JPH0664467B2 - Electronic musical instrument - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic musical instrument, and more particularly to playing based on an operation amount of a performance operator such as a string or a drum pad, that is, a vibration amount indicating strength, speed, depth or the like. The present invention can be applied to an electronic musical instrument such as an electronic percussion instrument that detects that an operation has been performed.

[Outline of Invention]

The present invention, in an electronic musical instrument adapted to detect a performance operation based on the vibration of a performance operator, is based on a comparison result between performance operation detection information and self-vibration detection data obtained at the present time. By controlling the generation of musical tones, it is possible to effectively prevent the possibility of generating unnatural musical tones due to abnormal self-vibration.

[Conventional technology]

As shown in FIG. 23, this type of electronic percussion instrument 1 is a pad PA as a plurality of, for example, six performance operators 3 on a board 2.
When D1 to PAD6 are arranged and the player strikes each pad PAD1 to PAD6 with a stick, for example, each pad PAD1 to PAD6
The performance operation information such as the strength, speed, and depth of hitting is generated based on the vibration generated in accordance with the performance operation amount, and the performance operation information is converted into electrical performance operation detection information, and the pads PAD1 ~ It has been proposed that the percussion instrument sound assigned to the PAD 6 is sounded according to the sounding conditions set by various setting operators of the setting operation unit 4.

For example, as percussion instrument sounds, pad PAD1 is a bass drum, pad PAD2 is a cymbal, pad PAD3 is a tom, pad PAD4 is a snare drum, and pad PAD5 is a hi-hat open, pad.
You can assign a hi-hat closed percussion sound to PAD6.

[Problems to be solved by the invention]

In this way, when the performance operator 3 to be operated by the performer is one having a relatively large operation surface such as the pads PAD1 to PAD6, the performer operates the pads PAD1 to PAD6.
When one of D6, for example, pad PAD1 is strongly hit, the pad PAD1 vibrates (this is called "self-vibration")
As a result, the operation information such as the operation strength can be input, and at the same time, the other pads PAD2 to PAD6 arranged on the board 2 vibrate under the influence of the vibration of the pad PAD1. ").

By the way, the pad PAD1 ~
In order to accurately detect the playing operation state of the PAD6, the envelope signal J2 shown in FIG. 24 (B) is formed based on the vibration detection signal J1 shown in FIG. 24 (A), and the envelope signal J2 is formed. When the peak occurs, it is determined that the performance operation has been performed, and as shown in FIG. 24 (C), the performance operation information acquisition signal J3
A method of causing

However, when the performance operation detection information is obtained by such a method and the pads PAD1 to PAD6 are struck abnormally strongly, as shown in FIGS. 24 (A) to (C), vibration is generated. Abnormal vibration detection signal J1X that causes the detection signal J1 to have only one peak and does not have a waveform that naturally decays, but to generate the second and third peaks following the first peak.
Can occur.

In such a case, an abnormal musical sound is generated each time the performance operation information acquisition signal J3X is generated each time an abnormal vibration detection signal J1X peaks, which may cause the musical sound to be unnatural as a whole. There is.

The present invention has been made in consideration of the above points, and proposes an electronic musical instrument capable of surely discriminating the abnormal self-vibration when the abnormal self-vibration occurs so as not to generate an unnatural musical tone. It is what

[Means for Solving the Problems] In order to solve the above problems, the present invention has a performance operator (3) and generates a musical tone corresponding to the operation amount of the performance operator (3). In electronic musical instruments, performance operation detection information (ADVAL) corresponding to the operation amount of the performance operator (3)
Performance operation detection means (25, 14, 31, LP1) forming
When the performance operation detection information (ADVAL) of the performance operator (3) is obtained at the present time, the previous operation obtained by the previous performance operation with the same operator (3) as the performance operator (3) is obtained. Elapsed time (LAP) from the time when the performance operation amount (ADATA (PADNO)) based on the performance operation detection information (ADVAL) and the previous performance operation detection information (ADVAL) were obtained to the current time point.
Self vibration detection data (DC) determined by CTR (PADNO)
VAL), and the performance operation detection information (A
DVAL) is compared with self-vibration detection data (DCVAL), and based on the comparison result, sound generation control means (31, 34A, LP4) which forms sound generation information when the previous performance operation detection information is sound-accepted. ) And a musical tone generating means (35, 36) for producing a musical tone in response to the pronunciation information.

[Action]

Performance operation detection information at the present point in time _{(t 0)} (PADNO, A
DVAL) is obtained, this is used as self-vibration detection data (DC
VAL) to determine whether or not to sound at the current time (t ₀ ), abnormal vibration occurs due to the previous performance operation, and the current time (t ₀ ) Performance operation detection information (PADNO, ADV)
AL) is obtained, self-vibration detection data (DCVA
It is possible to form pronunciation information containing the content that should not be pronounced in comparison with L).

Therefore, it is possible to prevent an unnatural percussion instrument sound from being generated when an abnormal vibration occurs.

〔Example〕

 An embodiment of the present invention will be described in detail with reference to the drawings.

[1] Configuration of First Embodiment FIG. 2 shows an embodiment in which the present invention is applied to an electronic musical instrument consisting of an electronic percussion instrument. An electronic musical instrument 11 is a performance of the electronic percussion instrument 1 described above with reference to FIG. It has a performance operation section 12 having the same structure as the operation section (corresponding parts to those in FIG. 23 are designated by the same reference numerals).

Pad PAs as performance operators 3 arranged on the board 2
When D1 to PAD6 are operated by the stick, the performance operation information signal S1 is output to the multiplexer 13.

The multiplexer 13 supplies a pad number signal S2, which represents the pad number of the performance-operated information among the performance operation information signals S1, to the pad number variation section 14A of the latch register 14,
The performance operation amount signal S3 indicating the strength, speed, depth, etc. of the performance operation for the pads PAD1 to PAD6 is passed through the analog / digital conversion circuit 15 as the performance operation amount data S4 in the latch register.
14 peak values are given to the latch section 14B.

Here, as shown in FIG. 3, the pads PAD1 to PAD6 convert the vibration of the pad vibrating body 21 into a vibration signal S11 (FIG. 4 (A)) by a vibration detector 22 such as a piezoelectric sensor or conductive rubber. The vibration signal S11 is converted and the diode 23 and the capacitor 2
It is configured to be smoothed by 4 and converted into a vibration envelope signal S12 (FIG. 4 (B)), which is sent as performance operation information S1 (FIG. 2).

The performance operation amount data S4 sent from the analog / digital conversion circuit 15 is given to the peak detection circuit 25, and when the vibration envelope signal S12 (FIG. 4 (B)) becomes the peak value PEAK, the peak interrupt signal S13. Is generated as detection timing information, and this is given to the latch register 14 as a latch signal. At this time, the latch register 14 latches the performance operation amount data S4 to the peak value latch section 14B, and at the same time, outputs the pad number signal S2 to the pad number variation section 14B.
Latch to A.

