JPH08123411A - Musical sound synthesizing device - Google Patents

Abstract

PURPOSE: To obtain faithful sustain pedal effects. CONSTITUTION: A closed loop consisting of a delay means 1, a filter 2, and a coefficient multiplier 3 is driven by supplying an excitation waveform to an adder 5. A KON signal which is set to '0' at key-OFF time and '1' at key-ON time and a dump signal DUMP which corresponds to the manipulated variable of a sustain pedal are inputted to a maximum value selecting means 4 and a larger signal is selected and applied as a multiplication coefficient M to the coefficient multiplier 3. Consequently, a musical sound having attenuation characteristics corresponding to the dump signal DUMP can be obtained from the closed loop.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to dump control, and is particularly suitable for application to a musical sound synthesizer having a damping system algorithm such as a piano string model.

[0002]

2. Description of the Related Art Conventionally, in a piano, a damper is in contact with a string when the keyboard is not operated, and in this state, vibration of the string is suppressed. When the keyboard is keyed on, the damper is struck away from the string corresponding to the keyon, and a musical sound is generated from the string. Further, when the key is keyed off, the damper comes into contact with the string again, the vibration of the string is rapidly suppressed, and the sound is damped (dumped). Further, when the sustain pedal is operated, the damper is separated from the string and the string vibration is allowed.

That is, when the sustain pedal is operated and sustain is instructed, the vibration of the sounding string is allowed. Therefore, even if the key is turned off, the mute of the sounding string is delayed and the musical sound is naturally attenuated. I will do it. In addition, strings other than the struck strings are also resonated and vibrated through the frame and the main body, so that some sounds are produced. Further, there is a performance mode in which the performer does not fully push in the sustain pedal and operates the damper so that the damper slightly contacts the strings. In this way, the sound quality of the piano to be played changes greatly depending on the operation mode of the sustain pedal.

[0004]

However, the conventional tone synthesizer has a problem that the dump control is unique and the half pedal state cannot be faithfully realized. Therefore, an object of the present invention is to provide a musical sound synthesizer capable of faithfully realizing a sustain pedal effect such as a half pedal by arbitrarily controlling a dump.

[0005]

In order to achieve the above-mentioned object, a musical tone synthesizing apparatus of the present invention comprises a musical tone generating means for generating a musical tone, and a damping for controlling a decay state of the musical tone generated by the musical tone generating means. The decay state control means includes state control means, first performance information generation means for instructing the generation and termination of the musical sound, and second performance information generation means for instructing a decay mode of the musical sound. In addition, a control signal corresponding to both the values of the performance information generated by the first performance information generating means and the performance information generated by the second performance information generating means is supplied.

Further, the musical sound synthesizer of the present invention comprises an excitation signal generating means for generating an excitation signal for driving a closed loop for generating a musical sound, a loop gain control means for controlling a loop gain of the closed loop, and a generation of the musical sound. And performance information generated by the first performance information generating means, the first performance information generating means for instructing the end of the musical sound and the second performance information generating means for instructing the attenuation mode of the music. And the loop gain is determined according to both values of the performance information generated by the second performance information generating means. Further, in the musical sound synthesizer, the performance information generated by the first performance information generating means is compared with the performance information generated by the second performance information generating means, and based on the comparison result, The performance information may be selected and supplied to the loop gain control means.

[0007]

According to the present invention, the attenuation state of the musical tone is controlled by the attenuation state control means, and more specifically, the loop gain of the closed loop is controlled by the control signal from the sustain pedal or the like. The faithful pedal effect of the sustain pedal can be realized.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the schematic construction of an embodiment of the musical tone synthesizer of the present invention. This musical tone synthesizer is a stringed instrument.
It is composed only of strings. This schematic configuration will be described with reference to FIGS. 1 and 2. In FIG. 1, a delay means (Delay) 1 and a filter (F
ilter) 2 and a coefficient multiplier (MPY) 3 that represents reflection / attenuation at the frame, and this closed loop is driven by supplying an excitation waveform to the adder circuit 5. . A multiplication coefficient M that mainly determines the damping characteristic (dump characteristic) of the closed loop is applied to the coefficient multiplier 3, and this multiplication coefficient M is output from the maximum value selecting means 4. The maximum value selection means 4
A key-on signal KON and a dump signal DUMP are input to the input terminal, and a signal having a larger value among them is selected by the maximum value selection means 4 and output. Since the high frequency component of the musical sound of the stringed instrument is attenuated as the musical sound is repeatedly reflected, the filter 1 is generally composed of a low-pass filter.

