JPH07109558B2 - Electronic musical instrument - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of use) The present invention relates to the same key in which the same key is pressed again so that the pronunciation corresponds to the pronunciation corresponding to the key depression of a key once released. The present invention relates to an electronic musical instrument that processes continuous tapping.

(Prior Art) Conventionally, in the electronic musical instrument as described above, when the same key is newly pressed again, a tone generation different from the tone generation channel to which the tone is assigned by the previous key depression is generated. A process of simply assigning a musical sound to the new key depression was performed on the channel.

Further, in addition to the above-described processing, processing for rapidly attenuating the tone generation of the tone generation channel by the previous key depression after the start of the tone generation of the tone generation channel by the new key depression Was being conducted.

(Problems to be Solved by the Invention) However, according to the former, there is a problem that the amount of pronunciation increases unnecessarily, or the number of musical tone generation channels becomes insufficient and the pronunciation of other keys that are currently sounding stops when continuous tapping continues. There is a point.

Further, according to the latter, the above-mentioned problems of the former are solved, but there is a problem that when the weak key is pressed after the strong key is pressed, the sound suddenly becomes small.

The present invention has been made for the purpose of solving such a problem.

(Means for Solving the Problems) In order to achieve the above-mentioned object, the constitutional features of the electronic musical instrument according to the present invention are, as shown in FIG. 1, (a) a newly depressed key Continuous hit detection means (1) for detecting whether or not a musical instrument having the same key as that already assigned to a tone generation channel, (b) a first tone amount of a tone to be sounded by the new key depression or its pronunciation A second value of the tone corresponding to the value corresponding to the amount and the time point to be sounded by the new key depression by the same key already assigned to the tone generation channel.
Detecting means (2) and (c) for detecting the sound output amount of the key or the key corresponding to the sound output amount, and comparing the first and second sound output amounts or the values corresponding to the sound output amounts, and the continuous hit detecting means. Is provided with comparison and selection means (3) for selecting a side having a large sound generation amount or a value corresponding to the sound generation amount to give priority to sound generation when a continuous stroke is detected.

(Operation) A side having a large sound amount or a value corresponding to the sound amount of the same key is selected and sounded from the tone generation channel, while a side having another small sound amount or a value corresponding to the sound amount is rapidly attenuated. As a result of processing and the like, the occupation of the tone generation channel can be released quickly.

(Effect of the invention) Therefore, it is possible to prevent the sound generation amount from being unnecessarily increased at the time of repeated hits, and to avoid stopping the sound generation of other keys as much as possible because the tone generation channel is not unnecessarily occupied. You can

Furthermore, the sound being sounded at the time of repeated hits and the new sound are naturally connected without any difficulty.

(Examples) Next, specific examples of the electronic musical instrument according to the present invention will be described with reference to the drawings.

First Embodiment: FIG. 2 schematically shows an electronic musical instrument, in other words, an electronic keyboard musical instrument, for producing a musical sound of a damped sound system (percussive system) to which the present invention is applied. In FIG. 2,
The key operation detection circuit 22 operates which of the keys is operated based on the operation of pressing or releasing each key on the keyboard 20 which is composed of a plurality of keys for designating the pitch of the pronunciation to be emitted. Whether or not the key is pressed or released. The pitch information indicating the pitch of the key operated in the built-in buffer and the key press indicating the key pressed / released state are detected.・ Accumulate key release information. Under the control of the microcomputer 21, the stored information is stored in the bus 23 as data of the key code BKYC, the key status flag BKYS, and the total number KEN of changed keys indicating the number of keys that have changed during the storage period.
Is supplied to the microcomputer 21 via. Similarly, the key pressing speed based on the key operation, the key touch strength of the key pressing force, etc. are detected by the touch response detection circuit 24 and stored as touch response information in the buffer built in the touch response detection circuit 24. The accumulated information is supplied to the microcomputer 21 via the bus 23 as touch response data BKTD corresponding to the above-mentioned data BKYC, BKYS, etc. under the control of the microcomputer 21. Furthermore, this microcomputer
21 includes an operator group 25 for switching or adjusting the tones of the piano, harpsichord, etc., and the generated volume.
The operation state of is detected by the operator detection circuit 26 and supplied as operator data MNPh. In addition, a damper pedal that prohibits the dumping process that accelerates the decay of the musical sound by the depression operation to lengthen the decay time, and a dumping process of the key during key depression at the time of depression to prevent the rapid decay of the musical sound, The pedal detection circuit 28 detects the stepping state of the pedal group 27 composed of the sostenuto pedals in which the musical tone is rapidly attenuated by the dump process by releasing the stepping operation after releasing the key, and the damper state flag FCDS and the sostenuto state are detected. The flag is supplied. The manipulator data MNPh, the damper state flag FCDS, and the like represent the manipulator state and the stepped state at the time of supply to the microcomputer 21 in the control of the microcomputer 21. The key code BKYC, the key status flag BKYS, and the touch response data BKTD described above form the key data BKYD.

The details of the sostenuto state flag and the processing related thereto are left to the publicly known documents and the like, and are omitted for simplification of description.