In this way, the pad number variation data S latched in the pad number variation portion 14A and the peak value latch portion 14B is used.
14 and peak value latch data S15 and peak detection circuit 25
The central processing unit (CPU) 31 executes the program of the ROM-configured program / table data memory 32, and the peak interrupt signal S13 sent from the CPU is taken into the RAM-configured data / working memory 34 via the bus 33. At the same time, the arithmetic processing is performed together with the table data in the program / table data memory 32.

The CPU 31 supplies the processed data via the bus 33 to the musical tone signal generation unit (TG) 35 serving as a rhythm sound source as rhythm sound generation information S31. The musical tone signal generator (TG) 35 sequentially receives the rhythm sound generation information S31 about the maximum number of tones of the percussion instrument (six tones in this case) for the time slots assigned to the first to sixth pronunciation channels, and the corresponding rhythm. A musical tone signal S32 representing a musical instrument sound is sent to the sound system 36, and thus the musical tone signal S32 is converted into a percussion instrument sound in the sound system 36.

The CPU 31 uses the pads PAD1 to PAD as the performance operators 3 by the performer.
When one of the AD6 is operated at the current time t ₀ , the data based on the performance operation information signal S1 is transferred to the register 34A (FIG. 5) of the data / working memory 34 every time the peak interrupt signal S13 is generated. By executing the pad-on interrupt routine RT1 shown in FIG. 6, it is fetched and processed.

Further, the elapsed time of the past performance state is judged by the CPU 31 executing the timer interrupt routine RT2 of FIG. 7 every time the interrupt timer 37 generates the timer interrupt signal S34.

In the case of this embodiment, the CPU 31 sets the past performance states of the pads PAD1 to PAD6 into three states, that is, the status "0",
It is classified into "1" and "2", and as shown in FIG.
Each pad PAD1, PAD is stored in the pad status register REG11 which constitutes the register 34A of the data / working memory 34.
2, ..., Pad status corresponding to PAD6 PDKON (i)
(I = 1, 2, ..., 6) is held.

Status state of "0", as shown in FIG. 8 (A1), from the present time t ₀ a first reference time TR1 (for example, 15 [m
s]), the i-th pad has not been operated until time t _{(-15) in the} past. In this state, the operation is performed for the first time at the current time t ₀ as indicated by the symbol X1. Represents the state of

In this status "0", a sounding operation can be performed any time a performance operation is performed. At this time, the CPU 31 uses the elapsed time data LAPCTR (i) (i = 1, 2,
.., 6), as shown in FIG. 8 (A2), the numerical value 15 representing the first reference time TR1 is preset, and then 1 [m
s], the elapsed time data LAPCTR (i) (i = 1, 2,
The first reference time TR1 is measured by subtracting -1 from 6).

This first reference time TR1 is, as shown in FIG. 4 (A),
From the time when the pads PAD1 to PAD6 are operated and the vibration starts, a sufficient time (for example, 15 [m
s]).

The state of the status "1", as shown in FIG. 8 (B1), based on the present time t ₀ a first reference time TR1 before time t _(-15) other than time t = t _{(ST1 )} , The current time t ₀ is reached after the first performance operation is performed as indicated by the symbol X1.
In the figure, as shown by the symbol X2, the second performance operation is performed.

In this state, at the time point t = t _(ST1) , the pad which is vibrated by the first performance operation is vibrated without finishing the vibration state, and the second performance operation is performed. This means that, for example, when the player holds the player's hands in both hands and hits repeatedly for a short time interval.

In such a case, it takes a certain amount of time or more before the stick hitting the pad is once released from the pad surface and hit the pad surface again. Generating a percussion instrument sound may be considered a malfunction.

Therefore, in such a state, as shown in FIG. 8 (B2), as the elapsed time data LAPCTR (i), the second reference time TR2 from the current time t ₀ when the second performance operation is performed.
(In this embodiment, 30 [ms] is selected) is timed, and during the second reference time TR2, the musical tone generator (TG) 35 is controlled not to generate a sound as a pause period. .

Furthermore, the status "2" is as shown in Fig. 8 (C1).
At the time points t = t _(ST1) and t = t _(ST2) other than the time point t _(-15) before the first reference time TR1 with respect to the current time point t ₀ , the same pad is used as indicated by symbols X1 and X2. After being struck twice twice, at the current time t ₀ , as shown by the symbol X1, 3
When the same vibration as the performance operation of the second time occurs, the elapsed time LAPCTR (i) of the elapsed time register REG12 is included in the timing operation of the second reference time TR2 (see FIG. 8 (C2 )), The CPU 31 controls so that the sounding operation is not performed.

[2] Principle processing procedure In the embodiment shown in FIG. 2, the CPU 31 produces a sound based on the performance operation detection information obtained at the present time t ₀ according to the principle processing procedure as shown in FIG. Determine whether or not.

That is, the CPU 31 controls the pad PAD1 that constitutes the performance operator 3.
When the peak interrupt signal S13 is generated from one of the PAD6 to PAD6, the sounding confirmation processing routine is entered from the processing step PR1 and a peak is generated in the processing step PR2, and the performance operation detection information (that is, the patch) from the pad (that is, the relevant pad). Donan variation data S14 and peak value latch data S15) are loaded.

Subsequently, the CPU 31 determines in the processing step PR3 whether or not the pad has been sounded within a predetermined reference time, that is, the first reference time TR1, and when a positive result is obtained, moves to the processing step PR4 and proceeds to the previous step. From the time of pronunciation to the current time t
_The self-vibration detection data is obtained according to the elapsed time until ₀ .

Here, the self-vibration detection data is the current time t ₀ , considering the performance operation amount of the previous performance operation and the elapsed time to the present.
The playing operation amount of is not based on the self-vibration generated based on the new playing operation, and as described above with reference to FIG. 24, the envelope is not normal during the damping process of the vibration because the previous playing operation is abnormal. It is pre-selected as a value at which it can be determined that a peak interrupt has occurred due to a variation that deviates from the attenuation curve.

Here, the vibration of the pad is the vibration detection signal J1 shown in FIG.
As shown by, the self-vibration detection data for determining whether or not an abnormal vibration has occurred is based on the fact that the vibration decays smoothly over time. It is selected so as to draw a curve that decreases according to the elapsed time up to t ₀ .

In the subsequent processing step PR5, the CPU 31 determines whether or not the performance operation amount at the present time is larger than the self-vibration detection data.

If an affirmative result is obtained here, this means that the value of the currently detected performance operation amount is larger than the self-vibration detection data and is therefore based on the normal performance operation. After moving to processing step PR6 to generate a musical tone, the processing ends the pronunciation confirmation processing routine in processing step PR7.

On the other hand, when a negative result is obtained in the processing step PR5, this means that the current play operation amount is not generated based on the normal play operation,
At this time, the CPU 31 jumps to the tone generation processing of step PR6 and ends the tone generation confirmation processing routine in processing step PR7.

If a negative result is obtained in the above-mentioned processing step PR3, this means that the time from the previous performance operation to the current time t ₀ is longer than the reference time TR1, and therefore abnormal vibration monitoring based on the previous performance operation is performed. Means that it is not necessary to perform (in other words, it means that it may be determined that it is based on a regular performance operation.

Therefore, at this time, the CPU 31 jumps to the processing step PR6 without performing the pronunciation confirmation processing of the processing steps PR4 and PR5, and executes the pronunciation processing based on the performance operation detection information.