In the tone synthesizer thus constructed, the pitch of the tone output from the closed loop corresponds to the total delay time of the closed loop, and when the pitch of the tone output is changed, the delay of the delay means 1 is delayed. The time should be controlled. The loop gain of the closed loop is mainly determined by the multiplication coefficient M applied to the coefficient multiplier 3. That is, the attenuation characteristic of the output musical sound is an attenuation characteristic according to the multiplication coefficient M. Therefore, when the coefficient multiplier 3 is controlled according to the key-on signal KON which becomes a coefficient of "0" at the time of key-off and a coefficient of "1" at the time of key-on, the coefficient of the coefficient multiplier 3 is set to "0" at the time of key-off and the loop gain becomes the minimum. Therefore, the tone signal circulating in the closed loop is rapidly attenuated, and the coefficient of the coefficient multiplier 3 is set to "1" at the time of key-on, and the loop gain is maximized.
The musical tone signal circulating in the closed loop is circulated while being attenuated mainly according to the characteristics of the filter 2.

When the coefficient multiplier 3 is controlled according to the dump signal DUMP, the dump signal DUMP is, for example, a signal corresponding to the amount of depression of the sustain pedal and is a signal indicating the degree to which the string is released. The size is "0
Since an arbitrary value of "1.0" can be taken, the attenuation characteristic of the tone signal can be an arbitrary attenuation characteristic. For example, when the dump signal DUNP having a magnitude of about a half pedal is applied to the coefficient multiplier 3, the tone signal circulating in the closed loop is attenuated by the attenuation characteristic of the half pedal. Therefore, by setting the signal of the sustain pedal to the dump signal DUMP, the faithful pedal effect of the sustain pedal such as the half pedal can be realized.

When the key is turned on while the sustain pedal is being operated in the actual piano, the damper of the keyed string is completely separated from the string and the string is released regardless of the operation amount of the sustain pedal. Therefore, if the maximum value of the key-on signal KON and the dump signal DUMP, which has the maximum coefficient of "1" at the time of key-on, is applied to the coefficient multiplier 3 so as to be selected by the maximum value selection means 4, this is simulated. be able to. The operation of the maximum value selecting means 4 will be described with reference to FIG. 2. First, the dump signal DUMP is turned on at time t1 and turned off at time t4 as shown in FIG. The key-on signal KON is turned on at a time point t2 after the time point t1 as shown in FIG.
Keyed off at 3. However, the DUNP signal is set to "0.5" corresponding to a half pedal.

Then, the maximum value selecting means 4 selects and outputs "0.5" of the DUNP signal from the time t1 to the key-on from time t2, as shown by the multiplication coefficient M in FIG. The key-on signal KON of "1.0" is selected and output from the time t2 to the key-off from the time t3, and the DUNP signal is selected and output again from the time t3 to the time t4. As a result, the strings are somewhat released from the time t1 to the time t2 as shown in (d) of the figure, so that, for example, a musical tone signal that resonates with the vibration of the other struck strings is output from the closed loop. And
Between the time points t2 and t3, the excitation waveform supplied with the key-on circulates in the closed loop while naturally attenuating mainly according to the attenuation characteristic of the filter 2, so that the generated tone signal is output from the closed loop. And from time t3 to t4
Until then, the tone signal circulating in the closed loop is damped by the attenuation characteristic of the half pedal. Then, at time t4, the M value becomes "0", so that the musical sound in the closed loop is rapidly dumped.