The microcomputer 21 has a central processing unit (CPU) 21A for executing a predetermined program, a read-only memory (ROM) 21B for storing the program, a working memory necessary for executing the program, and the operations described above. It is composed of a writable memory (RAM) 21C as various registers assigned to store the child data MNPh, damper status flag FCDS, key data BKYD, etc., and a timer circuit 21D for measuring the time during this program. Then, the manipulator data MNPh and the damper state flag F described above are used.
By executing the program based on the CDS, the key data BKYD, etc., in the present embodiment, the musical tone generating circuit 29 having 16 musical tone generating channels is controlled, and the desired musical tone is generated by the predetermined musical tone generating channels. A signal is generated and generated as a musical sound from the speaker 31 via the amplifier 30.

Next, the basic operation of the electronic musical instrument according to the present invention configured as described above will be described in detail for each step based on the flowchart of the basic program of FIG.

A When the power is turned on, a predetermined program is started to clear the contents of the RAM 21C in the microcomputer 21 such as allocated as various registers, and the key operation detection circuit 22, the touch response detection circuit 24, and the operator detection circuit
26, the pedal detection circuit 28, and further, the musical tone generation circuit 29 are instructed to perform the desired setting.

B Read the manipulator data MNPh from the manipulator detection circuit 26,
With the manipulator data MNPh, parameters are read from a predetermined table stored in the ROM 21B and converted into a parameter group GTEm relating to musical tone generation, and the converted parameter group GTEm is written into a predetermined register GTEmR. The register GTEmR is provided corresponding to each of the 16 tone generation channels constituting the tone generation circuit 29, in other words, corresponding to each envelope waveform generation channel corresponding to each tone generation channel.

The damper state flag FCDS, which indicates the state of depression of the damper pedals of the pedal group 27 by "1", is read from the C pedal detection circuit 28 and written in the register FCDSR.

D The key code BKYC based on the pitch information and the key depression / key release information accumulated in the buffer after the last reading from the key operation detection circuit 22 and the key state flag BKYS indicating the key depression state as "1" In addition to reading in the time sequence that occurred, the total number KEN of change keys indicating the number of keys that have changed since the last read time is read based on these key press / key release information. Similarly, the touch response data BKTD is read according to the time sequence generated from the touch response detection circuit 24. These key codes BKY
C, key status flag BKYS and touch response data BK
Key data BKYD corresponding to each other as described above in TD
And write to the corresponding area of register BKYR according to the time sequence that occurred.

In addition, the total number of changed keys KEN is calculated as the number of new pending keys BKE
Write to register BKENR as N.

E Number of pending keys written in register BKENR B
Depending on whether or not KEN is "0", it is determined whether or not the key operation processing based on the key depression or key release is completed. When the number of processing waiting keys BKEN is “1” or more and the processing of the key operation is not completed, the process goes to step G.

F The number of keys waiting to be processed in the judgment in step E
When the BKEN is "0" and the key operation processing is completed, the predetermined envelope processing is sequentially performed in each envelope waveform generation channel as follows.

i) A predetermined table of envelope waveforms stored in the ROM 21B is read out, and is written in the corresponding tone register parameter GTE _m written in the corresponding register GTE _m R, and similarly in the corresponding registers KYCR, KTDR described later. Based on the embedded key code KYC and touch response data KTD, the rate RT indicating the change value of the envelope (including positive and negative depending on the increase and decrease of the envelope) per predetermined time and its accumulated rate. The rate RT _j group and the breakpoint LBP each consisting of the breakpoint LBP indicating the change point of the envelope slope at which the RT value changes
_Compute and generate group _j .

ii) Calculate the envelope level LEV based on the calculated predetermined rate RT _j group and breakpoint LBP _j group,
In other words, the envelope waveform is calculated (this envelope level LEV is calculated by first accumulating at a predetermined rate RT of the calculated rate RT _j group and calculating the accumulated value similarly at the breakpoint LBP. _When a predetermined break point LBP in the _j- group is reached, it is repeated by accumulating at the next predetermined rate RT, etc.).

iii) In the envelope waveform generated as described in the previous section, the attack part A in the so-called ADSR expression ends (breakpoint LBP corresponding to the end of the attack part A).
When it is judged whether the envelope level LEV has reached "-AT"), the attack end flag EV "-AT" is set to "0".
And when the decay section D is terminated (determined by whether or not the envelope level LEV reaches the breakpoint LBP "-DK" corresponding to the termination of the decay section D), the decay end flag EV "-DK" is set to "0". To Further, when the release section R is ended (determined by whether or not the envelope level LEV reaches the breakpoint LBP “-END” corresponding to the end of the release section R), the envelope end flag
Set EV "-END" to "0". This envelope end flag
The corresponding tone generation channel is released by EV "-END""0".

(These rates RT ₃ groups, breakpoint LBP _j group, target rate RT and breakpoint LBP, envelope level LEV and various flags EV "-AT", EV "-DK", E
V "-END" is set corresponding to each envelope waveform generation channel including a silence processing request flag DMPQ and a silence processing in progress flag RDMP which will be described later. Therefore, the register RT that writes and / or reads these data
_j R, LBP _j R, RTR, LBPR, LEVR, EV-ATR, EV-DKR, EV-ENDR, DM
PQR and RDMPR are also provided for each envelope waveform generation channel. Then, these form a group for each envelope waveform generation channel and are handled as a group. ) When the key release envelope processing flag RKOF written in register RKOFR is set to “1” and the key release processing (step K) described later is started, the damper state written in register FCDSR is written. When the flag FCDS indicates "0" that the damper pedal of the pedal group 27 is not depressed, the attack end flag EV "-AT" written in the corresponding register EV-ATR indicates that the attack portion A After indicating the end "0", the key release envelope processing flag RKOF is reset to "0" and the envelope waveform is changed to a predetermined key release envelope. In addition, the corresponding register DMPQR
If the mute processing request flag DMPQ written in the field indicates the mute processing request “1”, this mute processing request flag DM
In order to reset the PQ to "0" and generate a sound characterizing the attack to give an attack feeling, similarly, after the attack end flag EV "-AT" indicates the end of the attack part A, a predetermined value is set. Change to the muffling envelope of. The method of generating the key release envelope and the muffling envelope conforms to the above-described method of generation.