The principle processing procedure shown in FIG. 1 is more specifically expressed in the pad-on interrupt routine RT1 of FIG.

[3] Timer interrupt processing The timing operation of the elapsed time data LAPCTR (i) described above with reference to FIGS. 8 (A2), (B2), and (C2) is performed every time the interrupt timer 37 generates the timer interrupt signal S34.
This is done by the PU 31 executing the timer interrupt routine RT2 (FIG. 7).

When the timer interrupt signal S34 is generated, the CPU 31 sends the pad number working register REG10 at step SP1.
Numerical data 1 is written in (Fig. 5) as pad number working data PN.

This pad number working data PN is data used for processing the first to sixth sounding channels one by one, and the CPU 31 sends the PN in the next step SP2.
= 1 Pronunciation elapsed time data LAPCTR (PN)
It is determined whether (PN = 1) is 0 or not. If a negative result is obtained here, it means that the timer operation of the first or second reference time TR1 or TR2 is being performed for the first pad PAD1 at the time when the timer interrupt signal S34 is generated. .

At this time, the CPU 31 shifts to step SP3 and the elapsed time data LA
After subtracting "-1" from PCTR (PN), it is determined in step SP4 whether or not the elapsed time data LAPCTR (PN) after subtraction of "-1" has become zero.

If a negative result is obtained here, this means that it is necessary to continue the timing operation. At this time, the CPU 31 shifts to step SP5 and adds "+1" to the pad number working data PN. Now you can specify the second pronunciation channel.

On the other hand, if a positive result is obtained in step SP2 described above, this means that the first or second reference time TR1 or T
This means that the timekeeping operation for R2 has already ended, and at this time the CPU 31 jumps to steps SP3 and SP4 and moves to step SP5.

If a positive result is obtained in step SP4 described above, this means that the timing operation of the first or second reference time TR1 or TR2 has ended at the current time t ₀ when the timer interrupt signal S34 is obtained. And then the CPU
31 is the pad status data PDKON in step SP6
After clearing (PN) to 0, move to step SP5.

In step SP5, the CPU 31 specifies the next pronunciation channel by adding "+1" to the pad number working data PN, and then in step SP7, the pad number working data PN after the addition "+1" is the maximum number of pronunciation channels. Judgment whether or not it exceeds.

If a negative result is obtained here, this means that the processing for all the pronunciation channels has not been completed, and at this time, the CPU 31 returns to the above-mentioned step SP2 and the new pronunciation channel, that is, PN = 2. The processing of the elapsed time data LAPCTR (PN) (PN = 2) is executed for.

Similarly, when the processing of the elapsed time data LAPCTR (PN) for the third to sixth pronunciation channels is completed,
If a positive result is obtained at step SP7, C
PU31 returns to the main routine from step SP8.

In this way, the CPU 31 executes the timekeeping operation of the first or second reference time TR1 or TR2 for the first to sixth sound generation channels as needed at predetermined time intervals, that is, every 1 [ms].

[4] Pad-on-interrupt processing When the performer performs a performance operation on one of the pads PAD1 to PAD6 of the performance operation unit 12, the CPU 31 determines, based on the performance operation information signal S1 obtained from the performance-operated pad. Each time the peak detection circuit 25 obtains the peak interrupt signal S13, the pad-on interrupt routine of FIG.
Execute the processing of RT1.

That is, the CPU 31 first enters the performance operation information acquisition processing loop LP1 and at step SP11 the pad number variation section 14A.
The pad number variation data S14 is fetched from (FIG. 2) and is used as pad number data PADNO representing the operated pad number data.
Figure).

Then, the CPU 31 determines the peak value latch unit 1 at step SP12.
The peak value latch data S15 is fetched from 4B (Fig. 2) and the peak level data ADVA is stored in the peak level register REG1.
Write as L.

Thus, the CPU 31 causes the register 34A of the data / working memory 34 to store the data representing the number of the pad operated at the present time point and the performance operation amount thereof as basic performance operation information, and the succeeding immediately preceding operation pad detection processing loop.
Enter LP2.

This processing loop is a processing loop for recognizing a past performance state at the current time t ₀ in FIG. 8 and determining whether or not to sound based on the performance processing information about the pad operated at the current time t ₀ . Then, CPU31 first step SP1
In the pronunciation channels other than the pronunciation channels to which the pad of the pad number data PADNO fetched in the pad number register REG2 in 3 is assigned, the pad status data PDKON (i) is status "1" or "2". Moreover, after detecting the pronunciation channel with the minimum value of the elapsed time data LAPCTR (i) (that is, the channel that was pronounced immediately before), the elapsed time data LAPCTR (i) of the relevant pronunciation channel is written to the previous operation pad time interval data register REG4. At the same time, the channel number data PADNO of the sounding channel is written in the immediately previous operation pad number register REG5 as the immediately previous operation pad number data MINPD.

The processing of step SP13 can be executed by using the immediately preceding operation pad detection subroutine shown in FIG.

That is, the CPU 31 sets the value 1 as the pad number working data PN in step SP13A, sets the value 30 as the immediately preceding operation pad time interval data MINLAP, and further sets the value 0 as the immediately preceding operation pad number data MINPD. Pad number data PADNO in step SP13B
Is not PN (= 1), and the contents of the pad status data PDKON (PN) are confirmed in step SP13C.

Here, if the pad status data PDKON (PN) is the status "0", the CPU 31 directly moves to step SP13D to add "+1" to the pad number working data PN.

On the other hand, if it is confirmed in step SP13C that the pad status data PDKON (PN) is the status "1", the CPU 31 determines in step SP13E that 15-LAPCTR.
(PN) is executed and the calculation result is written to the elapsed time detection working register REG9 as the elapsed time detection working data LAP. Then, in step SP13F, this elapsed time detection working data is used as the immediately previous operation pad time interval data. Make sure it is smaller than MINLAP
Working data LAP for detecting the elapsed time in SP13G
Is stored as the immediately previous operation pad time interval data MINLAP, and the pad number working data PN is held as the immediately previous operation pad number data MINPD.

On the other hand, in step SP13F, the elapsed time detection working data LAP is the previous operation pad time interval data MIN.
If it is smaller than LAP, jump to step SP13D and jump to step SP13D.

On the other hand, when it is confirmed that the pad status data PDKON (PN) is the status "2" in the above-mentioned step SP13C, the CPU 31 determines in step SP13H 30-
The calculation of LAPCTR (PN) is performed and this is held as working data LAP for elapsed time detection.

Thus CPU31 previous state status than the current time point t ₀ "0" (Figure 8 (A1) and (A2)), the status "1" (Figure 8 (B1) and (B2)), or status "2 "(C1) and (C2) in Fig. 8", and when the status is "1" or "2", set the immediately previous operation pad number data MINPD and the immediately previous operation pad time interval data MINLAP. Hold in register 34A.

When the pad number working data PN is the relevant pad in step SP13B, the CPU 31 is not another channel, so the control is immediately jumped to step SP13D.

After confirming in step SP13I that this processing has been executed for all the pronunciation channels, the CPU 31 returns from step SP13J to the pad-on interrupt routine RT1.

