JPH11219185A - Musical sound synthesizing device and synthesizing algorithm derivation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a musical sound synthesizing device enabling even a person incapable of performing an actual wind instrument to have pseudo-experience with the substantial performance of a wind instrument. SOLUTION: This device has a processor 103 for generating approximately equal time waveforms in a wind instrument sound via the solution of a nonlinear differential equation derived from the addition of a forced vibration term such as a voice to a synthesizing algorithm based on the sound production mechanism of a wind instrument, and a microphone 101 for inputting a voice signal to the processor 103 as the forced vibration term of the differential equation. Also, the processor 103 generates a desired musical sound on the basis of each information (e.g. pitch, level and waveform) of the voice signal. In this case, a tone quality change peculiar to the wind instrument can be realized simply by controlling a voice generation mode. For expressing a growl (a trembling phoneme appearing at the rise of a sound), for example, a trembling sound is pronounced directly by a voice, and the input of a voice in the 'ta'-column or the 'ka'-column is simply required for expressing tanging.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical sound synthesizer used for a sound source of an electronic musical instrument.

[0002]

2. Description of the Related Art A tone synthesizer for synthesizing a tone by simulating the sounding structure of a natural musical instrument with an electronic circuit has been proposed (for example, JP-A-6-222777). According to this type of synthesizer, the tone color of a wind instrument such as a trumpet can be changed like a musical instrument during sounding.

[0003] A conventional tone synthesizer as described above will be described below with reference to the drawings.

FIG. 8 is a block diagram of a conventional tone synthesizer. In FIG. 8, 601 is a subtractor, 602 is a vibrating unit that simulates lip vibration when a trumpet is played, 603 is a table that stores acoustic admittance characteristics of the lip gap, 604 is a multiplier, and 605 is a subtractor. Bowl, 60
6 is an integrator, 607 is an adder, and 608 is a delay.

FIG. 9 is a circuit diagram of the vibration unit 602. FIG.
, 701 is a subtractor, 702 is an adder, 703,
704 is a delay unit for delaying data by one sampling time Ts, and 705, 706, and 707 are coefficients a and
This is a multiplier for multiplying b and c by corresponding data.

FIG. 10 is a longitudinal sectional view showing a vibration state of a lip when playing a trumpet, and FIG. 11 is a schematic diagram showing a vibration of a lip when playing a trumpet, expressed by a mass point, a spring, and a dashpot. 12 is a graph showing the acoustic admittance characteristics of the lip gap when playing the trumpet.

FIG. 13 is a circuit diagram of the integrator 606. In FIG. 13, reference numeral 901 denotes an adder; 902, a delay unit for delaying data by one sampling time Ts; and 903, a multiplier for multiplying a coefficient d by data.

FIG. 14A is a graph showing the time characteristic of the intraoral pressure data P corresponding to the pressure in the mouth when playing the trumpet, and FIG. 14B corresponds to the tightness of the lips when playing the trumpet. FIG. 14C is a graph showing the time characteristic of the ambusher data Am, and FIG.
FIG. 14D is a graph showing the time characteristic of the synthesized sound data.

FIG. 15A is a graph showing the time characteristic of the intraoral pressure data P corresponding to the pressure in the mouth when playing the trumpet, and FIG. 15B corresponds to the tightness of the lips when playing the trumpet. FIG. 15C is a graph showing the time characteristic of the ambusher data Am, and FIG.
FIG. 15D is a graph showing the time characteristic of the synthesized sound data.

FIG. 16 is a longitudinal sectional view showing a vibration state of a reed when playing a saxophone, FIG. 17 is a schematic diagram showing vibration of a reed when playing a saxophone, represented by a mass point, a spring, and a dashpot. 3 is a graph showing acoustic admittance characteristics of a void of a lead when a saxophone is played.