After the envelope processing, the process returns to step B.

G If the key operation processing is not completed in the judgment at step E, the oldest key data BKYD among the key data BKYD written in the register BKYR is read (first-in first-out method), and this key data is read. It is determined whether or not the key corresponding to the key data BKYD read by the key status flag BKYS included in BKYD is in the depressed state. When the key status flag BKYS indicates "0" and the key is not in the key released state, the process goes to step K.

If the key state flag BKYS indicates "1" in the key depression state in the judgment of H step G, the register BKENR
Subtract "1" from the number of pending keys BKEN written in. The number after this subtraction is written in the register BKENR as the number of new waiting keys BKEN.

(Musical tone is assigned to each tone generation channel by the key code KYC, key status flag KYS, touch response data KTD, and pitch data FQY set for each musical tone generation channel. Code KYC,
Key code BKYC, of the predetermined key data BKYD read from register BKYR in registers KYCR, KYSR, and KTDR for key status flag KYS and touch response data KTD, respectively
By writing the key status flag BKYS and touch response data BKTD that are "1" as the key code KYC, key status flag KYS and touch response data KTD of the key data KYD, the register GTE _m corresponding to the register FQYR is further written.
Parameter group for tone generation written in R GTE _m
And the pitch data FQY calculated by the key code KYC written in the register KYCR. Further, the mute processing request flag DMPQ, the mute processing flag RDMP, and the key release envelope processing flag RKOF are reset to “0” and written to the predetermined registers DMPQR, RDMPR, and RKOFR corresponding to the corresponding envelope waveform generation channel. , The envelope level LEV is cleared by clearing the register LEVR. Also, the above-mentioned register RT
_{This is performed} by writing the rate RT _j group and the breakpoint LBP _j group to _j R and LBP _j R, and setting various flags of the registers EV-ATR, EV-DKR and EV-ENDR to "1". ) The assignment to the tone generation channel is performed as follows.

i) Ending and releasing the sound from the key status flag KYS written in the register KYSR of each tone allocation channel and the envelope end flag EV "-END" written in the register EV-ENDR of each envelope waveform generation channel. The generated tone generation channels are detected and assigned as described above to instruct the start of sound generation. The assigned tone generation channel number is stored in the register ANCHR as the new key press channel number ANC.
Further, as H, the key code BKYC written in the register KYCR by allocation is written in the register ANKCR as the key code ANKC of the new key depression. The time timer TST from the start of sounding, which is written in the register TSTR provided for each of the musical tone allocation channels, is reset and the next step I
go to.

ii) If no released tone generation channel is detected, the envelope level LEV and attack end flag EV “-AT” written in the registers LEVR, EV-ATR of each envelope waveform generation channel are set. Envelope level LEV that has finished attack part A while producing sound
Detects the smallest tone generation channel of. Then, as described above, it is assigned and the start of sounding is instructed. The assigned tone generation channel number is stored in register ANCHR, and the key code BKYC written in register KYCR is stored in register ANKC.
Write to R. Further, the time timer TST from the start of sound generation is reset and the process goes to the next step I. In this case, although the register LEVR is reset to stop the sound generation, it is preferable to perform the rapid attenuation process.

I Continuous hit detection routine. Details will be described later with reference to the flowchart of the continuous hit detection routine shown in FIG.

J A mute processing target detection and instruction routine. The details will be described later with reference to the flowchart of the detection and instruction routine of the noise reduction processing target shown in FIG.

If the key state flag BKYS indicates "0" in the judgment of step G and the key is not in the key release state, "1" is subtracted from the number of waiting keys BKEN written in the register BKENR. Then, the number after this subtraction is written in the register BKENR as the number of new waiting keys BKEN.

By the key code BKYC contained in the key data BKYD written in the register BKYR, the key code KYC written in each of the registers KYCR and KYSR in each tone allocation channel and the key code BKY in the key status flag KYS.
C and KYC are the same, and the key status flag KYS detects a tone allocation channel that indicates "1" during key depression, and sets the key release envelope processing flag RKOF to "1" during key release envelope processing. The key state flag KYS is changed to "0" indicating that the key has been released, the start of the key release process is instructed, and the process returns to step E.

When the tone generation channel as described above is not detected, the process directly returns to step E.

Next, the continuous hit detection routine (step I) will be described step by step with reference to FIG.

Set the number of loops n written in the I-1 register nR to "1",
Also, the mute processing start flag DMP written in the register DMPFR.
Initialize F to "0", which does not indicate the start of mute processing.

1-2 With the damper state flag FCDS written in the register FCDSR, it is determined whether or not the damper pedal of the pedal group 27 is depressed and the dump processing is prohibited even if the key is released. If the damper pedal is not in the depressed state and the damper state flag FCDS is "0" indicating that dump processing is not prohibited, the duration of sound generation in the tone generation channel subject to dump processing is short and exceptional. The routine is terminated because there is no danger of causing trouble even if the processing is not performed.

