JPH02281296A - Musical tone synthesizer - Google Patents

Abstract

PURPOSE:To shorten the computing time per piece of junctions by using square calculation grids as the junctions for signal scattering of a bidirectional transmission means. CONSTITUTION:The information indicating the various intensities of blowing supplied from a musical tone control information generating circuit 21 is supplied as a DC bias VA to a nonlinear element 11. The resonance frequency of a resonance circuit 23 is switched according to the information indicating the scale supplied from the musical tone control information generating circuit 21. This resonance frequency switching is enabled by switching the effective number of steps of delay circuits DF1, DF2,..., DR1, DR2,... in the resonance circuit 23. The lesser number of the multiplication times required for signal processing is necessitated in the case of using the square calculation grids as the junctions JA1, JA2,... than for the biquadratic or cubic calculation grids in the case of executing the signal processing by the computation of signal processors, etc. The computing power and processing speed required for the signal processing processor are relieved in this way.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a musical tone synthesis device suitable for use in electronic musical instruments and the like.

``Prior Art'' Conventionally, electronic musical instruments have been well known in which a musical sound waveform generated by a natural musical instrument is stored in a waveform memory in advance, and the musical sound waveform is read out from the waveform memory in response to an operation by a performer to generate a musical sound. ing. In addition, in high-performance electronic musical instruments, calculations are performed on the musical sound waveforms read from the waveform memory, or processing is performed to synthesize multiple musical sound waveforms, resulting in more natural musical sound reproduction. There is.

However, the actual sounds of natural musical instruments vary widely depending on the way they are played, the environment in which they are played, and other factors.

For example, in a wind instrument such as a clarinet, even if the scale is the same, the sound waveform changes in a variety of ways, depending on whether the scale is played or played softly, and the audience perceives this change as a change in timbre. I can feel it. Note that this phenomenon will be explained in detail later.

If we try to create an electronic musical instrument that faithfully reproduces the various sound waveforms of natural instruments using the method using the waveform memory described above, we will need a waveform memory that can store many Ill waveforms and a complicated A calculation means capable of waveform processing is required, which is difficult to realize.

Against this background, attempts have been made to generate musical tones of natural instruments by modeling the sound production mechanism of natural instruments and operating the created models without using waveform memory or the like.

This type of technology is disclosed in, for example, Japanese Patent Laid-Open No. 63-40199.

Hereinafter, modeling of two sets of natural musical instruments and a musical tone synthesis device using a model obtained by this modeling will be explained using a wind instrument as an example.

The simplest model of a wind instrument, as shown in FIG. 3, is constructed by combining a resonance pipe 1 (a model of the wind instrument's pipe section) and a reed 2 made of an elastic body. In this configuration, when the blow player blows 2A of exhaled air into the reed 2, the reed 2 is pushed toward the inside of the tube by this exhalation pressure PA (in the direction of the arrow 2F).

Further, since the reed 2 is an elastic body, it vibrates when the exhaled air 2A is sprayed onto it. As a result, an air pressure wave (compression wave) is generated inside the tube of the lead 2, and this becomes a traveling pressure wave F toward the terminal end IE of the resonator 1. The traveling pressure wave F is then reflected at the terminal end IE and returns to the lead 2 as a reflected pressure wave R, and the lead 2 receives the pressure PR from the reflected pressure wave R. As a result, lead 2 is P
=PA-PR...(1) Receives a pressure P. Eventually, the reed 2 will vibrate according to this pressure P and the elastic properties of the reed 2.

FIG. 4 exemplifies the elastic characteristics of the lead 2, ie, the relationship between the pressure P (input) that the lead 2 receives and the amount of displacement (output) of the lead 2.

As shown in this figure, the amount of displacement of the lead 2 is non-linear with respect to the pressure P, and when the pressure P increases to a certain extent, the amount of displacement is saturated.

Now, a resonance phenomenon occurs when the frequency of vibration in the reed 2 is equal to the resonance frequency of the resonance tube 1, which will be explained below.
A large pressure wave is obtained in the resonance tube 1, and this is output as the musical sound of the wind instrument.