The operation of the conventional tone synthesizer constructed as described above will be described. First, the overall operation will be described with reference to FIG. First, pressure is generated in the player's mouth when playing the trumpet. The corresponding mouth pressure data P is sent to the subtractor 601. Subtractor 60
1, the data P corresponding to the sound pressure in the mouthpiece
The difference value ΔP between m and the mouth pressure data P is calculated, and ΔP
Is sent to the vibration unit 602 as a forced vibration term for vibrating the lips. On the other hand, ambusher data Am corresponding to the degree of tightening of the lips is also input to the vibration unit 602, and the natural frequency (the vibration pitch of the lips) in the vibration unit 602 is determined. ΔP excites the vibration in the vibration unit 602, and data x corresponding to the displacement of the lips (the degree of opening of the lips) is sent from the vibration unit 602 to the table 603. The table 603 stores data A corresponding to the acoustic admittance of the lip gap according to the data x.
dm (x) is referred to and sent to the multiplier 604. The multiplier 604 multiplies ΔP by Adm (x), and data U corresponding to the volume flow velocity flowing into the mouthpiece from the player's mouth is sent to the subtractor 605. Subtractor 605
In, the volume flow velocities Ui and U flowing backward from the tube into the mouthpiece are subtracted to determine data Um corresponding to the volume flow velocity in the mouthpiece. Um is the integrator 606
In, data Pm corresponding to the sound pressure in the mouthpiece is determined by an integration process caused by the acoustic compliance (so-called hollow state) in the mouthpiece. On the other hand, U
m is added to Ui, data Uo corresponding to the volume flow velocity flowing into the main body side of the pipe is calculated, and output as desired synthesized sound data. The delay 608 simulates the process (delay) of Ui passing through the main body of the tube, being reflected near the morning glory, and returning as Uo again in a concentrated manner, and the length of this process, that is, the delay length Is controlled by pitch data Pi corresponding to the pitch of the synthesized sound data.
Note that, for simplicity, a circuit corresponding to the radiation characteristics of the morning glory has been omitted.

Now, referring to FIG. 9 to FIG.
2 will be described. An important factor in lip vibration is data x (upper lip displacement) in FIG. Assuming that the upper lip and lower lip vibrate symmetrically,
The lip gap can be considered to be proportional to x.
The vibration model shown in FIG. 11 simulates the vibration of the upper lip in a simple manner. This vibration model can be described by (Equation 1).

[0013]

(Equation 1)

The circuit shown in FIG. 9 is a circuit obtained by converting the linear differential equation (Equation 1) into a digital circuit. The coefficients a, b, and c are represented by (Equation 2). It is assumed that the spring constant k is equal to the ambusher data Am. Increasing the tightness of the lips means increasing the spring constant k (Ambusher data Am), which increases the natural frequency of the lips.

[0015]

(Equation 2)

X generated in the vibrating section 602 is an element for determining the acoustic admittance Adm (x) of the lip gap;
That is, it is transmitted as the address value of the table 603. The acoustic admittance characteristic values shown in FIG. 12 are stored in the table 603, and the acoustic admittance Adm (x) referred to based on x is multiplied by ΔP in the multiplier 604, and the volume flowing through the gap of the lips The flow velocity U is calculated. It should be noted that the ambusher data Am and the pitch data P
The respective values of i are given by (Equation 3) and (Equation 4). When they are both values T, the synthesized sound data has a period T
Is obtained.

[0017]

(Equation 3)

[0018]

(Equation 4)

Next, the internal operation of the integrator 606 will be described with reference to FIG. The integrator 606 is obtained by converting the differential equation described by (Equation 5) into a digital circuit. The coefficient d is represented by (Equation 6).

[0020]

(Equation 5)

[0021]

(Equation 6)

The operation described above is an operation per sampling time Ts. By repeating this operation over a plurality of samples, the synthesized sound data shown in FIG. 14D is output. The rise time of the synthesized sound data depends on the rise time of the intraoral pressure data P shown in FIG. Further, the stability of the amplitude level of the synthesized sound data immediately after the sound has risen is guaranteed when both of the following two conditions are satisfied.

<Condition 1> Intraoral pressure data P must be constant. <Condition 2> Pitch data Pi and unbushing data A
In both cases, the value corresponding to the period T must be kept constant.

Since the synthesized sound data shown in FIG. 14D satisfies both <condition 1> and <condition 2>, its amplitude level is constant (stable).