I-3 If the damper state flag FCDS indicates "1" and the dump processing is prohibited in the judgment in step I-2, the channel number ANCH of the newly depressed key written in the register ANCHR is the register nR. It is determined whether or not it is equal to the loop number n written in If the number of loops is not equal to n, go to step I-6.

I-4 If the channel number ANCH of the newly depressed key is equal to the loop number n in the judgment in step I-3, "1" is added to the loop number n and the new loop number n is written in the register nR. .

This step I-4 is for avoiding the contradiction that the partner of the repeated hit detection is the own tone generation channel.

In the present embodiment stored in the I-5 ROM 21B, 16
The number N of musical tone generation channels is compared with the number n of loops written in the register nR, and if the number n of loops is not large, the process returns to step I-3 and repeats. The routine ends because there is no corresponding continuous hit point.

I-6 If the number of loops n is not equal to the channel number ANCH of the new key depression in the determination in step I-3,
Key code ANK of new key pressed written in register ANKCR
C and the key code KYC written in the register KYCR of the musical tone allocation channel for the musical tone generation channel of the channel number corresponding to the loop number n written in the register nR.
It is determined whether or not Key code for new key press AN
If KC and the key code KYC are not the same, go to step I-4 to search for another.

I-7 If the key code ANKC and key code KYC of the newly depressed key are the same in the judgment in step I-6, the register RDMPR of the tone allocation channel for the tone generation channel of the channel number corresponding to the loop number n It is determined whether or not the silencing processing flag RDMP written in “1” is “1” indicating that the silencing processing is in progress. Silent processing flag RDMP
Indicates "1" and the mute processing is being performed, the mute processing has been performed for the same key repeated hit processing by the previous repeated hit detection, so the process goes to step I-4 to search for another.
When the key is depressed in step H, it is assigned to a tone generation channel, and when the rapid deceleration process is performed in ii), the same process is performed when the rapid decay process is in progress.

I-8 If the mute processing flag RDMP indicates "0" in the judgment in step I-7 and the mute processing is not in progress, the previous key depression of the same key (hereinafter referred to as the previous key depression) having an appropriate repeated hitting relationship In the register AOCHR), the loop number n representing the channel number of the old key depression is written in the register AOCHR as the channel number AOCH of the old key depression. Further, the muffling process start flag DMPF is set to "1" indicating that the muffling process is started.
And write to register DMPFR.

In the above-mentioned repeated hit detection routine, in short, the same key is already assigned to one of the tone generation channels, and the tone generation channel that is in the active tone generation is searched, and the channel number of that tone generation channel is written to the register AOCHR. The mute processing start flag DMPF is set to "1" indicating the mute start processing. Therefore, even if the same key is used, the musical tone generation channel for which the mute processing has already been instructed to start in the repeated hitting processing, which has a short sounding duration, is excluded. It should be noted that since the repeated hits are detected as described above for each new key depression, only one musical tone generation channel is being generated during effective sound generation.

Next, the muffling process target detection and instruction routine (step J) will be described step by step with reference to FIG.

It is determined whether or not the silencing process start flag DMPF written in the J-1 register DMPFR is "1" indicating the silencing process start. When the muffling process start flag DMPF indicates "0" and the muffling process has not been started, the routine ends.

J-2 When the mute processing start flag DMPF indicates "1" and the mute processing is started in the judgment at step J-1, the envelope waveform is simulated as follows to calculate the remaining tone generation amount. To do. Incidentally, simulation of the envelope waveform, the key code BKYC (KYC) from ROM21B table, touch response data BKTD (KTD) and rate RT _j group and breakpoints required to generate a predetermined envelope waveform based on operator data MNPh The level LBP _j group is calculated and created, the generation operation of the envelope waveform is simulated at high speed, and the envelope level LEV at an actual predetermined time interval (for example, 1 msec) is sequentially calculated in a short time.

1) Volume WNL to be sounded by the newly pressed key Simulates the envelope waveform of the newly pressed key, and after the envelope waveform of the newly pressed key reaches the sustain section S, the pronunciation by the newly pressed key ends. To the register WNL according to the following formula that cumulatively adds the envelope level LEV for a predetermined time interval up to
Obtain the pronunciation WNL of the new keypress written in R.

2) Volume to be played by the old key press WOL Simulating the envelope waveform of the old key press, and the predetermined time from when the envelope waveform of the new key press reaches the sustain section S until the old key press finishes sounding. The tone generation amount WOL of the old key depression written in the register WOLR is obtained by the following equation for cumulatively adding the envelope level LEV of the interval.

Envelope level at V (t) t LEV T1 Time from start of new key press to end of decay waveform D of envelope waveform T2 From start of old key press to start of new key press Time TN-END Time from the start of sounding a new key press to the end of the release section R of the envelope waveform TO-END Time from the start of sounding of the old key press to the end of the release section R of the envelope waveform.

(These times T1, TN-END, TO-END are obtained by simulating the envelope waveform, and time T2 is the start of sounding that is written in the corresponding register TSTR that indicates the elapsed time from the start of sounding the old key. It is obtained by reading the current value of the elapsed time timer TST.) J-3 Sound volume of new keypress written in register WNLR
WNL is the pronunciation amount WOL of the old keypress written in the register WOLR.
Determine if it is greater.