Furthermore, when air vibrations of a specific frequency determined by the dimensions of the air column in the resonance tube I (this frequency is called the resonance frequency) occur, the air pressure wave is constant in the dimension direction inside the resonance tube I. Current waves are generated and large vibrations are obtained in the resonant tube I. This phenomenon is known as a resonance phenomenon.

Here, the relationship between the dimensions of the resonance tube 1 and the wavelength λ of the standing wave will be explained. As shown in FIG. 5, when both ends of the resonance tube are open, air particles can move freely at the end, so the amplitudes of pressure waves F and R reach their maximum value. Further, in this case, at the terminal end, the reflected pressure wave It has a reverse phase with respect to the traveling pressure wave IF and is inverted. Therefore, in this case, the wavelength λ of the settled wave that can be generated in the resonant tube I is λ-2L / n (2) (1 echo, n = 1.2, 3...) . FIG. 5 shows the state of standing waves in the case of n=1, 2.3. On the other hand, as shown in FIG. 6, when one end of the resonance tube 1 is closed, air particles cannot move at this end, so the amplitudes of the pressure waves F and R become zero. . Therefore, the wavelength λ of the standing wave in this case is λ-4L/(2n-1) (3) (where n=1, 2, 3, etc.). FIG. 6 shows the state of standing waves in the case of n=1.2.3.

Then, the lead 2 is attached to the resonance tube 1 having such characteristics.
Accordingly, when air vibration with a resonance frequency fn shown in the following equation (4) is applied, a 2-order resonance phenomenon occurs in the resonance tube I.

rn= c/λ (4) (C represents the propagation velocity of pressure waves F and R) In a wind instrument, the reed 2 vibrates in synchronization with the pressure standing wave in the resonant pipe I, so that Resonance in the resonance tube 1 is maintained. In other words, for example, the lead 2 swings in the direction of the arrow 2F, which generates a traveling pressure wave F, which is reflected at the terminal end IE and returns to the lead 2 as a reflected pressure wave R. , the lead 2 swings in the direction of the arrow 2R, generating a traveling pressure wave F, which is reflected at the terminal end IF and returns to the lead 2 as a reflected pressure wave R.
swings again in the direction of arrow 2F, and the vibration of reed 2 and the reciprocating motion of the pressure wave (in other words, the vibration of the standing wave)
and continue in sync.

In this manner, in the wind instrument, the reed 2 and the pressure standing wave of the resonance tube 2 vibrate in synchronization, thereby maintaining resonance and generating musical tones. Here, since the vibration of the reed 2 is a nonlinear vibration, the pressure waves F and R obtained by this vibration contain many harmonic components, and the resonance pipe I is also expressed in the above equations (2) and (3). It has many resonant frequencies. Therefore, air vibrations of many resonance frequencies are obtained within the resonance tube I.

FIG. 7 shows the configuration of a musical tone synthesis device obtained by simulating the sound generation mechanism of a wind instrument as described above. Note that the configuration of this musical tone synthesis device is of course applicable not only to wind instruments but also to other instruments such as stringed instruments.

In FIG. 7, 11 is a nonlinear element that simulates the operation of lead 2, 12 is a resonant circuit that simulates resonance tube 2, and I3 is the above equation (1) performed in lead 2.
This is a subtractor that simulates pressure subtraction. This subtracter 13 subtracts the input signal VA corresponding to the exhalation pressure PA from the output signal from the resonant circuit 12 (the signal corresponding to the reflected pressure wave R described above), and supplies the subtracted signal to the nonlinear element 11 .

According to this configuration, the nonlinear element 11 is DC-biased by the input signal VA. Then, the output signal of the nonlinear element 11 is input to the resonant circuit 12, and the output signal of the resonant circuit I2 is input again via the subtracter I3.
Nonlinear element II is excited. In this way, the circuit of FIG. 7 operates in oscillation.