On the other hand, as shown in FIG. 15B, when the emboucher data Am is small near the rising edge of the sound, takes a sufficiently long time than the period T, and gradually approaches the value corresponding to the period T, Since the <condition 2> cannot be satisfied between the time 0 and the time ta, the amplitude level of the synthesized sound data becomes inconsistent as shown in FIG. In this phenomenon, the natural mode of the lips does not match the vibration mode of the air column (corresponding to the pitch data Pi) generated in the main body of the tube because the ambucher data Am is different from the value corresponding to the period T (resonance). No). FIG.
The period of the envelope (shown by the broken line) in (D) is
Corresponding to the natural frequency of the lips, that is, the Ambusher data Am, the frequency gradually approaches a very low frequency with respect to the period T, and becomes a periodic waveform having a constant amplitude after time ta. In playing brass instruments such as trumpets, this phenomenon is an element of musical expression (called growl).
It is actively used as For example, in a jazz-type performance, in order to give a sense of fancy, a tone without a gross feeling (FIG. 14D) and a tone with a gross feeling (FIG. 15D) are intentionally set according to the composition. We may use properly. Further, the blowing pressure may be changed at the same time to make the timbre change more complicated.

[0026]

However, in the configuration shown in FIG. 8, in order to realize the above-mentioned tone color change, a plurality of parameters such as the intraoral pressure data P and the ambucher data Am are set to different operators. And it is necessary to control them at the same time, and it is very difficult to change the timbre.

The present invention solves the above problems,
A tone synthesizer capable of easily achieving the above-described tone change is provided.

[0028]

In order to achieve this object, a musical tone synthesizing apparatus according to the present invention employs, based on a non-linear differential equation, a time waveform (a desired musical tone) substantially equal to the shape of the musical instrument sound. Signal), and external signal input means for inputting an external signal such as a voice as a forced vibration term of an operation based on a non-linear differential equation in the vibration simulation means. Thus, since the vibration state of the vibration simulation means is directly controlled by using an external signal such as a voice, it is possible to easily control a change in timbre (for example, a sense of grace).

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention provides a sound detecting means for detecting a sound, an analog-to-digital converting means for performing an analog-to-digital conversion of the detected sound, and a time waveform of a musical instrument sound by an operation based on a nonlinear differential equation. And a digital-to-analog converting means for converting a digital output output from the vibration simulating means into a desired musical tone signal. With this configuration, based on the sound signal detected by the sound detecting means, the vibration simulating means generates a time waveform (desired tone signal) substantially equal to the shape of the time waveform of the musical instrument sound by an operation based on a nonlinear differential equation. Since the nuances of the timbre change of a wind instrument are similar to the nuances of the timbre change in the difference in sound generation,
Depending on the difference in the manner in which the sound is generated, it is possible to perform a tone change like a wind instrument.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a musical sound synthesizer according to an embodiment of the present invention. FIG.
In the figure, 101 is a microphone for detecting a voice, 102 is an analog-to-digital converter (ADC), 103 is a difference equation (an arithmetic expression obtained by replacing a nonlinear differential equation described later in a recurrence form) shown in FIG. A processor as a vibration simulation means executed for each Ts, 104
Is a digital-to-analog converter (DAC), 105 is a ΔT control unit that controls the value of the minute time term ΔT in the difference equation of FIG. 2, and 106 is the difference equation of FIG.
a loop control unit for controlling the number of executions N in s, 107
Denotes a mixing control unit for performing weight control of data output from the processor 103; 108, a selector;
Is a pitch extractor for extracting the pitch of the voice. In addition,
Since the internal configuration of the pitch extractor 109 is generally well known, a description of the internal configuration will be omitted.

FIG. 2 is a flowchart of the difference equation executed by the processor 103. FIG. 3 is a simplified circuit diagram of the conventional tone synthesizer of FIG. In FIG. 3, 2
01 is a subtractor. Other blocks are the same as those of the tone synthesizer of FIG.

FIG. 4 shows a delay 60 of the tone synthesizer of FIG.
8 is a tone synthesizer in which 8 is deleted and replaced with an external periodic waveform data w instead of the feedback input from the delay 608. In FIG. 4, reference numeral 301 denotes a subtractor. FIG. 5 shows a tone synthesizer in which the position of an external input in the tone synthesizer of FIG. FIG.
In the figure, reference numeral 302 denotes an adder. It is assumed that the audio data s is input to the adder 302.

FIG. 6A is a time characteristic diagram of the audio data s,
FIG. 6B is a time characteristic diagram of the synthesized sound data. Also,
Similarly, FIG. 7A is a time characteristic diagram of the audio data s, and FIG.
(B) is a time characteristic diagram of the synthesized sound data.

