JPH04181296A - Musical sound synthesizer - Google Patents

Abstract

PURPOSE:To obtain expressing ability in brass instrument performance which is extremely close to real musical sound by supplying an exciting signal to a 1st waveguide network. CONSTITUTION:The 1st waveguide network 10 which is an oral cavity/glottis forming circuit simulates the shape of either one of the oral cavity and glottis. Then the device is equipped with a mousepiece simulating circuit as a 2nd waveguide network 11 which simulates the mousepiece or reed of a wind instrument and is cascaded to the network 10 and a tube body forming circuit which simulates the tube body of a brass instrument and is cascaded to the network 11. Then the network 10 supplies the exciting signal to reflect the influence of a wave motion, propagated to the mouse or glottis of a player, upon a musical instrument, any difference from a real musical instrument is not generated, and the expression ability in brass instrument performance which is extremely close to real one can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a musical tone synthesis device suitable for use in electronically synthesizing musical tones of a wind instrument or the like.

"Prior Art" A musical tone synthesis device is known that operates a model obtained by simulating the sound generation mechanism of a natural musical instrument, thereby generating musical tones of a natural musical instrument. and,
Simulations of tubular bodies have been carried out using linear circuits and nonlinear circuits using waveguide digital filters that describe the tubular bodies (Japanese Patent Application Laid-Open No. 63-40199, etc.).

``Problem to be solved by the invention'' By the way, in the conventional musical tone synthesis device described above, only the simulation from the player's mouth to the pipe body is performed, and the wave propagating to the player's mouth, vocal tract, etc. The impact was ignored. Therefore, there is a problem that there is a discrepancy with the behavior of an actual musical instrument, resulting in poor reproducibility.

This invention was made in view of the above circumstances,
It is an object of the present invention to provide a musical tone synthesis device capable of obtaining the expressive power of wind instrument performance that is extremely close to reality.

"Means for Solving the Problems" In order to solve the above problems, a first waveguide network that simulates the shape of at least one of the oral cavity and the vocal tract, and at least the mouthpiece and reed of a wind instrument are provided. a second waveguide network that simulates one of the first waveguide networks and is connected in cascade to the first waveguide network; A third waveguide network is connected thereto, and an excitation signal is supplied to the first waveguide network.

"Operation" According to the above configuration, the oral cavity and vocal tract are simulated by the first waveguide network, and the influence of waves propagating to the player's mouth, vocal tract, etc. on the musical sound is reflected.

"Example" Hereinafter, the present invention will be described in detail with reference to the drawings.

(1) Configuration of Embodiment FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, 1 is a keyboard, which is composed of white keys and black keys. Reference numeral 2 denotes an operation panel on which various parameters for controlling the timbre of musical tones can be set. A CPU (central processing unit) 3 executes a predetermined program and controls each part of the electronic musical instrument. The table ROM 4 stores an operation program for the CPU 3 and coefficients to be applied to a musical tone synthesis circuit 7, which will be described later. Further, the RAM (random access memory) 5 is a CP
It is used as a temporary storage area during calculation by U3. 6
are operators for performing various operations, and control signals corresponding to the operations are supplied to the CPU 3. The musical tone synthesis circuit 7 is constituted by a circuit that simulates a natural musical instrument, and synthesizes musical tone signals under the control of the CPU 3.

Reference numeral 8 denotes a sound system, which causes a speaker 9 to sound based on the musical tone signal outputted from the musical tone synthesis circuit 7.

Next, the above-mentioned musical tone synthesis circuit 7 will be explained with reference to the block diagram shown in FIG. In this figure, the musical tone synthesis circuit 7 is connected to the body of a wind instrument such as a clarinet, the mouthpiece of a wind instrument, and the oral cavity.

It consists of a closed loop circuit that simulates the vocal tract. That is, as shown in FIG. 2, an oral vocal tract formation circuit 10 that simulates the oral cavity and vocal tract, a mouthpiece simulating circuit 11 that simulates the state of the mouthpiece and reed, and a pipe body portion of a wind instrument are simulated. It is composed of a tube forming circuit 12.

The mouthpiece simulation circuit 11 includes adders 15 to 1.
9, multipliers 20 and 21, a delay 22, and a function generator 23, and performs, for example, the following simulation.

Here, we will take the case of single read as an example.

Now, as shown in FIG. 3, the pressure inside the mouthpiece MP is q, the pressure inside the mouth M is p, and these are further divided into two as shown in the following equation.

q = q I+ q 0 ・・・・・・・・・
・・・・・・・・・ (1) p = I) I+ p-
・・・・・・・・・・・・・・・ (2) Then, if the characteristic impedance of the pipe body is 27 and the characteristic impedance of the inside of the oral cavity is ZM, then the volumetric flow rate f at the connection part between the mouth and the instrument is is equal in the tube body and the oral cavity, so f=(q.-ql)/Z↑=(p+-1). ) / Z
M・・・・・・・・・・・・(3). Transforming this, we obtain the following equation.

q−=q++Zyf ・・・River・・・・・・
(4)p-=I)+ZM-f...
・・・・・・・・・(5) Here, f is a nonlinear function of the difference Δq between the internal pressure and external pressure of the mouthpiece, and f=F(Δq) ・・・・・・・・・・・・・・・
・・・(6)Δq=q−p ・・・・・・・・・
......(7). The mouthpiece simulation circuit shown in FIG. 2 realizes the above-mentioned equations (1) to (7). Note that the circuit configuration of the mouthpiece simulation circuit 11 differs depending on the mode of simulation.