Here, the nonlinear element 11 is designed so that its input/output characteristics can simulate the nonlinear characteristics of the first and second types described above. Note that this nonlinear element 11 may be realized by a nonlinear element such as a diode, or alternatively, for example, a data table of a desired nonlinear function may be stored in a ROM and read out. By adjusting the input/output characteristics of the nonlinear element 11 to the nonlinear characteristics of the actual lead in this way, it is possible to obtain a waveform that closely matches the vibration waveform of the actual lead as the output signal of the nonlinear element 11, as explained below. .

FIG. 8 shows an example of a reed vibration waveform in an actual clarinet. As shown in this figure, in a wind instrument such as a clarinet, even when producing tones of the same scale, the vibration waveforms change between hits played and soft sounds, resulting in different tones. To explain in more detail, the waveform in the case of a weak tone is relatively close to a sine wave, but when it becomes a playing tone, the amplitude is determined by the elastic limit of the reed within the vibration range L to U.
Since it is limited by L, the peak portion is crushed, resulting in a distorted waveform. This phenomenon can be reproduced as a change in the output signal waveform when the bias point of the nonlinear element 11 is changed by the input signal VA corresponding to the exhalation pressure PA. That is, in the case of a weak performance, the exhalation pressure PA is small, so the bias point of the nonlinear element 11 is set within the linear region, and the output signal of the nonlinear element 11 has a waveform close to a sine wave. On the other hand, in the case of playing, the exhalation pressure PA is large, so the bias point of the nonlinear element 11 falls within the nonlinear region. As a result, a distorted waveform containing many harmonic components is obtained from the nonlinear element I+.

Next, the resonant circuit 12 will be explained in detail. This resonant circuit I2
is designed to match the shape of the resonant tube of the wind instrument to be realized. Here, let us give an example of the transmission frequency characteristics of a resonant tube of an actual natural musical instrument. FIG. 9 shows the transmission amount frequency characteristics of the clarinet tube, and FIG. 10 shows the transmission amount frequency characteristics of the oboe tube. As shown in these figures, the pipe section of a wind instrument has multimodal transmission frequency characteristics with poles at resonance frequencies determined by its shape. The relationship between this resonance frequency and the tube shape is as described above (Equation (2), (
3)). The air vibrations generated by the reeds of each wind instrument are supplied to the resonant pipes having such transmission frequency characteristics, thereby generating musical tones with tones unique to each wind instrument.

FIG. 11 shows an example of a circuit simulating the transmission amount and frequency characteristics of the wind instrument's pipe section, and this circuit is used as the resonant circuit I2 in FIG. 7. In this figure, DF,
-DFn and DR, ~DRn are delay circuits each formed by a multistage shift register (usually three or more stages are used), and simulate the transmission delay of air pressure waves in the pipe. Here, the delay circuits DF and DRn correspond to the portion of the pipe portion closest to the lead 2 side, and the delay circuits DF n and DR correspond to the portion of the pipe portion closest to the terminal portion IE. The output signal of the nonlinear element +1 shown in FIG. 7 is input to the delay circuit DF, and the output horn signal of the delay circuit DRn is input to the subtracter 13 shown in FIG.

Jl, J2. . . is a junction (coupling circuit), which simulates the scattering phenomenon of air pressure waves that occurs at a point where pipes of different diameters are connected. In this figure, a case is shown using a 4-multiplying grid composed of multipliers M1 to M4 and adders A1A2. Here, "I+kn", "-kn", "I-kn", "k
n'' is a coefficient by which the input signal is multiplied by each of the multipliers M1 to M4, and is determined according to the signal scattering characteristics within the wind instrument.
,,J3. Signal transmission between adjacent ones of the delay circuits DF1 to DFn and Drt, to DRn is performed via the delay circuits DF1 to DFn and Drt, to DRn. For example, the output signal of the delay circuit DF is sent to the delay circuit DI via the multiplier M1 of the junction J.
? The output signal of the delay circuit DRn-1 is sent to the delay circuit DRn via the multiplier M3 of the junction J.