In order to output a musical tone that resonates with the vibration of another string from the closed loop, it is necessary to supply the signal of the other string to the closed loop. For example, it is necessary to apply the signal of the other string to the adder 5. You can do it. When the half pedal is used as described above, the damping characteristic mainly between the times t3 and t4 is controlled so that the half pedal effect is obtained. If the pedal operation amount is arbitrary, the dump characteristic between the time points can be controlled according to the operation amount according to the DUNP signal which is the operation amount of the sustain pedal.

Next, a detailed block diagram of the tone synthesizer shown in FIG. 1 is shown in FIG. 3, but blocks having the same function are assigned the same reference numerals and their explanations are omitted. In this figure, 6 is an excitation waveform generating section for generating an excitation waveform (DW) when a performance operation section (for example, a keyboard) is keyed on, 7
Is a smoothing filter composed of a low-pass filter for smoothing the abrupt change of the multiplication coefficient M output from the maximum value selection unit 4, and 8 is output from a closed loop composed of the filter 1, the delay unit 2 and the coefficient multiplier 3. A tone color setting unit for setting the tone color of a tone, and a control unit 9 for controlling the entire tone signal synthesizer so that a tone having a predetermined pitch and having a set tone color and corresponding to the operation amount of the sustain pedal is output. And the sampling clock φ _S is supplied to each part.

Further, 10 is a performance operator such as a keyboard, 11 is a performance assisting operator such as a sustain pedal, and 12 is a coefficient supply capable of supplying the filter 1 with a coefficient IM and a filter coefficient FC corresponding to the tone color parameter TC. Part 13 is the coefficient IM output from the smoothing filter 7, and the adjustment coefficient Ad
A multiplier 14 which adjusts the coefficient IM by multiplying j, and 14 is an other-string signal generator that generates a signal of another string and supplies it to the adder 5. The adjustment coefficient Adj is supplied from the control unit 9.

In the musical tone signal synthesizer having the above-described structure, when the performance operator 10 such as a keyboard is not operated and the coefficient of the coefficient multiplier (MPY) 3 in the closed loop is substantially "0", When the performance assist operator 11 such as the sustain pedal is operated, the operation amount parameter SUSTA
IN is supplied to the control unit 9, and the dump signal DUMP output based on this parameter in the control unit 9 is input to the maximum value selection unit 4. In this case, the dump signal DUM
P is an arbitrary coefficient between "0 and 1", but when the performance assisting operator 11 is operated, it becomes a coefficient value exceeding "0", and when the key-on signal KON is not key-on, the coefficient is set. Becomes "0", the comparator 4-2 in the maximum value selection unit 4 judges that A <B and gives a selection signal to the selector 4-1. Then, the selector 4-1 selects and outputs the dump signal DUMP as the multiplication coefficient M. The output multiplication coefficient M is smoothed by the smoothing filter 7 and input to the multiplier 13.

The smoothing filter 7 is a delay circuit (DR).
7-1 and the adder 7-2 that adds the multiplication coefficient M delayed by one sampling in the delay circuit 7-1 and the currently input multiplication coefficient M, and the output added by the adder 7-2. It is composed of a multiplier 7-3 for halving. That is, the smoothing filter 7 averages and outputs as the coefficient IM, and the multiplier 13 further outputs the adjustment coefficient Adj.
Is applied to the coefficient multiplier 3 as a dump control parameter DC to control the loop gain of the closed loop, that is, the damping characteristic. In this case, the other string signal generation unit 14
When the signal of the other string is input to the adder 5, the signal of the other string is added to the coefficient multiplier 3 via the filter 1,
While this signal is multiplied by the dump control parameter DC which is an arbitrary value of "0 to 1", the signal of the other string circulates in the closed loop.

In this way, the signal of the other string circulates in the closed loop while being attenuated, and this musical tone signal is output from the closed loop, so that the resonance effect due to the vibration of the other string in the piano can be simulated. Although the tone color of the musical tone output from the closed loop is set by the tone color setting unit 8, the tone color parameter TC output from the tone color setting unit 8 is supplied to the control unit 9, and the control unit 9 is based on this parameter TC. By applying the filter parameter FC to the filter 1, the tone of the set tone color is output from the closed loop. Further, the coefficient supply unit 12 is provided so that the tone color parameter TC is controlled by the coefficient IM output from the smoothing filter 7 so that the tone color is changed depending on the contact state of the damper according to the operation of the sustain pedal. Then, the filter parameter FC obtained from the coefficient supply unit 12 may be supplied to the filter 1.