The operation of the tone synthesizer constructed as described above will be described. First, a method of deriving the difference equation in FIG. 2, which is a synthesis algorithm of the musical sound synthesizer of the present invention, will be described. First, the conventional tone synthesizer shown in FIG. 8 is changed to the configuration shown in FIG. This simply combines the subtractor 605 and the adder 607 of FIG. 8 into the subtractor 201 of FIG. Next, FIG. 4 is derived from FIG. The periodic waveform data out (Tt) is fed back from the delay 608 in FIG. The time T is a cycle corresponding to the pitch of the synthesized sound data, and is also a delay time of the delay 608. FIG. 4 shows a musical tone synthesizer to which a waveform substantially equal to the periodic waveform data out (T−t) is externally given. According to the tone simulation, out (T−t)
3 and FIG. 4 if the waveform shapes do not exactly match each other including the phase,
Have the same timbre.

Next, as a modification from FIG. 4 to FIG. 5, an adder 302 is added instead of removing the subtractor 301,
Further, if a configuration is adopted in which periodic waveform data w2 obtained by integrating the periodic waveform data w is input instead of the periodic waveform data w, FIG.
And FIG. 5 are equivalent circuits. The periodic waveform data w2 is data obtained by subjecting the periodic waveform data w to the integration action (high-frequency cutoff) of the integrator 606, and has a waveform shape with few overtone components as shown in FIG. Become. If this periodic waveform data w2 is externally input as voice data s (preferably voice with few overtones such as back voice), FIG.
Is substantially equal to the tone color of the synthesized sound data of FIG. Now, the difference equation in FIG. 2 is derived based on the circuit in FIG. The synthesized sound data of FIG. 5 is dy / dt (y is 1
Is the second derivative, which is expressed in the form of a difference y
dot), the output of the integrator 606 is y
Becomes Further, the conversion characteristic A of the table 603 shown in FIG.
Assuming that dm (x) is x ² , the output of the table 603 is x ²
Becomes The difference equation in FIG. 2 is derived using FIG. 5 and the differential equation of the vibration unit 602 shown in (Equation 1). First, since ΔP in (Equation 1) corresponds to (s−y + P),
(Equation 1) can be transformed into (Equation 7).

[0036]

(Equation 7)

In equation (7), a new variable dz / dt
And (g) can be transformed from (Equation 7) into (Equation 8).

[0038]

(Equation 8)

In (Equation 8), the second derivative term of t is 1
(9) can be derived by using the second derivative term.

[0040]

(Equation 9)

Further, based on FIG. 5, dy / dt (ydot
(Equation 10) can be derived.

[0042]

(Equation 10)

Here, a first-order differential term such as dx / dt is approximated to a first-order differential term, and each difference term is represented by xdot, ydot, zd.
If it is ot, x, y, and z can be represented by using a short time term ΔT as shown in (Equation 11).

[0044]

[Equation 11]

Now, the various variables in (Equation 11) are recurrence formulas updated at every sampling time Ts. FIG. 2 shows recurrence formulas updated at time intervals shorter than one sampling time Ts. This is the difference equation shown, and one sampling time Ts is divided into four. In this way, by making the update time interval shorter, a digital system (difference equation) in which the analog system (differential equation) is more strictly simulated can be obtained. Here, various constants such as A and B in (Equation 11) are strictly determined by measuring various physical quantities (for example, mass of lips) at the time of playing an actual trumpet or the like. Therefore, as a result of performing the timbre simulation based on the difference equation of FIG. 2, the values of the various constants are as shown in (Equation 12). In the case of this value, a tone color substantially similar to the tone color of the trumpet is obtained.

[0046]

(Equation 12)

The operation of the musical sound synthesizer of the present invention will be described with reference to the difference equation of FIG. 2 described above. First, as shown in FIG. 1, an audio signal is taken into a processor 103 as audio data s via a microphone 101 and an analog-to-digital converter 102. When synthesizing a normal trumpet tone, the audio data s may have a periodic waveform with a constant amplitude level as shown in FIG.
By this input, the synthesized sound data ydot becomes a periodic waveform having a constant amplitude level as shown in FIG. This waveform is a waveform having substantially the same shape as the synthesized sound data of the conventional musical sound synthesizer shown in FIG. By the way, if a voice that contains many overtone components as an audio signal (for example, a disturbed voice) is input, the overtone components of the voice itself are also included in the synthesized sound data and output, so that the actual trumpet The tone is different from the instrument tone. However, since the characteristic overtone structure of the musical instrument sound (for example, one corresponding to the natural vibration of the lips) is preserved, the image of the rough timbre does not change.