The above-mentioned Japanese Patent Application Laid-Open No. 63-40199 describes junctions J, , J2. ... discloses an example in which, in addition to the above-mentioned 4-multiply lattice, a 3-multiply lattice shown in FIG. 12 is used. In this figure, M5 to M7 are multipliers, A
3 to Δ5 are adders, and IV2 is an inverting circuit. here,
kn indicates a coefficient multiplied by the input signal by multiplier M7. Similarly, gm and 1/gm indicate coefficients by which the input signal is multiplied by multipliers M5 and M6, respectively.

However, gm= [(1-kn)/(1+kn))
gm is selected to be 0.5--(5). By doing so, the transmission gain is normalized.

In FIG. 11, TRM is a termination circuit, and resonance tube 1
This is a simulation of the terminal IE of. The signal output from the nonlinear element 11 is sent to the delay circuits DF, -DFn.
and junctions J, , J inserted in each delay circuit
, . . . and then input to this termination circuit TRM.

ML is a simulation of energy loss when a pressure wave is reflected at the termination part IE, and the output signal of the preceding stage delay circuit DFn is multiplied by the loss coefficient g& and outputted to the phase inverter IV. The phase inversion circuit IV simulates a phenomenon in which the phase of the reflected wave is inverted with respect to the traveling wave when the terminal end IE of the wind instrument is open. Note that if the terminal end IE is closed, this phase inverter IV is not necessary. Then, the DC component of the output signal of the phase inverter IV is removed by the DC removal circuit DCB, and the signal is input to the delay circuit DR1. And delay circuit D
The adder 1 in FIG. 7 is connected to the adder 1 in FIG.
3 is input.

In this configuration, delay circuits DF, ~DFn and DR
, -DRn is determined in accordance with the frequency of the generated musical tone. Furthermore, in an actual wind instrument, the time required for the propagation of the traveling pressure wave F (from the reed to the end) and the time required for the propagation of the reflected pressure wave R (from the end to the reed) are considered to be the same. , delay circuit D
Total delay time of F, ~DFn and delay circuits DR, ~DR
The total sum of n delay times is designed to be approximately equal.

As described above, the nonlinear element II and the resonant circuit I2 are constructed by simulating each part of an actual wind instrument, and a desired musical tone of a wind instrument is synthesized. In addition, musical instruments other than wind instruments,
For example, in the case of a stringed instrument such as a guitar, the nonlinear element 11 is designed in accordance with the elastic characteristics of the string, and the resonant circuit 12 is designed in accordance with the length of the string. Furthermore, as in the case of musical instruments, a resonant circuit can be used to construct a residual effect device.

``Problems to be Solved by the Invention'' By the way, when a musical tone synthesis device is realized by arithmetic processing of a signal processor, the conventional musical tone synthesis device uses a 4-multiplying lattice or a 3-multiplying lattice as a junction. There are many multiplications. Therefore, there has been a problem in that the signal processor is required to have high processing ability in order to achieve the required signal processing speed. In addition, when trying to realize junctions with various transmission characteristics by changing the coefficients of each multiplier that constitutes a junction to various values, it is necessary to use not only the above-mentioned 4-multiplying lattice and 3-multiplying lattice but also circuits with other configurations. Using it as a junction also allows for a richer variety of signal processing.

The present invention has been made in view of the above-mentioned circumstances, and it is an object of the present invention to provide a musical tone synthesis device that requires a short calculation time per junction.

It also aims to realize a rich variety of signal processing in musical tone synthesizers.

"Means for Solving the Problem" The first invention provides an excitation means for outputting an excitation signal based on an input signal and a feedback signal; means for transmitting the traveling wave signal and the reflected wave signal at the terminal end to the excitation means, m = a plurality of means for transmitting the traveling wave signal and the reflected wave signal for a predetermined time and transmitting the reflected wave signal in the direction of propagation of the signal; It consists of a delay path and a bidirectional transmission means constituted by a signal scattering junction that mediates signal transmission between the plurality of delay circuits, and by bringing the excitation means and the bidirectional transmission means into a resonant state, musical tones can be produced. A musical tone synthesis device configured to obtain a signal is characterized in that a square lattice is used as a signal scattering junction in the bidirectional transmission means.