When the performance operator 10 is operated to be keyed on, a key indicating the timing of the key-on is displayed.
The on signal is applied to the excitation waveform generator 6, and the generated excitation waveform (DW) is supplied to the closed loop via the adder 5, whereby the closed loop is driven. Further, a KEY signal indicating the key operated by the performance operator 10 is supplied to the control unit 9, and the control unit 9 determines the delay time control parameter DLY of the delay unit 2 based on this signal and applies it to the delay unit 2. To do. As a result, the total delay time in the closed loop is set to the delay time at which the musical sound of the pitch of the operated key can be output. Further, the key-on signal KON set to “1” is output to the maximum value selection unit 4. The maximum value selection unit 4 receives a dump signal DUMP corresponding to the operation of the performance operation auxiliary operator 11 such as a sustain pedal from the control unit 9 to the maximum value selection unit 4, but the maximum dump signal DUMP is "1". Therefore, the key-on signal KON is normally set to a larger value.

Therefore, the comparator 4-2 of the maximum value selection unit 4 does not give the selection signal to the selector 4-1. Therefore, the selector 4-1 selects the key-on signal KON and outputs it as the multiplication coefficient M. The multiplication coefficient M is set as a control parameter DC via the smoothing filter 7 and the multiplier 13 and applied to the coefficient multiplier 3. In this case, since the control parameter DC has the maximum value, the tone signal driven by the excitation signal DW circulating in the closed loop circulates without being attenuated by the multiplication coefficient unit 3, and mainly the characteristics of the filter 1 are obtained. It will naturally decay accordingly. That is, the musical tone in the sustain state is output from the closed loop.

Further, the control section 9 gives an excitation waveform parameter DP according to, for example, a key pressing speed to the excitation signal generation section 6 to control the waveform of the excitation signal DW. This excitation signal DW does not consist of a single frequency signal, but is assumed to be a signal containing many frequency signals, and the excitation signal generation unit 6 uses the above signals of a waveform memory system, a non-linear circuit, etc. It is composed of a signal generating circuit that can generate the signal. Further, the coefficient multiplier 3 may be omitted by providing the filter 1 with the attenuation characteristic changing function of the coefficient multiplier 3.

The key-on signal KON and the dump signal D
Instead of outputting the largest coefficient of UMP from the maximum value selection unit 4, an OR circuit is used to calculate the logical sum of the key-on signal KON and the dump signal DUMP as shown in FIG. Alternatively, the same figure (b)
The multiplication coefficient M may be generated by adding the key-on signal KON and the dump signal DUMP by using an adder and a limiter as shown in FIG. The input-output characteristic in this case is shown in FIG. 7C. It is common to use a limiter characteristic such as A where the output is “1” when the input is “1”. As shown in B, the output may be saturated at "1" before the input becomes "1".

Next, a modified example of the musical tone synthesizing apparatus of the present invention is shown in FIG. It is also possible to provide a musical sound generating means. In this figure, SG1 to SGn are musical sound generating means, all have the same configuration, and when simulating a piano, it is preferable that they correspond to strings of the piano, respectively. Reference numeral 23 denotes a resonator portion, to which the tone signal output from each of the tone generating means SG1 to SGn is input to generate a tone signal resonated in the resonator portion 23, and the generated tone signal corresponds to each tone generating means. It is returned to SG1 to SGn. The resonance body portion 23 corresponds to a soundboard of a piano, and the musical sound generated by the resonance body portion 23 is output.