On the other hand, when audio data s whose amplitude level fluctuates at the rising edge as shown in FIG. 7A is input, the synthesized sound data ydot has an unstable amplitude level as shown in FIG. 7B. A periodic waveform is obtained. This waveform is a waveform having substantially the same shape as the synthesized sound data of the conventional musical sound synthesizer shown in FIG. Note that the sound data s shown in FIG. 7A is a timbre when the sound is pronounced “blue”, and is substantially equal to a rising timbre (vibration sound of only the lips) when the trumpet has a gross feeling. . Further, in a wind instrument such as a trumpet, a musical tone may be expressed by selectively using a tone when tangling is performed and a tone when non-tangling is performed. By simply using the "phoneme" and the "ta-line phoneme", it is possible to realize a difference in timbre with or without tonging.

Here, ΔT introduced into the difference equation of FIG.
And the meaning of N will be supplementarily described. First, although ΔT, xdot, ydot, zdo in (Equation 11)
As a minute time term for calculating x, y, and z from t, ΔT is commonly used for each variable, but they are different physically (strictly). For example, yd
ΔT corresponding to ot corresponds to (Equation 6) and the accumulation coefficient d of the integrator 606 shown in FIG. 13 (inversely proportional to the acoustic compliance C). On the other hand, ΔT corresponding to x and z
Is a function of the spring constant k corresponding to the tightness of the lips. In the conventional tone synthesizer, the spring constant k is an element for controlling the tone color of the synthesized tone data as the unbushing data Am. However, in the present invention, since the tone color is directly controlled by the voice signal, the tone color control plane ( It is not necessary to independently provide a spring constant k (ΔT corresponding to the spring constant k) on the timbre changing surface). Even in the case of the timbre itself, there is not much problem in timbre simulation (audibility) even if ΔT is a common value as shown in (Equation 12). Here, changing the tone by changing the value of ΔT will be described. As described above, ΔT is inversely proportional to the acoustic compliance C. The acoustic compliance C corresponds to the size of the internal volume in the mouthpiece.
Therefore, when the value of ΔT is made smaller, the internal volume in the mouthpiece changes to a large tone (a tone such as a trombone or a tuba with respect to the trumpet).

Focusing on this point, the tone synthesizer of the present invention is provided with a ΔT control unit 105 as shown in FIG. The ΔT control unit 105 is a table shown in (Table 1) or a table shown in (Table 2).

[0051]

[Table 1]

[0052]

[Table 2]

In FIG. 1, when the MODE flag input to the selector 108 has a value of 0, the A input (instrument type instruction data) is selected. In this case, it is assumed that the ΔT control unit 105 is a table shown in (Table 1). ΔT control unit 10
5, the value of .DELTA.T corresponding to the musical instrument type instruction data is read out, set in the processor 103, and a tone corresponding to each musical instrument type instruction data is synthesized. On the other hand, MO
If the DE flag is 1, the B input (pitch extractor 109
Output) is selected. In this case, the ΔT control unit 105
Is a table shown in (Table 2). ΔT control unit 1
From 05, the value of ΔT corresponding to the pitch of the voice is read out, set in the processor 103, and a tone color corresponding to each musical instrument type instruction data is synthesized. By changing the tone according to the pitch in this way, the tone suitable for the pitch can be automatically switched (for example, a tuba for a low pitch, a piccolo trumpet for a high pitch). . Also (Table 2)
Is a table with a fine resolution in the address direction, it is possible to smoothly (continuously) switch these musical instrument types. The pitch extractor 109 may be of a type that extracts the pitch of the output (digital audio signal) of the analog-to-digital converter 102.

Next, in the meaning of N, N is the number of times the difference equation (Equation 11) is executed per sampling time Ts, and the larger N is, the more the differential equation (Analog 10) System) is more strictly digitally simulated, and the degree of approximation of the timbre is increased. However, when synthesizing the trumpet tone regardless of the value of N, it is necessary to adjust the value of ΔT according to the increase or decrease of the value of N while maintaining the relationship shown in (Expression 13).