At this time, the CPU 31 proceeds to step SP14 and determines whether or not the immediately previous operation pad number data MINPD is not zero. If an affirmative result is obtained here, this means that the first or second reference time TR1 or TR2 has elapsed in the pad with the minimum elapsed time from the time of the previous operation among the pads other than the one currently operated. It means not doing.

In this state, the CPU 31 proceeds to step SP15 and determines that the immediately preceding operation pad time interval data MINLAP is smaller than or equal to the third reference time TR3 (selected in this embodiment 5 [ms]). .

This third reference time TR3 is the third reference time TR3 (that is, 5 [m
s]) It means that the pad that is currently operated is in a state where there is a risk of jumping and causing vibrations because the above has not elapsed.

At this time, the CPU 31 shifts to step SP16, and based on the pad number data PADNO and the immediately preceding operation pad number data MINPD, the jump vibration detection coefficient stored in the table memory 32A (FIG. 10) of the program / table data memory 32. data table jumped from KTABLE vibration detection S data _{K 1} 2 _{~K 6 5} reads the register 34A (fifth
Write as jumping vibration detection coefficient data KDATA in the jumping vibration detection coefficient register REG3 in the figure).

The jump vibration detection coefficient data KDATA is obtained from the pad operated immediately before the pad (represented by the pad number data PADNO) (represented by the immediately previous operation pad number data MINPD). Represents the degree of influence of the jumping vibration of the., For example, the coefficient value is determined by the distance between the pad operated at the current time t ₀ and the pad operated immediately before.

If the coefficient value determined in this way is expressed by K _{(PADNO) (MINPD)} , the jump vibration detection coefficient table KTABLE is the 11th.
As shown in FIG. 12, for example, when the first pad PAD1 is operated at the present time t ₀ , the pad operated immediately before is sequentially PAD2, PAD3, PAD.
When 4, PAD5 and PAD6, each coefficient value K _{(PADNO) (MINPD)} is
It can be represented by K _{1 2} , K _{1 3} , K _{1 4} , K _{1 5} and K _{1 6} , the values of which are the second and fourth for the first pad PAD1.
K because the pads PAD2 and PAD4 are almost the same distance
_{Since 1 2} = K _{1 4} = 1/8 and the fifth pad PAD5 is at a slightly distant distance, K _{1 5} = 1/10 is set and the third and sixth pads PAD3 and PAD6 are set. Is the farthest distance, so K _{1 3} = K _{1 6} = 1/16 is set.

Thus, in a state where other pads PAD2~PAD6 other than the pads are oscillating by being operated immediately before, the pads PA in which the vibration has been operated at the present time t ₀
By weighting the degree of impact of jumping vibrations on D1, the conditions can be optimized again.

At the subsequent step SP17, the CPU 31 determines whether the vibration of the pad is larger than the jump vibration detection coefficient data KDATA using the jump vibration detection coefficient data KDATA.

Here, the current vibration state uses the ratio of the peak level data ADVAL held in the peak level register REG1 to the sounding level data DATA (MINPD) held in the immediately preceding operation pad number register REG5.

Incidentally, the peak level data ADVA held in the peak level register REG1 at the status "1" or "2".
L is the peak level data obtained from the pad operated at the current time t ₀ , while the sounding level data ADATA (i) (i = 1, 2, .., 6) is the peak level data (represented as ADATA (MINPD)) for the pad that was just played. Therefore the ratio
ADVAL / ADATA (MINPD) is the value from the pad operated immediately before when the pad whose peak is detected at the current time t ₀ is detected even though the peak is not actually operated. It can be considered that a peak is generated due to the jumping vibration of the pad. The peak level of the pad at which the peak occurred at this time is the peak level of the vibration of the pad operated immediately before × K _{(PADNO) (MINPD)} . It is considered to be below the value. This is because the jump vibration detection coefficient data K _{(PADNO) (MINPD)} represents the reference value of the damping ratio of the vibration from the pad operated immediately before to the pad where the current peak is detected.

On the other hand, the ratio ADVAL / ADATA (MINPD) is the magnitude of the vibration level when the pad that detected the peak is actually operated, and therefore the peak level data ADVA.
Since the value of L becomes a sufficiently large value, it becomes a value larger than the value of the jump vibration detection coefficient data KDATA.

Thus, when a negative result is obtained in step SP17, this means that the peak value may be due to the jump vibration, and at this time, the CPU 31 shifts to the jump vibration processing loop LP3.

On the other hand, when a positive result is obtained in step SP17, this means that the occurrence of the peak is due to the performance operation of the pad, and at this time, the CPU 31 shifts to the tone generation processing loop LP4.

When a negative result is obtained in step SP14 described above, this means that all the pads other than the pad in which the peak has occurred have the first reference time (that is, 15 [ms]) from the start of sounding. It means that the above has passed, and at this time, it is not necessary to execute the process for the jump vibration, so the CPU 31 executes steps SP15, SP16, SP1.
Immediately enter the tone generation processing loop LP4 without performing the processing of 7.

If a negative result is obtained in step SP15 described above, this means that the third reference time TR3 (that is, 5 [ms]) or more has elapsed from the time when the last operated pad started sounding. However, at this time, there is no risk of jumping vibration, so the CPU31 uses the step SP1
6. Directly enter the tone generation processing loop LP4 without processing SP17.

When the CPU 31 enters the tone generation processing loop LP4, it confirms the past operation state based on the pad status data PDKON (PADNO) of the pad number data PADNO peak-detected in step SP21.

When the pad status data PDKON (PADNO) is the status "0", this means that it is shown in FIGS. 8 (A1) and (A2).
As mentioned above, it means that the CPU 31 is ready to pronounce at any time.
Then, the data of the pad status data PDKON (PADNO) is rewritten from the status "0" to the status "1", and secondly the first reference time TR1 (that is, 15 [ms]) to the elapsed time data LAPCTR (PADNO). Data of
Thirdly, the peak level data ADVAL of the peak level register REG1 is transferred and held to the sounding level data ADATA (PADNO).

Thus, the CPU 31 switches to the control condition of the status "1" described above with reference to FIGS. 8 (B1) and (B2),
Move to next step SP23.

In this step SP23, the CPU 31 causes the tone signal generator (T
G) 35 transmits the pad number data PADNO of the pad number register REG2 and the peak level data ADVAL of the peak level register REG1 together with the key-on signal KON as rhythm-on occurrence information S31.

At this time, the musical tone signal generator (TG) 35 is set to the pad number data.
The musical tone signal S32 for generating the percussion instrument sound designated by PADNO at the volume designated by the peak level data ADVAL is sent to the sound system 36.

Thus as described above for Figure 8 (A1) and (A2), after generating the percussion sound corresponding to the pads the performer has operated at the present time t _0, the CPU31 step S
Return to the main routine from P24.

On the other hand, when the CPU 31 determines in step SP21 that the status is "1", this means that as described above with reference to FIGS. 8 (B1) and 8 (B2), the present time t
_This means that the same pad has been hit twice in a state in which there is a pad operated immediately before during a time period before the first reference time TR1 with reference to ₀ .