Further, since the musical tone generating means SG1 to SGn have the same configuration, only the configuration of SG1 is shown as an example. However, 20 is a dump coefficient generating section, which corresponds to the depression amount of the sustain pedal. Dump signal DUM
P and the key-on signal KON are supplied to multiply the multiplication coefficient M.
Is generated and supplied to the string section (STRING) 22 as a signal for controlling the attenuation. 21 receives the key-on signal KON and generates an excitation signal having a waveform corresponding to each key at the key-on timing to generate the string portion 22.
Is an excitation unit that drives the closed loop of the string unit 22 and may generate an excitation signal according to the speed signal of the key. The dump signal DUMP is used for each SG.
1 to SGn are commonly applied to each SG1 to SGn,
Key-on signal KONKON 1-KON corresponding to each string
n are supplied respectively.

Here, the string portion 22 is constituted by a closed loop as shown in FIG. 1 and FIG.
The configuration may be as shown in. In FIG. 6, the delay means 35 and 38 are divided into two, DELAY1n and DELAY2n, which are filters 33, which represent the characteristics of the strings.
36 is also divided into FILTER1n and FILTER2n. Then, the adder 31 is provided between the delay unit 38 (DELAY2n) and the filter 33 (FILTER1n).
To the delay means 35 (DELAY1n) and the filter 36.
By providing the adder 32 between (FILTER2n) and the excitation waveform Wn generated by the excitation waveform generation unit 21 to the adders 31 and 32, the excitation waveform can be supplied to the string striking position. it can.

Further, a control parameter DC for controlling the attenuation characteristic is applied to the coefficient multiplier 34, and the phase inverting circuit (REF) 37 is for simulating the reflection by the fixed end of the string. Further, the signal STRINGn output from the FILTER 1n is a musical tone signal output from the string unit 22 to the resonator unit 23, and the signal RESINn input to the coefficient multiplier 40 is fed back from the resonator unit 23 to the string unit 22. It is a resonated musical tone signal. The coefficient multiplier 40 is provided to control the dump with a feedback coefficient that determines the amount of signal fed back from the resonator unit 23 to the string unit 22, and the coefficient multiplier 40 includes a resonator feedback coefficient supply unit. From 39, a resonator feedback coefficient corresponding to the tone color parameter TC and the control parameter DC is applied as a multiplication coefficient. In this case, the control parameter DC
Is closer to "1", the feedback coefficient is such that the feedback amount of the resonance sound is increased.

As shown in this figure, the key depression speed signal (KEYVn) of the key of the performance operator 10 is applied to the excitation waveform generator (DRIVERn) 21 to generate an excitation waveform corresponding to this signal. Then, it may be supplied to the string portion 22. Further, FIG. 7 shows an example of the configuration of the resonator body 23. The resonator body 23 is a waveguide (WG) 41, 42, ..., 48, ...
・ Junction (J) 50, 51, 52, 53, ・
. Two-dimensionally configured by connecting as shown in FIG.

The junction 50 of the resonator body 23,
The musical tone signals output from the string portion 22 are input to 52, and the resonated musical tone signals generated by the resonator body portion 23 are connected to the junctions 50, 52 ,.
Is fed back to the string section 22. The tone signal generated by the resonator unit 23 may be taken out from any of the waveguides 41, 42, ..., But in FIG. 7, it is taken out from the WG 45. The waveguides 41, 42, ...
.. has been conventionally known in the art, and has a structure as described in, for example, Japanese Patent Laid-Open No. 63-40199.

As described above, according to the modification, since the resonator portion 23 is provided, it is possible to realize a physical phenomenon in which a string other than the struck string vibrates. In this case, the dump coefficient from the highest note to the key one octave and 5 degrees down is always set to "1.0", and if the dump function is not provided, the resonance of high-pitched strings like an actual piano is realized. You can do (on an actual piano,
There are no dampers on the high-pitched strings and they are in a released state. ).

The present invention is also effective as an input to a damper model by more faithful modeling described in Japanese Patent Laid-Open No. 4-52698. This state is shown in FIG.
In the figure, 01 is a string portion simulating the behavior of a string, and 02 is an exciting portion simulating the movement of a key, that is, the behavior of a hammer striking a string.
In the excitation unit 02, the damping information FD is given to the adder 79. The damping information FD is obtained by converting the pressure applied by the damper to the string into the acceleration of the damper, which is "1".
In the case of, it will be dumped. Therefore, when the control parameter DC is set to the dump state in the case of “0”, the damping information FD is set.
Since the control parameter DC may be subtracted from “+1” in order to obtain
The control parameter DC is subtracted from this to obtain the vibration suppression information FD. As a result, by inputting the control parameter DC into the damper model, it becomes possible to control the dump of the half pedal or the like.