[0055]

(Equation 13)

By using the relational expression of (Expression 13), it can be seen that the timbre (instrument type) can be changed also by N. In the tone synthesizer of the present invention, as shown in FIG.
06, and the loop control unit 106 is a table for outputting the value of N based on the musical instrument type instruction data. (Table 3)
Shows the relationship between instrument type instruction data and N.

[0057]

[Table 3]

In the difference equation shown in FIG. 2, the desired synthesized sound data is Ydot. However, the high frequency content of the synthesized sound data can be easily controlled (tone control) by an arithmetic expression shown in (Equation 14). Can be performed. Since y is equivalent to a value obtained by integrating ydot, that is, a value obtained by passing through a low-pass filter, the content of the high-frequency component is smaller than that of ydot.
Therefore, if the mixing gain Mg in (Equation 14) is increased, the high-frequency content increases, and if the mixing gain Mg is decreased, the high-frequency content decreases. It is assumed that the tone control data can be switched between values 1 to 4 in one step. Table 4 shows the relationship between the tone control data and the mixing gain Mg.

[0059]

[Equation 14]

[0060]

[Table 4]

Note that, in addition to y and ydot, audio data s
And the like may be used as synthesized sound data.

Although the brass instrument such as a trumpet has been described above, the present invention can also be applied to a woodwind instrument (a lead wind instrument) such as a saxophone. It is the acoustic admittance Adm of the gap between the lead and the mouthpiece shown in FIG.
It is only necessary to change (x). X ^{2 for} trumpet
However, in the case of a woodwind instrument, the relational expression shown in (Equation 15) may be used.

[0063]

(Equation 15)

In this case, the difference equation of (Equation 11) is as shown in (Equation 16).

[0065]

(Equation 16)

In the embodiment of the present invention, the processor 10
Although the synthesizing process is performed in the form of a difference equation (digital) using FIG. 3, the synthesizing process may be performed in the form of a differential equation (analog) represented by (Equation 7) or (Equation 10). For example, it is a generally known technique that the first-order differential form shown in (Equation 10) can be configured using an analog element such as an operational amplifier, a resistor, or a capacitor. If this technique is used, an analog circuit can be used. This eliminates the need for the analog-to-digital converter 102 and the digital-to-analog converter 104, and provides a cost advantage.

As described above, according to the embodiment of the present invention, the tone equation is synthesized by forcibly vibrating the difference equation (the equation corresponding to the sound generation mechanism of the trumpet) shown in FIG. A tone color change such as a tone color without a sense of grace (FIG. 6B) or a tone color with a sense of grace (FIG. 7B) can be controlled only by how to generate a voice. In addition, by changing the difference equation from the equation shown in (Equation 11) to the equation shown in (Equation 16), it is possible to support lead-type wind instruments such as saxophones.

When the MODE flag has a value of 0, the ΔT control unit 105 performs the conversion shown in (Table 1) based on the musical tone type instruction data, so that the musical instrument type can be easily changed. .

When the MODE flag has the value 1, based on the pitch of the voice extracted by the pitch extractor 109, ΔT
Since the control unit 105 performs the conversion shown in (Table 2), it is possible to automatically select the musical instrument type most suitable for the pitch or make a continuous transition.

Further, since the loop control unit 106 performs the conversion shown in (Table 3) based on the musical instrument type instruction data, the timbre can be easily changed.

Also, based on the tone control data, the mixing control unit 107 performs the conversion shown in (Table 4), and the processor 103 outputs the desired synthesized sound data while controlling the mixing gain Mg of (Equation 14). Therefore, the content of the high frequency component can be easily changed.

In this embodiment, audio is used as an external input. However, as shown in FIG. 6 (A) and FIG. 7 (A), a sound source which samples a plurality of audio signals is prepared and this output ( Line) is connected to the analog-to-digital converter 102. For example, if the waveforms of FIGS. 6A and 7A are switched in accordance with the touch of the keyboard, a similar effect can be obtained only by controlling the strength of the touch.

[0073]

According to the present invention, a vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by an operation based on a nonlinear differential equation, and an external signal such as a voice. Signal input means for inputting a signal as a forced vibration term of an operation based on a non-linear differential equation in the vibration simulation means. The external signal input means detects an external signal such as voice, and directly uses the signal to simulate vibration. Since the vibrating state of the sound means is controlled, it is possible to easily control a change in timbre (for example, a sense of grace).