J-4 If the pronunciation amount WNL of the new key depression is large in the judgment in step J-3, the old key depression channel number AOCH written in the register AOCHR is set in the register WDCHR as the mute processing target channel number WDCH. Write.

J-5 If the pronunciation amount WNL of the new key depression is not large in the judgment in step J-3, the channel number ANCH of the new key depression written in the register ANCHR is set in the register WDCHR as the mute processing target channel number WDCH. Write.

J-6 Mute processing request flag DMPQ and mute processing flag of the envelope waveform generation channel corresponding to the mute target channel number WDCH written in register WDCHR
Set RDMP to "1", which indicates that the mute processing request or mute processing is in progress.

In the above-described detection and instruction routines for the sound deadening processing target, in short, the attack portion A of the envelope waveform gives an attack feeling, and the sustain portion S and subsequent parts mainly contribute to the volume feeling.
After the time when the envelope waveform of the new key press should reach the sustain section S, the expected sound generation amount until the sound generation of both the old and new key generation is finished is calculated, and the sound generation amount is prioritized by the key generation with the larger sound generation amount. It is intended to release the tone generation channel to which a small number of key depressions are assigned. It should be noted that the attenuation rate used for the sound deadening process is set to a value gentler than the rapid attenuation rate in allocating the tone generation channel so that the sound is naturally connected.

If the old key press has not finished the attack portion A, it corresponds to the pronunciation amounts WNL and WOL of both the old and new key presses and the touch response data KTD-attack level LATK conversion graph shown in FIG. 6, for example. Since the ratio with the attack level LATK converted by the conversion table stored in advance in the ROM 21B is almost the same, the attack level LATK may be compared instead of comparing the sound production amounts WNL and WOL.

Furthermore, if the attack level LATK of the new key press is larger than the attack level LATK of the old key press, the pronunciation amount WNL of the new key press is always larger than the pronunciation amount WOL of the old key press, so first compare the attack level LATK, If the attack level LATK of the new key depression is larger than the attack level LATK of the old key depression, the pronunciation amount WNL of the new key depression is large, and conversely, if the attack level LATK of the new key depression is not large, the steps J-2 and J are performed. The process may be performed by -3.

According to the present embodiment, the same key is repeatedly struck and the pronunciation amount of a new key press.
When WNL is large, it is pronounced as shown in FIG. 7, and when the pronunciation amount WOL of the old key is large, it is pronounced as shown in FIG. Sounds and new sounds are naturally connected.
The rectangular waveforms shown in the lower part of each of FIGS. 6 and 7 show the key press / release states of the old key press and the new key press for the same key.

Next, modifications of the embodiment will be sequentially described.

-Embodiment 1- A modified example for simplifying and speeding up the processing will be described.

After the sustain portion S of the envelope waveform, the pronunciation amounts WNL.WOL of the new key depression and the old key depression of the same key have a close correlation with the envelope level LEV. In other words, with respect to the same key, the change value of the envelope per predetermined time in each envelope waveform after the sustain portion S is almost the same, and it is considered that the envelope waveforms after the sustain portion S have similar shapes. Therefore, the envelope level LEV after the sustain section S
Based on this, the values WNL ′, WOL ′ corresponding to the pronunciation amount can be obtained as follows.

1) A value corresponding to the volume that should be sounded by the newly pressed key WNL ′ The time when the envelope waveform of the new key is simulated and the envelope waveform reaches the sustain section S from the start of sounding the new key. Or the time when a predetermined time has passed from that time
The envelope level LEV at the time point of T3 is obtained, and this envelope level LEV is set to the value WNL ′ corresponding to the sound generation amount of the new key depression.

2) Value WO corresponding to the volume that should be produced by pressing the old key
L ′ Simulates the envelope waveform of the old key depression, and the time when the envelope waveform of the new key depression reaches the sustain section S or the corresponding time of the old key depression after a predetermined time has elapsed from that time.
The envelope level LEV at the time point of T2 + T3 is obtained, and this envelope level LEV is set to the value WOL 'corresponding to the pronunciation amount of the old key depression.

-Variation 2-For example, a musical tone to be produced is mainly composed of an initial portion (attack portion A and decay portion D) consisting of a hammer sound and a string-striking sound with a high frequency component immediately after keystroke, such as a piano sound. In the case where it is composed of the constituent sound A and the second constituent sound B which is mainly composed of the continuous portion (the sustain portion S and the release portion R), which is subsequent to the initial portion and is composed of the string sound with less high-frequency component and less timbre change, and which gives a volume feeling. A modified example will be described.

In the envelope waveform diagram of FIG. 9, the envelope waveform of the generated musical tone is a synthetic resin shape of the envelope waveform of the first component sound A and the envelope waveform of the second component sound B. As is apparent from FIG. 10 in which each of these envelope waveforms is expressed in a logarithmic expression, regarding the second constituent sound B, the deformation values of the envelope per predetermined time in each envelope waveform after the decay portion D are substantially the same, With the same key, the envelope waveforms of the second component sound B after the decay portion D are regarded as having similar shapes. It should be noted that the envelope level of the second component sound B after the decay portion D, particularly in the decay portion D, can be obtained from the envelope level LEV of the musical tone (synthesized sound) to be produced, by a conversion table or the like.