In the above description, the dump control is performed by directly controlling the reflection coefficient of the string, but a simple filter for the damper is inserted in the closed loop and the coefficient is controlled by the control parameter DC. Then, the model can be closer to the actual one. Alternatively, the in-loop filter representing the characteristics of the strings may be controlled by the control parameter DC. Also, the dump signal DUMP is "0.
Although it is set to a continuous value between "1", for example, "0.1 to 0.9"
The range width may be set so that the control is performed within a certain range such as "9". Further, the control parameter DC may be multiplied by an envelope signal that changes with time.

In addition to the control relating to the sustain pedal, the present invention performs the same processing as when the sustain pedal is operated, only for the string model corresponding to the key that was pressed when the sostenuto pedal was operated. If done, it can also be applied to simulate sostenuto pedal functions. In addition, the tone signal synthesizer of the present invention is not limited to hardware and is not limited to DS.
It can also be realized by an arithmetic processing means such as P. Furthermore, in the above description, the musical tone synthesizer that synthesizes the attenuated sound of a string instrument or the like has been described.
The present invention is not limited to this, and may be applied to a musical sound synthesizing device for synthesizing continuous sounds of wind instruments and the like. For example, it becomes possible to perform a mute operation on the trumpet.

[0033]

Since the present invention is configured as described above, the faithful pedal effect of a sustain pedal such as a half pedal can be realized. Furthermore, if signals from other string models are transmitted via a resonance system or the like, it is possible to simulate resonance strings.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a musical tone synthesizing apparatus of the present invention.

FIG. 2 is a timing diagram for explaining FIG.

FIG. 3 is a block diagram showing a detailed configuration of a musical sound synthesizer of the present invention.

FIG. 4 is a diagram showing another example for generating a multiplication coefficient.

FIG. 5 is a diagram showing a modified example of the musical sound synthesizer of the present invention.

FIG. 6 is a diagram showing another example of a string section and an excitation waveform generating section.

FIG. 7 is a diagram showing a configuration of a resonator section.

FIG. 8 is a diagram showing an application example of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 delay means, 2 filter, 3 coefficient multiplier, 4 maximum value selecting means (maximum value selecting section), 5 adder, 6 excitation waveform generating section, 7 smoothing filter, 8 tone color setting section, 9
Control unit, 10 performance operators, 11 performance assist operators, 1
2 coefficient supply section, 20 dump coefficient generation section, 21 excitation section, 22 string section, 23 resonator section, SG1, S
G2 ... SGn Musical sound synthesizing means

Claims (3)

[Claims] 1. A musical tone generating means for generating a musical tone, an attenuating state controlling means for controlling an attenuating state of the musical tone generated by the musical tone generating means, and a first instruction for generating the musical tone and ending the musical tone. The performance information generating means and the second performance information generating means for instructing the attenuation mode of the musical sound are provided, and the attenuation state control means includes the first performance information generating means and the second performance information generating means. A tone synthesizer characterized in that a control signal corresponding to both values of the generated performance information is supplied. 2. An excitation signal generating means for generating an excitation signal for driving a closed loop for generating a musical tone, a loop gain control means for controlling a loop gain of the closed loop, and an instruction for generating the musical tone and ending the musical tone. A first performance information generating means and a second performance information generating means for instructing a decay mode of the musical sound are provided, and the performance information generated by the first performance information generating means and the second performance information. A musical tone synthesizer characterized in that the loop gain is determined according to both values of performance information generated by a generating means. 3. The performance information generated by the first performance information generating means is compared with the performance information generated by the second performance information generating means, and any performance is performed based on the comparison result. 3. The musical tone synthesizer according to claim 2, wherein information is selected and supplied to the loop gain control means.