The present invention also provides a vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation; External signal input means for inputting an external signal, such as an external signal, as a forced vibration term of the difference equation in the vibration simulation means, and a minute time term of the difference equation in the vibration simulation means {T is changed according to the designated musical instrument type} T control means, and the ΔT control means controls the value of the minute time term ΔT corresponding to the difference in the type of the instrument (tone color) in accordance with the designated instrument type parameter. You can change the tone.

The present invention also provides a vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation; External signal input means for inputting an external signal such as a forced vibration term of the difference equation in the vibration simulation means, and a minute time term ΔT of the difference equation in the vibration simulation means is changed according to the pitch of the external signal such as voice. △ T control means, and a pitch extraction means for an external signal such as voice, the pitch extraction means detects a pitch of an external signal such as voice, and △ T control means, in accordance with the detected pitch, Since the value of the short time term ΔT is controlled, it is possible to synthesize an optimum timbre for the pitch.

The present invention also provides a vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of the time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation; External signal input means for inputting an external signal such as a forced vibration term of a difference equation in the vibration simulation means, and loop control means for controlling the number N of executions of the difference equation in one sampling unit time in the vibration simulation means. And the loop control means controls the value of N corresponding to the difference in the type of instrument (tone color) according to the designated instrument type parameter, so that the tone color can be easily changed.

Further, the present invention provides a vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation; An external signal input means for inputting an external signal such as a forced vibration term of a difference equation in the vibration simulating means, and a mixture of a variable y in the difference equation and its first-order difference term ydot in the vibration simulating means are output. And mixing control means for controlling the mixing ratio. The mixing control means changes the mixing ratio of the variable y, which is the element of the desired synthesized sound data, and its first-order difference term ydot. Thus, the high-frequency content of the desired synthesized sound data can be controlled.

Further, according to the present invention, the sounding mechanism of the musical instrument is analyzed, and a circuit corresponding to the oscillating portion of the musical instrument and a circuit corresponding to the resonance portion of the musical instrument (a portion for determining a desired musical tone pitch) are combined. A circuit (circuit A) is obtained, a circuit corresponding to the resonance portion is removed, and a circuit (circuit B) configured to input an external signal having a pitch such as voice to the circuit corresponding to the oscillation portion is obtained. By replacing the content of the signal processing with a nonlinear differential equation, it is possible to derive a synthesis algorithm corresponding to an analog circuit that can directly control timbre by an external signal such as voice.

Further, according to the present invention, the sounding mechanism of the musical instrument is analyzed, and a circuit corresponding to the oscillating portion of the musical instrument and a circuit corresponding to the resonant portion of the musical instrument (a portion for determining a desired tone pitch) are combined. A circuit (circuit A) is obtained, a circuit corresponding to the resonance portion is removed, and a circuit (circuit B) configured to input an external signal having a pitch such as voice to the circuit corresponding to the oscillation portion is obtained. By replacing the content of the signal processing with a nonlinear differential equation, and further calculating a nonlinear differential equation based on the nonlinear differential equation,
It is possible to derive a synthesis algorithm corresponding to a digital circuit capable of directly controlling the timbre by an external signal such as voice.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a musical sound synthesizer according to an embodiment of the present invention.

FIG. 2 is a flowchart of a difference equation executed by a processor of the musical sound synthesizer.

FIG. 3 is a simplified block diagram of a conventional tone synthesizer.

FIG. 4 is a block diagram showing another configuration method of the musical sound synthesizer.

FIG. 5 is a block diagram showing another configuration method of the musical sound synthesizer.

FIG. 6 is a waveform diagram of voice data and synthesized sound data.

FIG. 7 is a waveform diagram of audio data and synthesized sound data.

FIG. 8 is a block diagram of a musical sound synthesizer in a conventional example.

FIG. 9 is a circuit configuration diagram of a vibrating section of the musical sound synthesizer.

FIG. 10 is a longitudinal sectional view showing a vibration state of a lip when playing a trumpet.

FIG. 11 is a schematic diagram showing the vibration of a lip when playing a trumpet, represented by a mass point, a spring, and a dashpot.

FIG. 12 is a graph showing acoustic admittance characteristics of a lip gap when playing a trumpet.

FIG. 13 is a circuit diagram of an integrator.

FIG. 14 is a waveform diagram of each signal when a trumpet sound is synthesized.

FIG. 15 is a waveform diagram of each signal when synthesizing a trumpet sound.