Therefore, as in the first modification, as described above, the second
Values WNL 'and WOL' corresponding to the sound generation amounts of the old and new key depressions are obtained from the envelope level of the constituent tone B after the decay portion D.

Second Example: The tones to be sounded are composed of the first constituent sound A and the second constituent sound B as described above, but unlike the above-mentioned case, the first constituent sound A and the second constituent sound B respectively. Another embodiment in which a tone signal is generated by a different tone generation channel will be described. It should be noted that only the parts that are different from the first embodiment will be described, and the description of the overlapping parts will be omitted.

In the present embodiment, the tone generating circuit 29 is arranged from the first to
It consists of 32 musical tone generation channels up to the 32nd, and the 1st and 2nd, 3rd and 4th, ..., 31st and 32nd, respectively, form the set which produces the desired musical tone. To do.
Further, the first constituent sound A is assigned to the even-numbered tone generation channels.
However, the second component sound B is assigned to the odd-numbered tone generation channels to generate a tone signal. In this way, basically, based on the second constituent sound B, repeated hits are detected by searching the tone generation channel during the effective sounding of the same key, and detection and instruction of the sound deadening processing target are performed. The tone generation channel to which the key with a small amount of sound generation is assigned is released by giving priority to the key with a large amount of keys.

Next, based on the flowcharts of FIGS. 3 to 5 described in the first embodiment, the main differences other than those described above will be described.

The tone generation channels are allocated in step H by tone generation channels in units of groups. First, the tone generation channels released from the key status flag KYS and the envelope end flag EV "-END" of each tone generation channel forming the group. The key combination is detected and the new key is assigned. Further, when the set of released tone generation channels is not detected, based on the envelope level corresponding to each second component sound B and the attack end flag of each first component sound A and second component sound B. The attack portion A is terminated during sound generation, the set of tone generation channels having the smallest envelope level is detected, and new key depression is performed. The number of the tone generation channel to which the second component sound B is assigned by the assignment is written in the register ANCHR.

In step I-4, the tone generation channel 2
Since the number is set to one, "2" is added to the loop number n and written in the register nR. Further, for the same reason, in step I-5, the number N of musical sound generation channels is increased.
Is subtracted from "1", in other words, the number obtained by subtracting "1" from 32 is compared with the loop number n in this embodiment.

At step J-5, the second key B of the new key press.
Therefore, the second constituent sound B of the new key depression is not sounded. In other words, the register RT corresponding to the channel number ANCH of the new keypress written in the register ANCHR.
_j R, LBP _j R, EV-ATR, EV-DKR, EV-END contents are all cleared. Note that the muffling process for the first component sound A may not be performed. Others are substantially the same as those in the first embodiment, except that the first constituent sound A of the old key depression is muted together with the second constituent sound B.

11 and 12 are waveform charts showing envelopes of musical sound according to the present embodiment corresponding to FIGS. 7 and 8 in the first embodiment.

In the present embodiment, the tone generation channels are formed into a set, and the first component sound A and the second component sound B are assigned to the tone generation channels of this set. However, the tone generation to which the first component sound A is assigned is generated. As is clear from FIG. 10, the channels are released earlier than the tone generation channels to which the second constituent sound B is assigned. Therefore, if the assignment process is performed individually without forming a set, the tone generation channels will be effective. Available. In the present embodiment, it is needless to say that the first and second modified examples of the first embodiment can be applied with a slight modification, though the description thereof is omitted.

Next, a modification of this embodiment will be described.

-Modification 1-In order to further enrich the timbre change of the continuous portion of the musical tone to be sounded, when the continuous portion is composed of a plurality of second constituent sounds B, as shown in FIG. 13, for example. The second constituent sound B1 which has a high frequency high frequency component of the keystroke and has a relatively short envelope, and the second constituent sound B2 which has a low frequency high frequency component of the keystroke and a relatively long envelope.
A modified example in the case of being composed of

In the present modification, the tone generation circuit 29 is arranged from the first to
It consists of 48 tone generation channels up to the 48th, from the 1st to the 3rd, from the 4th to the 6th,
.., from the 46th to the 48th, form a set in which the desired musical tones are pronounced. The second component sound B1 is assigned to the first, fourth, ..., The second component sound B2 is assigned to the second, fifth, ..., The second component sound B2 is assigned to the third, sixth ,. One constituent sound A is assigned to generate a musical tone signal. In this way, based on the second constituent sound B2 having a relatively long envelope sound, repeated hits are detected by searching the tone generation channel during the effective sounding of the same key, and the detection and instruction of the mute processing target are performed. This is performed based on the sum of the pronunciation amounts of the two component sounds B1 and B2, and the tone generation channel to which the key with the smaller amount of sound is assigned is released by giving priority to the key with the larger amount of sound. Other than that, it is basically the same as this embodiment described above. The waveform diagram showing the envelope of the pronunciation of the musical sound corresponding to FIGS. 7 and 8 of the first embodiment according to the present embodiment is shown in FIG.
FIG. 14 and FIG. 15.

-Modification 2- In order to obtain a timbre change in the continuous portion and to reduce the number of constituent tones, the musical tones to be generated are not the first and second constituent tones A and B as described above. A modified example in which the sound of the initial part and the sound of the continuous part are included in different ratios will be described.