At this time, the CPU 31 shifts to step SP25 and outputs the sound level data ADATA (PADNO) and elapsed time data 15-LAPCTR (PA
DNO) self-vibration detection coefficient data table REG2 stored in the program / table data memory 32
2 (Fig. 10) to DCTBL (Fig. 13)
Read the figure and write it as self-vibration detection data DCVAL in self-vibration detection data register REG8.

Here, 15-LAPCTR (PADNO) represents the elapsed time from the operation of the pad to the present time, and is the first reference time TR.
A group of data groups DCTBL in which 1 (15 [ms] in this embodiment) is set as self-vibration detection coefficient data DV _{(ADVAL) (15-LAPCTR)} at predetermined time intervals (for example, 1 [ms] interval). (3
It is stored as 0) to DCTBL (10) (Fig. 14).

DCTBL (30) to DCTBL (10) of each data group are prepared for each operation intensity for the pad (that is, the peak level data ADVAL values 30, 29, ..., 10 at the elapsed time 15-LAPCTR = 0). It is constructed by data that draws a damping curve that corresponds to the damping curve of the vibration of the pad.

Note that the vibration of the pad is the first reference time TR after the performance operation.
While 1 decays as described above with reference to FIG. 4,
The attenuation curve draws different attenuation curves depending on the operation strength with respect to the pad (that is, the strength when the pad is hit by the stick). Each data group DCTBL (30) to DCTBL (10) of the self-vibration detection coefficient data table DCTBL is selected to be a value smaller by a predetermined amount than the value at each elapsed time point on the damping curve.

Thus, the self-vibration detection coefficient data DV _{(ADVAL) ( 15-LAPCTR)} , the vibration level of the pad in the natural damping state is always smaller than that in the corresponding elapsed time data.

Thus, the data group DCT determined from the self vibration detection coefficient data table DCTBL based on the peak level data ADVAL.
When the data DV _{(ADVAL) (15-LAPCTR)} at the corresponding elapsed time (15-LAPCTR) of BL (ADVAL) is read and compared with the peak level data ADVAL, if the peak level data ADVAL is large, the vibration of the pad is 2 It is possible to determine that it is self-vibration due to the second performance operation.

Therefore, if a positive result is obtained in step SP26,
This means that the second performance operation has been performed as described above with reference to FIGS. 8 (B1) and (B2). At this time, the CPU 31 moves to step SP27 and firstly, the pad status data PDKON (PADN
The status "1" that was written as O) is rewritten to the status "2", and secondly the elapsed time data LAPCTR
Write the value 30 representing the second reference time TR2 to (PADNO),
Thirdly, the peak level data ADVAL currently held in the peak level register REG1 is transferred and stored as the sounding level data ADATA (PADNO).

Thus, the CPU 31 switches the control condition of status "1" described above with reference to FIGS. 8 (B1) and (B2) to the control condition of status "2" shown in FIGS. 8 (C1) and (C2), and thereafter. Move to step SP28, pad number data PADNO of pad number REG2 and peak level register REG1.
The peak level data ADVAL of is sent to the musical tone signal generator (TG) 35 as the rhythm-on occurrence information S31 together with the key-on signal KON.

At this time, the musical tone signal generator (TG) 35 is set to the pad number data.
After the percussion instrument sound of PADNO is supplied to the sound system 36 so as to generate the tone level S of the peak level data ADVAL, the procedure returns from step SP29 to the main routine.

On the other hand, if a negative result is obtained in step SP26, this means that the detected peak value is lower than that in the case where the playing operation is performed normally, for example, abnormal vibration occurs because the first playing operation is abnormally strong. Means that At this time, the CPU 31 immediately proceeds to step SP29 without performing the processing of steps SP27 and SP28.

Further, when it is judged in the above-mentioned step SP21 that the pad status data is PDKON (PADNO) being the status "2", this means that the present time t as described above with reference to FIGS. 8 (C1) and (C2). ₀ are the two performance operation continues for the same pads during the past first reference time TR1 from, yet present time t ₀ from the second performance operation the second reference time TR2 (= 30 [ms ]), The occurrence of the peak at the present time t ₀ is abnormal.

At this time, the CPU 31 immediately returns from the step SP30 to the main routine, ignoring the occurrence of the peak and ending the processing program without controlling the tone generation.

When the jumping vibration processing loop LP3 is entered, the CPU 31 first determines in step SP35 whether or not the pad status data PDKON is "0". If a negative result is obtained here, the pad is in status "1" or "2",
Therefore, it is in a state of vibrating by the first or second performance operation, and at this time, the CPU 31 returns from step SP40 to the main routine.

However, in such a vibrating state, even if there is a jumping vibration from another pad, the vibration is very small. Therefore, it becomes unnatural when a special sound is newly generated based on the jumping vibration. Therefore, in the case of such a status condition, the processing program is terminated without making a special sound.

On the other hand, if a positive result is obtained in step SP35, the CPU 31 moves to step SP36 and transfers the pad number data P
ADNO and the previous operation pad number data MINPD are used as address data to store the low volume sounding coefficient data table DMTBL in the table memory 32A (FIG. 10).
The low-volume sounding coefficient data D _{(PADNO) (MINPD)} is read from.

This small volume sounding coefficient data D _{(PADNO) (MINPD)} is small depending on which pad gives the jumping vibration when the other pad has a jumping vibration. It represents a coefficient for determining the volume at which sound is emitted, and the low volume sound coefficient data D _{(PADNO) (MINPD)} read from the low sound volume coefficient data table DMTBL is jumped to the sound volume coefficient data DMRA.
Write as TIO in the jump volume coefficient register REG6 (Fig. 5).

In this embodiment, the jumping sound volume coefficient data D
_{(PADNO) (MINPD),} for example as shown in FIG. 16 for the first pads PAD1, jumped between the second pads PAD2 mutually adjacent volume sound volume coefficient data D _{1 2,} D _{2 1} a D _{1 2} = D _{2 1} = 1/10, using D _{1 3} = D _{3 1} = 1/20 as the jumping sound volume coefficient data D _{1 3} and D _{3 1} with the third pad PAD ₃ , jumped pronunciation volume coefficients data _{D 1} 4 between the pads _{PAD4, D 4 1} using the _{D 1 4 = D 4 1 =} 1/10 as, sound volume coefficient data _{D 1} 5 jumped for the fifth pads PAD5, _{D 5} with _{D 1 5 = D 5 1 =} 1/16 as _1, the _{D 1 6 = D 6 1 =} 1/20 as jumped sound volume coefficient data _{D 1 6, D 6 1} between the sixth pads PAD6 To use.

In this way, the pad PAD1 that detected the peak occurrence
Based on the fact that the closer the distance to the pad is, the greater the influence of the jumping vibration is, the larger the sounding volume coefficient data D _{(PADNO) (MINPD)} , the larger the coefficient data that is assigned. It is possible to make a special sound of a volume having a sounding effect similar to the jumping sound that occurs in a musical instrument.

When the processing of step SP36 is completed, the CPU 31 multiplies the peak level data ADVAL by the jump sound volume coefficient data DMRATIO in the next step SP37, and the multiplication result is used as jump sound level data DMADVL to jump sound level data register REG7 (fifth). (Fig.)