FIG. 16 is a longitudinal sectional view showing a vibration state of a lead when a saxophone is played.

FIG. 17 is a schematic diagram showing vibration of a reed when a saxophone is played by using a mass point, a spring, and a dashpot.

FIG. 18 is a graph showing acoustic admittance characteristics of a lead gap when a saxophone is played.

[Explanation of symbols]

 Reference Signs List 102 analog-to-digital converter 103 processor 104 digital-to-analog converter 105 ΔT control unit 106 loop control unit 107 mixing control unit 108 selector 109 pitch extractor

Claims (15)

[Claims] 1. A vibration simulating means for generating a time waveform (desired musical tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation; And an external signal inputting means for inputting a forced oscillation term of a nonlinear differential equation in the means. 2. The musical sound synthesizer according to claim 1, wherein the vibration simulating means generates a time waveform by solving a nonlinear differential equation in the form of a difference equation. 3. The tone synthesizer according to claim 2, wherein the vibration simulating means solves a second-order nonlinear differential equation in the form of a difference equation. 4. A vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation, and an external signal such as a voice. Signal input means for inputting as a forced vibration term of the difference equation in the vibration simulating means, and changing the short time term ΔT of the difference equation in the vibration simulating means in accordance with the designated musical instrument type ΔT A tone synthesizer comprising control means. 5. The musical sound synthesizer according to claim 4, wherein the vibration simulating means solves a second-order nonlinear differential equation in the form of a difference equation. 6. A vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation, and an external signal such as a voice. Is input as a forced vibration term of the difference equation in the vibration simulating means, and the short time term of the difference equation in the vibration simulating means is changed according to the pitch of the external signal such as voice. A tone synthesizer comprising: T control means; and pitch extracting means for extracting a pitch of an external signal such as voice. 7. The musical sound synthesizer according to claim 6, wherein the vibration simulating means solves a second-order nonlinear differential equation in the form of a difference equation. 8. The tone synthesizing apparatus according to claim 6, wherein the ΔT control means changes ΔT continuously in accordance with the pitch of an external signal such as a voice. 9. A vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation, and an external signal such as a voice. And an external signal inputting means for inputting as a forced vibration term of a difference equation in the vibration simulating means, the difference equation in the vibration simulating means is executed N times within one sampling unit time, and A tone synthesizer characterized by changing the value of N according to the type of musical instrument selected. 10. The tone synthesizer according to claim 9, wherein the vibration simulating means solves a second-order nonlinear differential equation in the form of a difference equation. 11. A vibration simulating means for generating a time waveform (desired tone signal) substantially equal to the shape of a time waveform of a musical instrument sound by solving a nonlinear differential equation in the form of a difference equation, and an external signal such as a voice. And a signal obtained by mixing the variable y in the difference equation and its first-order difference term ydot in the vibration simulation means. A tone synthesizer comprising: a mixing control means for controlling the mixing ratio. 12. The tone synthesizer according to claim 11, wherein the vibration simulating means solves a second-order nonlinear differential equation in the form of a difference equation. 13. The musical sound synthesizer according to claim 11, wherein the mixing control means simultaneously mixes a variable other than the variable y in the difference equation and its first-order difference term ydot in the vibration simulating means. . 14. Analyzing the sounding mechanism of a musical instrument, find a circuit A in which a circuit corresponding to an oscillating portion of the musical instrument and a circuit corresponding to a resonance portion of the musical instrument (a portion for determining a desired tone pitch) are connected. A circuit B configured to remove a circuit corresponding to the resonance portion and to input an external signal having a pitch such as voice to the circuit corresponding to the oscillation portion.
And deriving a synthesis algorithm by replacing the content of the signal processing of the circuit B with a nonlinear differential equation. 15. Analyzing the sounding mechanism of a musical instrument, find a circuit A in which a circuit corresponding to an oscillating portion of the musical instrument and a circuit corresponding to a resonance portion of the musical instrument (a portion for determining a desired tone pitch) are connected. A circuit B configured to remove a circuit corresponding to the resonance portion and to input an external signal having a pitch such as voice to the circuit corresponding to the oscillation portion.
And substituting the content of the signal processing of the circuit B with a nonlinear differential equation, and further obtaining a nonlinear difference equation based on the nonlinear differential equation, thereby deriving a recursive formula algorithm.