The tone sounded as shown in FIG. 16 has the same tone quality as the first component sound A'which mainly constitutes the initial part of the round key-sensitive keystroke with few high-frequency components, whose sound quality does not change much depending on the strength of the touch. , A second constituent sound B'which mainly produces a continuous portion of a hard key depression that is more pronounced when a strong touch is made and, in the case of a piano, contains a large amount of high frequency components. Incidentally, the touch response data KTD-attack level LATK conversion table showing the relationship between the touch response data KTD and the attack level LATK corresponding to FIG. 6 described above is as shown in FIG. Therefore, when the key is depressed, the second constituent sound B'is not sounded and is dominated by the first constituent sound A '.

It should be noted that the first constituent sound A ′ is used for the detection and instruction of the sound deadening target.
And the second constituent sound B ', the same processing as in the first embodiment is performed on the basis of the sum of the sound generation quantities, and the sound generation quantity of the new key depression is performed.
When WNL is not larger than the sound generation amount WOL of the old key depression, both the first constituent sound A ′ and the second constituent sound B ′ are subjected to the muffling process.

Third Embodiment: Next, in order to effectively use the tone generation channel, when the pronunciation amount WNL of the new key depression is larger than the pronunciation amount WOL of the old key depression, the old key depression is replaced with the new key depression. An embodiment for producing a musical sound will be described based on FIGS. 18 to 20. In addition,
The same reference numerals as those in the first embodiment indicate the same contents,
In particular, only the parts different from the first embodiment will be described, and the description of the overlapping parts will be omitted.

The flow chart of the basic program shown in FIG. 18 in this embodiment is different from the flow chart of the basic program shown in FIG. 3 of the first embodiment in steps A to G and step K. The steps are the same corresponding to the steps A to G and K of the first embodiment, and the steps H ′, H ″, step I ′, step L and step J ′ are as follows.

If the key state flag BKYS indicates "1" in the key depression state in the judgment of H'step G, the register BKENR
"1" is subtracted from the number of pending keys BKEN written in the register BKENR and the number after the subtraction is written in the register BKENR as the number of new pending keys BKEN.

I'Continuous hit detection routine. Details will be described later with reference to the flowchart shown in FIG.

It is judged whether the mute processing start flag DMPF written in the L register DMPFR is "1" indicating the start of the mute processing or not, and continuous hits are judged. If the muffling process start flag DMPF is "0" and the muffling process start is not indicated, and if the consecutive hits are not made, the process proceeds to step H ".

J'When the mute processing start flag DMPF is "1" in the judgment in step L and the sound mute processing start is indicated and there is continuous hitting, the mute processing target detection and instruction routine is entered. Details of this mute processing target detection and instruction routine will be described later with reference to the flowchart shown in FIG. When the mute processing target detection and instruction routine ends, the process returns to step E.

H ″ If the mute processing start flag DMPF is “0” and the mute processing start is not indicated in the judgment at the step L, and continuous hitting is not performed, allocation to the tone generation channel is performed at the step H of the first embodiment. (The following description including parentheses in Step H),
Return to step E.

Next, regarding the continuous hit detection routine (step I ′),
This will be described with reference to FIG.

The continuous hit detection routine in this embodiment is different from the continuous hit detection routine shown in FIG. 4 of the first embodiment in steps I'-1, I'-2 and I'-.
4, step I'-5, step I'-7 and step I'-8 are steps I-1, I-2, I-4, I of the first embodiment.
-5, I-7 and I-8 are the same, step I-6 is deleted and step I'-9 is as follows.

I'-9 If the damper state flag FCDS indicates "1" and the dump processing is prohibited in the judgment in step I'-2, the key code BKYC of the new key pressed written in the register BKYCR is detected. , It is determined whether or not it is the same as the key code KYC written in the register KYCR of the sound generation channel of the channel number corresponding to the loop number n written in the register nR. The judgment based on the key code BKYC of the new key depression is made in step I- in the first embodiment.
This is because unlike 6 the channel number of the new key press has not been determined yet. Others are the same as the continuous hit detection routine of the first embodiment.

Next, the mute processing target detection and instruction routine (step J ') will be described with reference to FIG.

The noise reduction processing target detection and instruction routine in this embodiment is different from the noise reduction processing target detection and instruction routine shown in FIG. 5 of the first embodiment in step J ′.
-1 to step J'-3 are the steps J of the first embodiment.
-1 to J-3 are the same, and step J-
6 is deleted and steps J'-4 and J'-
5 is as follows.

J'-4 When the tone generation amount WNL of the new key depression is large in the judgment in step J'-3 (J-3), the allocation to the tone generation channel performed in step H is performed for the new key depression. To do. At this time, the tone generation channel to which the old key depression is assigned is preferentially assigned. The tone generation channel of the old key depression may be preferentially assigned only when there is no tone generation channel released, or the tone connection may be preferentially assigned to another tone generation tone generation channel. . When assigning to another tone generation channel, the channel number AOCH of the old key depression written in the register AOCHR is written in the register WDCHR as the mute processing target channel number WDCH. Then, the mute processing target channel number W written in the register WDCHR
The mute processing request flag DMPQ and the mute processing flag RDMP of the envelope waveform generation channel corresponding to the DCH are set to “1” indicating the mute processing request or the mute processing.