Subsequently, the CPU 31 proceeds to step SP38, where the pad number data PADNO stored in the pad number register REG2 is stored.
And the jumping tone level data DMADVL held in the jumping tone level register REG7 together with the key-on signal KON as rhythm-on occurrence information S31.
G) Send to 35. At this time, the musical tone signal generator (TG) 35 sends the musical tone signal S32 to the sound system 36 so that the percussion instrument sound corresponding to the pad number data PADNO jumps to a special tone with a small volume determined by the tone level data DMADVL.

In this way, the CPU 31 ends the processing of the jump vibration processing loop LP3, and returns from the step SP39 to the main routine.

[5] Operation of the first embodiment In the above-mentioned configuration, the CPU 31 operates the past pads when the peak is detected at the present time t ₀ based on the performance operation information signal S1 sent from the pads PAD1 to PAD6. The pronunciation of the percussion instrument sound is controlled in different control modes depending on the state, that is, the status "0", "1", and "2".

(1) In case of status “0” In this case, as described above with reference to FIGS. 8 (A1) and (A2), the pad in which the peak is detected within the first reference time TR1 from the current time t ₀ Regarding the fact that the performance operation is not performed within the first reference time TR1, the sound is ready to be sounded whenever the performance operation is performed.

At this time, the CPU 31 determines that the pad-on interrupt routine RT1
After writing the pad number data PADNO and the peak level data ADDVAL in the pad number register REG2 and the peak level register REG1, respectively, in steps SP11 and SP12 which constitute the performance operation information acquisition processing loop LP1 of FIG. For the pad operated immediately before in step SP13 other than the relevant pad, the immediately previous operation pad time interval data MINLAP representing the operation time interval with the relevant pad.
Also, the immediately previous operation pad number MINPD is detected, and the immediately previous operation pad time interval data MINLAP is the third reference time TR3.
When it is confirmed in step SP15 that the value is less than (= 5 [ms]), jumping vibration from the immediately preceding operation pad may occur. Therefore, in step SP16, the jumping vibration detection coefficient data KDATA is set in the jumping vibration detection coefficient register. Hold in REG3.

Here, when the pad is operated normally at the current time t ₀ , the peak level data obtained from the pad
Since the value of ADVAL is quite large, CPU31
It is detected in SP17 and the process proceeds to the tone generation processing loop LP4.

At this time, the CPU 31 detects in step SP21 that the status is "0", and then in step SP22,
The first reference time TR1 in the elapsed time data LAPCTR (PADNO)
(That is, by setting the value 15 of 15 [ms]) to start the timing operation of the first reference time TR1 and
In step SP23, as shown in FIG. 17, percussion instrument sound is generated in the tone signal generator (TG) 35 based on the pad number PADNO and the peak level data ADVAL.

In this way, in the status "0", the CPU 31 immediately generates the corresponding percussion instrument sound when the pad is operated for performance, and is in the status "1" state in which the elapsed time for the first reference time TR1 is measured. (Set in (B1) and (B2) of FIG.

(2) Status "1" Peak interrupt signal S13 in status "1"
As described above with reference to FIGS. 8 (B1) and (B2), the CPU 31 generates the percussion sound corresponding to the peak detection by performing the second peak detection at the current time t ₀ as described above with reference to FIGS. The timekeeping operation of the second reference time TR2 is started.

That is, when the peak detection interrupt S13 occurs, the CPU 31 causes the pad number data PADNO and the peak level data ADV in the performance operation information acquisition processing loop LP1 (FIG. 6).
Immediately after AL is held in the register 34A, the previous operation pad detection processing loop LP2 is executed.

Here, if the peak detection occurring at the current time t ₀ is based on the regular performance operation for the pad, the CPU 31 performs the same processing as described above for the status “0” in the immediately preceding operation pad detection processing loop LP2. It executes and enters the tone generation processing loop LP4.

Here, when the CPU 31 determines in step SP21 that the past state is the status "1", step SP25
Sound level data ADATA (PADNO) and elapsed time data LAPCTR (PADN held in register 34A at
O) using coefficient data table DCTBL for self-vibration detection
From Fig. 13 and Fig. 14, self-vibration detection coefficient data DV
_{(ADVAL) (15-LAPCTR)} is read and held in the register 34A as self-vibration detection coefficient data DCVAL, and then this self-vibration detection data DCVAL is compared with the peak level data ADVAL (step SP26).

If the peak level data ADVAL is large, as shown in FIG. 18 (A), the peak detection state generated at the present time t ₀ is based on the self-vibration caused by the performance operation of the pad in question. Therefore, the CPU 31 sounds the percussion instrument sound designated by the pad number data PADNO at the sound volume of the peak level data ADVAL in steps SP27 and SP28, and outputs the pad status data PDKON as status. 2 "data and set the second reference time TR2 (that is, 30 [ms]) to the elapsed time data LAPCTR (PADNO).

Thus, the pad in the status "1" is the first
When the second performance operation is performed during the reference time TR1 of, the percussion instrument sound corresponding to the pad is generated.

On the other hand, when the peak level data ADVAL is smaller than the self-vibration detection data DCVAL held in the self-vibration detection data register REG8 for the pad where peak detection has occurred, as shown in FIG. 18 (B). Since the peak generated at the time t ₀ is not self-oscillation, the CPU 31 terminates the pad-on interrupt routine without making a sound for the pad number data PADNO by jumping the processing of steps SP27 and SP28. .

(3) Case of status "2" When a peak is detected in the status "2" described above with reference to FIGS. 8 (C1) and (C2), the CPU 31 performs the performance operation information processing loop LP1 and the immediately preceding operation pad detection processing. After the processing of the loop LP2 is executed, it is determined in step SP21 of the tone generation processing loop LP4 that the state of the pad is the status "2".

In this case, as shown in FIG. 19, the peak detected at the current time t ₀ occurs during the second reference time TR2, and since a normal playing operation by the stick cannot be actually performed, the CPU 31 Ends the pad-on-interrupt routine without performing sound generation processing.

(4) In case of jumping vibration When it is judged that the vibration of the relevant pad for which peak detection has occurred in the past operation state of status "0" is small, CP
U31 performs a special sounding control that produces the same effect as the jumping sound in a natural musical instrument by sounding at a low volume.

That is, the CPU 31 registers the pad number data PADNO and the peak level data ADVAL in the performance operation information acquisition processing loop LP1 of the pad-on interrupt routine RT1 in the register 34.
It is determined that the vibration is a jumping vibration in step SP17 of the operation pad detection processing loop LP2 immediately before being captured in A.
Incidentally, since the vibration of the pad (peak level data ADVAL) is small, the ratio ADVAL / ADATA (MINPD) becomes smaller than the jump vibration detection coefficient data KDATA during the immediately preceding operation pad. That is, when a pad other than the peak detection pad is operated for performance, as described above with reference to FIG. 12, a vibration attenuated by the ratio of K _{(PANDO) (MINPD) arrives at the} pad, but the damped vibration. If the ratio of the vibrations currently generated is smaller, it may be determined that the vibration is a jumping vibration.