Further, a second invention is characterized in that a 1-multiply lattice is used as a signal scattering junction in the bidirectional transmission means.

Further, a third aspect of the present invention is characterized in that a 4-multiply normalization lattice is used as a signal scattering junction in the bidirectional transmission means.

"Operation" According to the first and second inventions, the number of multiplications per signal scattering junction can be small. Further, in the third invention, since the transmission gain of the signal scattering junction is normalized, it is easy to control the transmission gain in the bidirectional transmission means. Furthermore, according to the first to third inventions, it is possible to realize a wide variety of signal processing variations.

"Embodiment" Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a musical tone synthesizer according to an embodiment of the present invention. In this figure, parts corresponding to those in FIGS. 7 and 11 described above are designated by the same reference numerals 4-f. In the same figure, 21 detects the operation of various operators (not shown) equipped on the instrument body,
Accordingly, musical tone control information (scale, degree of blowing strength,
This is a musical tone control information generation circuit that generates note-on, note-off, etc.). 22 is an excitation circuit, which includes a nonlinear element 11 and a subtracter 13 having the same configuration as in the case of FIG. 7 described above.
Consisted of. Here, information indicating the strength of the blowing, which is supplied from the musical tone control information generation circuit 21, is supplied to the nonlinear element 11 as a DC bias VA.

23 is a resonant circuit that simulates a resonant tube. The resonance frequency of this resonance circuit 23 is the same as that of the musical tone control information generation circuit 2.
Switching is performed according to information indicating the scale supplied from 1 to 1. This resonant frequency switching is performed by the delay circuits D F , , D F 7 . in the resonant circuit 23 . ..., D Rl+DI'(
This is realized by switching the effective number of stages 2, . . . The simplest method for realizing this is, for example, by switching the connection destination (output light) of the nonlinear element 11 to the delay circuit DFi (i=
1.2,...), and subtractor 1 accordingly.
3 connection destination (input signal generation source) is DRjU=n, n-1
,...) can be used. 1st
In the figure, nonlinear element 11→resonant circuit 23−subtractor I
3 - In the closed loop formed by the nonlinear element It,
There is a risk of a free-running oscillation state due to noise, etc. Therefore, in this musical tone synthesizer, the output of the nonlinear element 11 is enabled only when a note is on, thereby stopping free-running oscillation when it is not necessary (when a note is off).

JA,, JA2. ... is a junction. These junctions include the 4-multiplying lattice and 3-multiplying lattice mentioned above, as well as the 2-multiplying lattice often used in the field of speech synthesis.
Multiplying lattice (see Figure 2(a)), 1-multiplying lattice (see Figure 2(a))
(see Fig. 2(c)), and a 4-multiply normalized grid (see Fig. 2(c)).

Now, the 4-multiply lattice shown in FIG. 11, the 3-multiply lattice shown in FIG. 12, the 2-multiply lattice shown in FIG. 2(a), and the 2-multiply lattice shown in FIG.
The 1-multiply normalized lattice shown in FIG. 2 and the 4-multiply normalized lattice shown in FIG. 2(c) have the same transfer characteristics when the coefficient kn in each circuit is the same value.

Therefore, the same musical tone output can be obtained regardless of which of these circuit configurations is used for the junctions JA, JAR, . . . in the musical tone synthesizer shown in FIG.

However, when signal processing in this musical tone synthesizer is performed by calculations by a signal processor, etc., the square grid (Figure 2 (a)) is used rather than the quadruple grid or the triple grid.
and 1 multiplication grid (Fig. 2(b)) at junction J
A1. JA2. ..., the number of multiplications required for signal processing is smaller. Therefore, in this case, the computational power and processing speed required of the signal processor are relaxed.

