram of a quantizing effect.

FIG. 12 is a diagram for explaining the operation of a conventional automatic musical instrument.

[Explanation of symbols]

 2 Automatic performance memory (performance information storage means) 3 Tone generator (reproduction means) 6 CPU (reading means, calculation means, reproduction means)

Claims (2)

[Claims] 1. A performance information storage means for storing performance information of a plurality of tracks each including musical tone generation timing data, a reading means for reading performance information from the performance information storage means, and a performance read by the reading means. After the sounding timing is calculated, a calculating means for calculating the sounding timing so that the sounding timing of the information is compared with the corresponding tone generation timing data and converges to the reference beat timing of the corresponding track. An automatic performance device comprising: reproduction means for reproducing performance information corresponding to the sounding timing when a time corresponding to the sounding timing has elapsed. 2. A performance information storage means for storing performance information of a plurality of tracks each including musical tone generation timing data, a reading means for reading performance information from the performance information storage means, and a performance read by the reading means. When the musical tone generation timing data of the information belongs to a predetermined time range of a predetermined track, a calculating means for calculating the sounding timing so that the reproduction time is delayed, and the sounding timing after the sounding timing is calculated. An automatic performance device comprising a reproduction means for reproducing performance information corresponding to the sounding timing when time passes.