At this time, the CPU 31 reads the corresponding jump sound volume coefficient data D _{(PANDO) (MINPD)} from the small sound volume coefficient data table DMTBL (FIG. 15) in steps SP36, SP37 and SP38 and registers it as jump sound volume coefficient data DMRATIO. It is held in 34A (FIG. 5) and is multiplied by peak level data ADVAL to form jumping tone level data DMADVL.

Thus, as in the case where the past state is the status "0", in the pad where the peak is detected in the state where the percussion instrument sound is not generated, the percussion instrument sound of a small volume is generated like the jumping sound produced in the natural musical instrument. By doing so, a more natural percussion instrument sound can be generated in the sound system 36.

On the other hand, when a small vibration peak occurs in the statuses "1" and "2", the CPU 31 does not respond to this (steps SP35 and SP40) to prevent unnatural sounding. .

[6] Effect of the first embodiment According to the above-mentioned embodiment, in the jump vibration processing loop LP3 composed of steps SP35 to 40, for a pad in which a relatively small vibration peak is detected, this is referred to as jump vibration. By making a judgment and making a special sound of a low volume, it is possible to generate a percussion instrument sound having a natural feeling like a natural instrument.

Further, in the case of the above-described embodiment, the peak level data of the peak generated when the past state is the status "1" in the tone generation processing loop LP4 including the steps SP25, SP26 and SP27.
ADVAL is the self-vibration detection coefficient data DV read from the self-vibration detection coefficient data table DCTBL.
_{(ADVAL) (15-LAPCTR)} (that is, self-vibration detection data DCVA
L) The percussion instrument sound is generated only when it is larger than L. Even if there is a peak in the same pad that would not occur in the case of natural attenuation in the same pad, the percussion instrument is erroneously determined based on this. No sound can be generated and thus erroneous pronunciation can be reliably prevented.

Further in accordance with the above embodiment, steps SP25-SP29 and
As in SP30, when the status "1" has already been pronounced once and then pronounced the second time, CPU31
Sets the value of the second reference time TR2 (for example, 30 [ms]) as the elapsed time data LAPCTR (PADNO) and sets the status "2" data as the pad status data PDKON (PADNO). As a result, it is possible to prevent the occurrence of erroneous pronunciation by preventing unnatural percussion instrument sound after the pad is repeatedly hit twice.

[7] Other Embodiments (1) In the above embodiments, the present invention is configured to be realized by control by software, but the present invention is not limited to this, and dedicated hardware may be used. May be.

(2) In the case of the above-mentioned embodiment, the algorithm as shown in FIG. 6 is used as the pad-on interrupt routine, but the algorithm and the numerical value representing the method of detecting the sounding state are not limited to the above-mentioned embodiment, and other Various ones can be applied.

(3) In the above-mentioned embodiment, an embodiment in which the present invention is applied to an electronic percussion instrument has been described, but the present invention is not limited to this, and an electronic percussion instrument forming a part of an electronic or electric musical instrument such as an autorhythm device or an electronic keyboard instrument. You may make it apply.

(4) In the case of the embodiment shown in FIG. 2, the latch register 14 for latching the pad number variation data S14 and the peak value latch data S15 is provided in common to a plurality of pads PAD1 to PAD6 and operated in a time division manner. By doing so, the performance operation information of a plurality of pads is fetched, but instead of this, for example, a latch register is provided for each of a plurality of pads PAD1 to PAD6, and the latch data of each latch register is sequentially assigned in a predetermined priority order. Therefore, various modifications can be made, such as processing.

Similarly, the analog / digital conversion circuit 15 and the peak detection circuit 25 may be provided for each pad.

(5) In the above embodiment, the first, second, and third reference times TR1, TR2, and TR3 are set to TR1 = 15 [ms] and TR2, respectively.
= 30 [ms] and TR3 = 5 [ms] have been described, but this time is not limited to this, TR1 = 5 to 30 [m]
s], TR2 = 10 to 50 [ms], and TR3 = 2 to 10 [ms].

(6) In the above embodiment, the self-vibration detection coefficient data DV is calculated from the self-vibration detection coefficient data table DCTBL.
_{(ADVAL) (15-LAPCTR)} is configured to be stored in self-vibration detection data register REG8 as self-vibration detection data DCVAL. Instead of this, self-vibration detection data DCVAL is calculated by the following formula DCVAL = ADATA (PADNO)-[ 15-LAPCTR (PADNO)] ……
Level data ADATA (PADNO) at the time of sounding using the arithmetic expression of (1)
It may be calculated from (Fig. 20).

(7) In the above-described embodiment, the electronic percussion instrument 1 has a configuration in which a plurality of pads PAD1 to PAD6 are arranged on the board 2 as described above with reference to FIG.
The present invention is not limited to this, and in short, it can be widely applied to a structure in which adjacent pads are affected by vibrations when a plurality of pads are played.

(8) Pad-on interrupt routine RT1 in FIG.
In the case of the above embodiment, the determination data KDAT is used when determining whether or not the vibration is a jumping vibration in step SP17.
A is the jump vibration detection coefficient data table KTABLE
Ratio data K for each pad from (Fig. 11)
_{I tried} to read _{(PADNO) (MINPD)} , but instead of this,
The same ratio data may be used for all pads.

(9) Pad-on interrupt routine RT1 in FIG.
In step SP37, in obtaining jumping sound level data DMADVL, self vibration detection coefficient data DV _{(ADVAL) (15-LAPCTR)} stored for each pad as shown in FIG. 13 is used as self vibration detection coefficient data. Table DCTBL
This self-vibration detection coefficient data table DCTBL
The peak level data ADVAL from the pad where the peak is detected by reading the data from is read out is multiplied, but the peak volume data ADVAL is replaced by a predetermined calculation to determine the low volume sound level. You may do it.

(10) Pad-on interrupt routine RT1 in Fig. 6
In the case of, when the low volume is pronounced, the volume is set to a value determined by the jumping tone level data DMADVL, but instead of or in addition to this, for example, a high-frequency signal component is set. Change the timbre as needed, such as by cutting (for example, by using a low-pass filter) to make the generated low-volume percussion instrument sound inconspicuous, or conversely to change to a timbre specific to pronunciation. It may be done.

When making a jumping sound, the sound level may be controlled, for example, by making the attack rate gentle.

(11) In the above-mentioned embodiment, the case where only two tones are sounded at the same time during the first reference time TR1 has been described, but the present invention is not limited to this, and a plurality of tones may be sounded. good.

In this case, if there is a pad in which peak detection has occurred for a plurality of sounds within the third reference time, special sounds may be generated for all of them.

(12) In the above-described embodiment, an embodiment in which the present invention is applied to an electronic percussion instrument has been described. However, the present invention is not limited to this, and a vibrating body is vibrated by a performance operation, such as an electronic stringed instrument. It can be widely applied to electronic musical instruments.