In the 4-multiply normalized lattice shown in FIG. 2(c), the transmission gain is normalized. Therefore, when this 4-multiply normalized lattice is used as a junction J A l + J A z, . . . , there is a risk that the signal level in each part of the resonant circuit 23 may become extremely high or, conversely, become extremely low. do not have. Therefore, level adjustment of the signal within the resonant circuit 23 becomes easy.

By independently changing the coefficients of each multiplier in each of the above-described junctions, it is possible to variously change the transmission characteristics of the junction. Here, it is better to use not only the 4-multiplying lattice and the 3-multiplying lattice but also the configuration shown in FIG. 2(a) or (b) as the junction configuration to realize more types of transmission characteristics. be able to.

By doing so, it becomes possible to perform delicate musical tone control that has not been possible in the past.

In addition, in the above-mentioned embodiment, the traveling wave delay circuit D F
, D F , . . . and reflected wave delay circuits DR, , D R2, . It does not matter if the delay times are unbalanced. In this case, by making the sum of the delay times of both the traveling wave and reflected wave delay circuits equal to that of this embodiment, a musical tone synthesizer that generates musical tones equivalent to that of this embodiment can be constructed. can do. Further, in this embodiment, a case has been described in which the present invention is applied to a musical tone synthesis device that simulates a wind instrument, but the present invention is not limited to this, and can be applied to a reverberation effect device, etc. Further, in this embodiment, the input signal VA and the feedback signal are input to the nonlinear element 11 via the subtracter 13, but the input signal VA and the feedback signal may be directly applied to the nonlinear table.

"Effects of the Invention" As explained above, according to the first and second inventions,
Since the number of multipliers in the signal scattering junction can be reduced, the amount of required calculations can be reduced when the signal processing of the musical tone synthesizer is performed by calculations by a signal processor or the like. Therefore, in this case, the effect of easing the computational power, speed, etc. required of the signal processor can be obtained. Further, according to the first to third inventions, various musical tone controls can be performed by selectively using the configurations of these inventions as necessary.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a musical tone synthesizer according to an embodiment of the present invention, and FIG. 2 shows junctions J A +, J A 2 . in the same embodiment. Figure 3 is a diagram showing the simplest model of a wind instrument; Figure 4 is a diagram showing the nonlinear characteristics of reed 2 in Figure 3; Figures 5 and 6 are diagrams showing a resonance pipe model. Figure 7 is a block diagram of a conventional tone synthesizer simulating a wind instrument, Figure 8 is a vibration waveform diagram of a clarinet reed, and Figure 9 is the transmission amount frequency characteristics of a clarinet tube. , FIG. 10 is a diagram showing the transmission amount frequency characteristics of the oboe tube, FIG. 11 is a block diagram showing an example of the configuration of the resonant circuit 12 in FIG. It is a figure showing an example of composition. 11...Nonlinear element, 13...Adder,
23・Resonant circuit, D F 1. DF? + ~, D R
+, D R2, ~...Delay circuit, J Al,
JA2. 〜・・・Junction.

Claims (3)

[Claims] (1) An excitation means that outputs an excitation signal based on an input signal and a feedback signal, and transmits the output signal of the excitation means as a traveling wave signal toward a terminal end, and excites a reflected wave signal at the end end. a plurality of delay circuits that delay the traveling wave signal and the reflected wave signal for a predetermined time and transmit the signals in the traveling direction of the signals; and a signal that mediates signal transmission between the plurality of delay circuits. A musical tone synthesizer comprising a bidirectional transmission means constituted by a scattering junction, and configured to obtain a musical tone signal by bringing the excitation means and the bidirectional transmission means into a resonant state, wherein the signal in the bidirectional transmission means A musical tone synthesis device characterized in that a squared lattice is used as a scattering junction. (2) The musical tone synthesis apparatus according to claim 1, wherein a 1-multiply lattice is used as a signal scattering junction in the bidirectional transmission means. (3) The musical tone synthesis apparatus according to claim 1, characterized in that a quadruply normalized lattice is used as a signal scattering junction in the bidirectional transmission means.