J'-5 If the tone generation amount WNL of the new key depression is not large in the judgment in the step J'-3 (J-3), the old key depression is prevented from being assigned to the assigned tone generation channel. The assignment to the musical tone generation channel at H is performed. Therefore, if the released tone generation channel is not detected and the tone generation channel of the old key depression is selected, it is assigned to the tone generation channel of the next rank. Regarding the mute instruction for a new key depression, the mute processing request flag DMPQ and the mute processing flag RDMP are set to "1" instead of "0" at the time of allocation. Others are the same as the detection and instruction routine of the sound deadening process target of the first embodiment.

It is needless to say that each modification of the first embodiment can be applied to the present embodiment as well, and in the case of the second embodiment, the features of the present embodiment can be adopted without deteriorating by appropriately modifying.

Next, a case where the tone generation channel is further effectively used by applying it to the second modification of the second embodiment will be described.

When the pronunciation amount WNL of the new key depression is not larger than the pronunciation amount WOL of the old key depression, the second component sound B'is not assigned and the first component sound A'is changed to the first component sound A of the old key depression. 'Is assigned to the tone generation channel to which it is assigned. The first constituent sound A'and the second
It is advisable to perform assignment processing individually without forming a set with the constituent sound B '.

If the touch of the new key is weak, the second component sound B'is not replaced as shown in FIG. 21, while the first component sound A'is not replaced.
Are replaced with each other and an attack feeling is obtained, but the envelope waveform becomes as shown by the dotted line X, and becomes stepwise.
In order to solve this, when assigning the first constituent sound A ′, if the attack level LATK of a new key depression is smaller than the envelope level of the old key depression at the time of a new key depression at the time of continuous tapping, a new key depression is performed. The attack level LATK may be changed to the envelope level of the old key depression as shown by the solid line Y.

Note that all the registers used in each of the embodiments and the like are provided by an area virtually assigned to the RAM 21C of the microcomputer 21 as described above.

Further, in each of the embodiments, the electronic keyboard musical instrument has been described, but other electronic musical instruments of attenuating sound system other than the electronic keyboard musical instrument, for example, an electronic drum that continuously strikes the same sounding body (film, striking surface, etc.) Needless to say, the present invention can also be applied to. In this case, the same key (same string) in each embodiment may be read as the same sounding body, and the unique processing of other electronic musical instruments may be changed accordingly.

[Brief description of drawings]

FIG. 1 is a block diagram corresponding to the configuration of the present invention described in the claims, and FIGS. 2 to 10 are drawings for explaining a first embodiment of an electronic musical instrument according to the present invention. FIG. 2 is a schematic diagram, FIGS. 3 to 5 are flowcharts of a basic program of a program executed by a microcomputer, a repeated hit detection routine, and a mute processing target detection and instruction routine, and FIG. Touch response data attack level conversion graph for the embodiment, FIG. 7 and FIG. 8 are waveform charts showing the envelope of the pronunciation of the musical sound processed according to the flowcharts of FIG. 3 to FIG. 5, FIG. FIG. 10 is an envelope waveform diagram relating to the second modification of the present embodiment, and FIGS. 11 to 17 are drawings for explaining the second embodiment of the electronic musical instrument according to the present embodiment. FIG. 11 and FIG. 12 are waveform diagrams showing envelopes of the pronunciation of musical tones corresponding to FIGS. 7 and 8 of the first embodiment, respectively, and FIG. 13 is a first constituent sound according to the modified example 1 of the present embodiment. A, envelope waveform diagrams of the second component sound B1 and the second component sound B2, FIG. 14 and FIG. 15 respectively correspond to FIG. 7 and FIG. 8 of the first embodiment relating to the modified example 1 of the present embodiment. FIG. 16 is a waveform diagram showing the envelope of a musical sound, and FIG. 16 and FIG. 17 are envelope waveforms of the first constituent sound A ′ and the second constituent sound B ′ in the modification 3 of the present embodiment, respectively. Touch response data / attack level conversion graphs corresponding to FIG. 6, FIGS. 18 to 21 are drawings for explaining a third embodiment of the electronic musical instrument according to the present invention, and FIGS. FIG. 5 is a flow chart corresponding to each of them, and the present embodiment Is a waveform diagram showing the envelope of the sound of a musical tone corresponding to Figure 8 of the first embodiment when applied to a second modification of the embodiment. 20 ... keyboard, 21 ... microcomputer 21A ... central processing unit 21B ... read-only memory 21C ... writable memory 21D ... timer circuit, 22 ... key operation detection circuit 23 ... bus 24 ... touch response detection circuit 25 ... operator group, 26 ... Operator detection circuit 27 ... Pedal group, 28 ... Pedal detection circuit 29 ... Musical sound generation circuit, 30 ... Magnifier 31 ... Speaker

Claims (1)

[Claims] 1. An electronic musical instrument which processes repeated hits of the same key such that the same key is pressed again so that the sound corresponding to the sound of a key released once is overlapped. A continuous hit detecting means for detecting whether or not a musical tone having the same key as the depressed key has already been assigned to the musical tone generating channel, (b) a first sound generation amount of the musical tone to be generated by the new key depression. Alternatively, the second value of the tone corresponding to the value corresponding to the sound amount and the time point to be sounded by the new key depression by the same key already assigned to the tone generation channel.
And (c) the first and second sounding amounts or the values corresponding to these sounding amounts are compared with each other, and consecutive hitting is performed by the consecutive hitting detecting means. An electronic musical instrument characterized by comprising a comparison / selection means for selecting a side having a large amount of sound emission or a value corresponding to the amount of sound generation and giving priority to sound when detected.