(13) In the above-described embodiment, as described above with reference to FIG. 18, when the self-vibration occurs in the status “1”, the peak occurrence based on the self-vibration is the first occurrence.
Based on the self-vibration detection data DCVAL that is during the reference time TR1 of and the vibration level is the self-vibration detection coefficient data DV _{(ADVAL) (15-LAPCTR)} read from the self-vibration detection coefficient data table DCTBL. Although it was designed to prevent erroneous pronunciation due to self-vibration by determining whether or not to perform the sounding operation for the second peak detection, instead of this, as shown in FIG. 20, self-vibration detection data DCVAL is detected.
May be calculated by using an arithmetic expression (for example, a linear expression) in which the elapsed time data 15-LAPCTR (PADNO) is used as a variable.

(14) Also, to prevent erroneous pronunciation due to self-vibration,
As shown in FIGS. 21 and 22, when peak detection occurs within the fourth reference time TR4, it is possible not to generate sound based on the peak detection.

That is, as shown in FIG. 21, when in the state of the status "0" at time t _{1 0} previously when the peak occurs at time t _{1 0} (Figure 21 (A)), the peak detection circuit 25 As shown in FIG. 21 (B), when the peak interrupt signal S13 is generated, the CPU 31 enters the pad-on-interrupt routine RT11 shown in FIG. 22, and the pad number variation data S14 representing the pad operated and played is padded. The number data PADNO is taken into the register 34A, and the peak value latch data 14B is written into the register 34A as the peak level data ADVAL at step SP42.

Thus, the CPU 31 ends the performance operation information acquisition processing loop LP1 and proceeds to the next step SP43.

The step SP43 is a fourth reference time from the current time point t _{1 0} for the pad of the same pads that peak detection occurs TR4
When a negative result is obtained in the step of judging whether or not the pronunciation processing is executed within (for example, 15 [ms]), this means that the fourth criterion has not been reached since the same pad was pronounced immediately before. It means that the time TR4 has not elapsed.

Actually, this fourth reference time TR4 is the mask time data MSK
_It is held in the register 34A as _(PADNO) and is timed by the CPU 31 by the timer interrupt routine RT2 (FIG. 7) like the elapsed time data LAPCTR (i) in FIG.

At this time, the CPU 31 shifts to step SP44 and sets the fourth reference time TR4 to the mask time data MSK _(PADNO) , and then at step SP45, outputs the pad number data PADNO and the peak level data ADVAL together with the key-on signal KON to the musical tone signal. Sound system by sending to generator (TG) 35
At 36, a percussion sound is generated.

In this way, the CPU 31 ends the pad-on interrupt routine RT11 and returns from step SP46 to the main routine.

As shown in FIG. 21 (A) In contrast, the peak interrupt signal S13 for the pad peak at time t _{1 1} before the fourth reference time TR4 after time t _{1 0} elapses occurs and When this occurs (FIG. 21 (B)), the CPU 31 executes the processing of steps SP41 and SP42, and then obtains a positive result in step SP45, so that the pad-on interrupt routine is executed without executing the processing of steps SP44 and SP45.
Exit RT11 and return from SP46 to the main routine.

In this way, the CPU 31 executes the tone generation process once for the same pad, and then detects the peak again when the fourth reference time TR4 (and therefore the mask time data MSK _(PADNO) does not elapse again _). This control is performed so as not to perform the sound generation process of, for example, in an electronic musical instrument in which a pad is operated by one stick,
For example, even if the pad is operated so strongly that a peak is generated due to self-vibration, it is possible to prevent an unnatural percussion instrument sound from being generated.

(15) In the above-described embodiment, a plurality of, for example, six pads PAD1 to PAD6 are provided as the performance operators 3, but even if the number of the performance operators 3 is one, the same as in the case described above. The effect can be obtained.

〔The invention's effect〕

As described above, according to the present invention, when the performance operation detection information is obtained due to the occurrence of the abnormal vibration in the performance operator, the abnormal vibration is detected by using the self-vibration detection data so that the sounding operation is not performed. By doing so, it is possible to easily realize an electronic musical instrument in which an unnatural musical sound is not generated due to abnormal vibration.

[Brief description of drawings]

FIG. 1 is a flow chart showing a principle processing procedure in an embodiment of an electronic musical instrument according to the present invention, FIG. 2 is a system diagram showing an overall configuration of the first embodiment, and FIG. 3 is a pad PAD1 of FIG.
~ Connection diagram showing the performance operation information output circuit of PAD6, Fig. 4 is a signal waveform diagram showing the signals of the respective parts of Fig. 3, and Fig. 5 is the second
FIG. 6 is a schematic diagram showing a detailed configuration of a register provided in the data / working memory 34, FIG. 6 is a flow chart showing a pad-on interrupt routine, FIG. 7 is a flow chart showing a timer interrupt routine, and FIG. Is a time chart used to explain the basic control method of the CPU 31 in FIG. 2, FIG. 9 is a flow chart showing the detailed processing step of step SP13 in FIG. 6, and FIG. 10 is the program / table data in FIG. FIG. 11 is a schematic diagram showing the structure of the table memory of the memory 32, FIG. 11 is a table showing the jump vibration detection coefficient data table of FIG. 10, and FIG. 12 is a diagram showing the relationship between the data of FIG. 11 and the pads PAD1 to PAD6. FIG. 13 is a schematic diagram showing the self-vibration detection coefficient data table of FIG. 10, FIG. 14 is a schematic diagram showing each data group thereof, and FIG.
Chart showing the low volume pronunciation coefficient data table of Figure 10,
FIG. 16 is a schematic diagram showing the relationship between the data of FIG. 15 and the pads PAD1 to PAD6, FIGS. 17, 18, and 19 show the vibration states of status “0”, “1”, “2”. Signal waveform diagram,
FIG. 20 is a characteristic curve diagram showing another embodiment of the self-vibration detection data, FIGS. 21 and 22 are signal waveform diagrams and a flow chart used for explaining another embodiment of the electric musical instrument according to the present invention,
FIG. 23 is a schematic plan view showing the structure of an electronic percussion instrument, and FIG. 24 is a signal waveform diagram used for explaining abnormal vibration. 2 ... Board, 3 (PAD1 to PAD6) ... Performance operator (pad), 11 ... Electronic musical instrument, 14 ... Latch register, 14A ...
… Pad number variation part, 14B …… Peak value latch part, 2
5 …… Peak detection circuit, 31 …… CPU, 35 …… Music signal generator (TG), 36 …… Sound system, 37 …… Interrupt timer, 34 …… Data / working memory, 34A ……
register.

Claims (1)

[Claims] 1. An electronic musical instrument which has (a) a performance operator and generates a musical tone corresponding to an operation amount of the performance operator, and (b) a performance operation detection corresponding to the operation amount of the performance operator. (C) When the performance operation detection information of the performance operator is obtained at the present time, the performance operation detection means for forming information is obtained by the previous performance operation with the same operator as the performance operator. The self-vibration detection data determined by the amount of performance operation based on the previous performance operation detection information and the elapsed time from the time when the previous performance operation detection information was obtained to the current time is formed, and the performance operation at the current time is performed. Sound generation control means for comparing the detection information with the self-vibration detection data and forming sound information based on the comparison result when the previous performance operation detection information is sound-allowed. Electronic musical instrument, characterized in that in response to the sound information comprising a tone generating means for pronunciation operations of the musical sound.