Here, the mouthpiece simulation circuit 1 shown in FIG.
1 and the above equations (1) to (7) are explained as follows.

According to the above simulation, the oral vocal tract formation circuit 10 is supplied with the signal p0 and outputs the signal p1, and the tube formation circuit 12 is supplied with the signal q0 and outputs the signal q1. Become. In other words, the input/output relationship is as shown in FIG. Therefore, the addition result of adder 18 corresponds to equation (1), and the addition result of adder 16 corresponds to equation (2).
) corresponds to the expression. Then, the adder 17 performs the calculation (qp) of equation (7), and the result of this calculation is supplied to the function generator 23 via the delay 22, where the calculation of equation (6) is performed. It will be done. Next, the output signal of the function generator 23 is multiplied by a coefficient Zv by the multiplier 20 and then input to the adder 15. In the adder 15, the input signal and Pl are added (algebraic addition), and as a result,
The calculation shown in equation (5) is performed. Also, function generator 2
The output signal of No. 3 is input to the adder 19 after being multiplied by a coefficient ZT by the multiplier 21 . In this adder 19, the signal q is added to the input signal, so that the calculation shown in equation (4) is performed. The above is the processing content of the mouse simulation circuit 11.

Next, the configuration of the tube forming circuit 12 will be explained with reference to FIG. 4. In the figure, JO-T to J+-IT are junctions, respectively, and have the configuration shown in FIG. 6. In FIG. 6, 25, 26, and 27 are adders, and 28, 29 are multipliers. Multiplier 28.29
The coefficients of are each a.

b, and when simulating the open state of a tone hole with no height, a + b < 2, and when simulating the closed state of a tone hole with no height, a + b
=2. Further, those indicated by the symbol SR and a suffix in FIG. 4 are shift registers, and the number of shift stages of each is controlled by the CPU 3. The mode of simulation changes depending on the difference in the number of shift stages. Further, the LPF shown in FIG. 4 is a low-pass filter that approximates the loss at the end of the tube, and γ7 is the reflection coefficient at the end of the tube.

Next, the configuration of the oral vocal tract forming circuit 10 will be explained with reference to FIG.

J, -4 to JOM shown in FIG. 5 are junctions, and like the above-mentioned junctions, they are constructed of the circuit shown in FIG. 6. However, the multiplication coefficient is set to a+b=2 when the nasal cavity is not considered. In FIG. 5, the symbols SR and suffix indicate shift registers, which are similar to those shown in FIG. 4. Also, L
PF is a low-pass filter that approximates the loss at the end of the vocal tract, and 7M is a reflection coefficient at the end of the vocal tract. The symbol pr shown in FIG. 5 is the force forcibly applied to the system by the performer. In this embodiment, a signal pr corresponding to the pressing force of the keyboard is used.

The coefficients a and b shown in FIG. 6 are stored in advance in the ROM 4 shown in FIG. 1, read out as appropriate under the control of the CPU, and supplied to each junction. Here, the method for determining the coefficients will be explained.

First, the determination of coefficients in the oral vocal tract formation circuit 10 is as follows:
This is done by predicting the shape of the oral cavity.

For example, as shown in FIG. 7, human speech is recorded with a microphone 30, and the recorded waveform is processed by a linear prediction circuit 31 to determine filter coefficients. In this case, a set of coefficients is obtained when the vocal tract and oral cavity are 3-rated with a filter, and this set of coefficients is written into the ROM 4.

Furthermore, in determining the coefficients, coefficients for various pronunciations are determined and written in the ROM 4 all together in order to accommodate the diversification of performance methods.

Here, a specific example of filter coefficients written in the ROM 4 is shown in FIG. As shown in this figure, combinations of coefficients are written separately for forms A, B, C, etc. of the oral vocal tract to be simulated.

Further, the coefficients at each junction of the tube forming circuit 12 are set in advance so that the coefficient corresponding to the tube to be simulated is R.
Stored in OM4.

(2) Operation of the embodiment Next, the operation of this embodiment with the above configuration will be explained.

First, coefficients to be supplied to the oral vocal tract formation circuit 10 are selected using the operator 6. By this operation, ROM4
One of the coefficient sets A, B, C, . . . is selected. Further, the coefficients for the tube forming circuit 12 are also selected by operating the operator 6. Next, when the operator performs a performance using the keyboard, a signal indicating the pressing force of the pressed key is supplied from the keyboard l to the CPU 5. Thereby, the CPU 3 creates a signal pr and supplies it to the oral vocal tract formation circuit 10 in the musical tone synthesis circuit 7 (see FIG. 5). Further, the CPU 3 reads out the selected coefficient set from the ROM 4 and supplies it to each junction in the oral vocal tract formation circuit 10. The oral vocal tract forming circuit 10 performs a simulation according to the supplied coefficients, and starts forming a musical tone signal using the signal pr as an excitation signal. And oral vocal tract formation circuit 10. A loop circuit constituted by the mouthpiece simulation circuit 11 and the tube body forming circuit 12 allows
The oral cavity, vocal tract, mouthpiece and tubular body are simulated, and the musical tone signal obtained as a result of this simulation is supplied to the sound system 8.

As a result, musical tones are emitted from the speaker 9.

In the above operation, the wave propagating to the player's mouth, vocal tract, etc. by the oral vocal tract formation circuit is also simulated, so the influence of this wave on the musical tone is reflected.

(3) Modifications In the above embodiment, the following modifications are possible.

(2) Although the set of coefficients to be applied to the oral vocal tract formation circuit 10 is selected by the operator 6, it may be configured to be selected by the initial touch or aftertouch of the key instead. When playing a wind instrument, the shape of the performer's oral cavity and vocal tract changes at any time depending on the sound being uttered, so rather than setting the simulation mode fixedly, it is better to A musical tone that is closer to the actual performance will be produced if the change is made accordingly.

Further, although the embodiment is an example of an electronic musical instrument using a keyboard, the present invention can also be applied to an electronic musical instrument using a mouthpiece.

In this case, the configuration may be such that the coefficient to be applied to the oral vocal tract forming circuit 10 is selected depending on the strength with which the mouthpiece is bitten (ampsure).

■The shift register SR used in the example is a RAM
Alternatively, other delay means may be used.

(2) Not limited to the pronunciation algorithm used in the embodiment, other algorithms, such as pronunciation algorithms based on string plucking or string striking, may be used. In this case, when the string is plucked, it can be thought of as the vibration of the hair of a bow, for example.

(2) Although the embodiments described above were realized by hardware, they may also be realized by microprograms or software.

(2) Each coefficient supplied to the oral vocal tract formation circuit 10 and the tube body formation circuit 12 may be key scaled. That is,
It may be configured such that each coefficient is allocated in advance according to the pitch range, and a set of coefficients corresponding to the played pitch is selected at any time.

■When the tube formation circuit 12 simulates a tone hole with a height, and when the oral vocal tract formation circuit 10 also considers the branching of the nasal cavity, the junctions shown in FIG. Become.

■To change the shape of the oral cavity as smoothly as possible,
Create a table corresponding to Figure 8 so that the shape of the oral cavity and vocal tract changes little by little in the order of A, B, C, etc., and the values of the operators corresponding to changes in the vocal tract change greatly. However, instead of suddenly achieving the desired vocal tract shape, the value JIM a gradually shows the shape by using intermediate table values through interpolation.

b.J! You may change a, b... of M.

"Effects of the Invention" As explained above, according to the present invention, there is provided a first waveguide network that simulates the shape of at least one of the oral cavity and the vocal tract, and at least one of the mouthpiece and reed of a wind instrument. a second waveguide network that simulates one of the first waveguide networks and is cascade-connected to the first waveguide network; and a second waveguide network that simulates a wind instrument body and is cascade-connected to the second waveguide network. Since the first waveguide network is provided with a third waveguide network and an excitation signal is supplied to the first waveguide network, the influence of waves propagating to the player's mouth, vocal tract, etc. on the musical sound is reflected. , there is no discrepancy with the behavior of real musical instruments, and it is possible to obtain the expressive power of wind instrument performance that is extremely close to reality.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of a mouthpiece simulating circuit 11 of the same embodiment, and FIG. FIG. 4 is a block diagram showing the configuration of the tube formation circuit 12 of the same embodiment, and FIG. 5 is a block diagram showing the configuration of the oral vocal tract formation circuit 10 of the same embodiment. , FIG. 6 is a block diagram showing the structure of the junction between the oral vocal tract forming circuit 10 and the tube forming circuit 12, and FIG. 7 shows the structure for determining the coefficients for the oral vocal tract forming circuit 10 in the same embodiment. The block diagram, FIG. 8, is a conceptual diagram showing the state of a set of coefficients stored in the ROMJ. 10...... Oral vocal tract formation circuit (first waveguide network), 11...... Mouthpiece simulation circuit (second waveguide network), 12...Tube forming circuit (third waveguide network).

Claims (1)

[Scope of Claims] A first waveguide network that simulates the shape of at least one of an oral cavity and a vocal tract; and a first waveguide network that simulates at least one of a mouthpiece and a reed of a wind instrument; a second waveguide network connected in cascade to the waveguide network; and a third waveguide network simulating the body of a wind instrument and connected in cascade to the second waveguide network. and supplying an excitation signal to the first waveguide network.