JPH06167981A - Musical sound synthesizing device - Google Patents

Abstract

PURPOSE:To provide the musical sound synthesizing device which has high fidelity although it can be manufactured at low cost. CONSTITUTION:Similarly to a general physical model sound source, various signals are exchanged between a linear part 104 and a nonlinear part 204 and then a musical sound is synthesized. The nonlinear part 204 is composed of an analog circuit to improve the stability and fidelity. Further, a junction 102 and the linear part 104 are composed of digital circuits to simplify the constitution.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone synthesizer suitable for use in electronic musical instruments.

[0002]

2. Description of the Related Art Heretofore, there has been known a musical tone synthesizing device for synthesizing a natural musical instrument sound by operating a model simulating a sounding mechanism of a natural musical instrument. As an example thereof, FIG. 3 shows the configuration of a musical tone synthesizer using a wind instrument physical model.

In the figure, the musical tone synthesizer comprises a non-linear section 101 for simulating the lead section of a wind instrument, a linear section 103 for simulating a resonance tube, a junction 102 interposed between the two sections, and a sound system 75. It consists of Inside the non-linear unit 101, 51 is a subtractor, which is a blowing pressure signal PRESS indicating the blowing pressure of the performer.
And the pressure signal q indicating the air pressure in the mouthpiece is supplied from the linear portion 103 via the junction 102. In the subtractor 51, the latter is subtracted from the former, and the subtraction result is output as the pressure difference signal Δq. The pressure difference signal Δq is a value indicating the difference between the pressure in the mouth and the pressure in the mouthpiece during the performance of the natural wind instrument, and after the high frequency component is removed via the averaging filter 52, the reed filter. 53 and the Graham function table 54.

The read filter 53 is a second-order low-pass filter, and its amplitude increase ratio Q and resonance frequency fc are externally given as parameters. Reference numeral 55 denotes a multiplier, which multiplies the output signal of the read filter 53 by the compliance (gain) C and supplies the multiplication result to the adder 56. That is, the reed filter 53 and the multiplier 55 simulate the frequency characteristic of the reed portion of the natural wind instrument, and the multiplier 55 outputs the pressure difference signal Δq ′ to which the frequency characteristic is added.

Next, a slit function table 57 stores the displacement of the opening area of the reed corresponding to the pressure of the reed portion in the wind instrument, that is, the slit function. Therefore, if the output signal of the multiplier 55 is supplied to the slit function table 57, the displacement of the opening area of the lead considering the frequency characteristic is output. But,
In the illustrated circuit, the multiplier 55 and the slit function table 57
An adder 56 is inserted between the two, and the output signal of the multiplier 55 is added with the embouchure signal emb. Here, the embouchure signal emb is a signal obtained by previously performing an inverse function of the slit function on the opening area of the lead (hereinafter referred to as aperture apr) when there is no vibration. Therefore, the slit function table 57 outputs the opening area signal S _R indicating the opening area of the lead in consideration of vibration.

Here, according to Graham's law, when the difference between the pressure in the mouth and the pressure in the mouthpiece is Δq, the flow rate (air flow in the unit area per unit time at the opening of the reed (air The velocity v) is expressed by the following equation (1). v = {2 (Δq) / ρ} ^1/2 ≡L (Δq) (1)

Equation (1) is called a Graham function. When the pressure difference signal Δq is supplied to the Graham function table 54, the expression
The flow velocity signal v is output based on (1). Next, the flow velocity signal v and the opening area signal S _R are multiplied in the multiplier 58. This multiplication result is the volume flow velocity signal f indicating the flow rate of the air flowing through the opening of the mouthpiece per unit volume per unit time.
It is output as [cm ³ / s]. The volume flow velocity signal f includes
In the multiplier 59, the constant z is further multiplied. Here, the constant z is a resistance to an air flow of the mouthpiece and the resonance tube of the wind instrument, that is, a proportional constant between the volume flow velocity and the air pressure. Therefore, the multiplier 59 outputs the signal zf indicating the air pressure. This signal zf is added to the reflected wave pressure signal q _i via the adder 61 of the junction 102, and the addition result is supplied to the linear unit 103 as the traveling wave pressure signal q ₀ .

Next, in the linear portion 103, 70 is a delay circuit, which simulates the propagation delay of the pressure traveling wave from the mouthpiece of the natural musical instrument to the end of the resonance tube. Further, 71 is an LPF (low pass filter),
Simulate attenuation of high frequency components during propagation of air pressure wave. 72 is a multiplier, in which the traveling wave pressure signal q ₀ output from the LPF 71 is multiplied by “−1”. This simulates that the pressure traveling wave that has reached the opening in the resonance tube is reflected, and the phase is inverted before propagating toward the mouthpiece as a pressure reflected wave. The reflected wave pressure signal q _i output from the multiplier 72 is again propagated and delayed via the delay circuit 73, and then supplied to the junction 102.

The adder 62 in the junction 102 adds the traveling wave pressure signal q ₀ to the reflected wave pressure signal q _i , and outputs the addition result as a new pressure signal q. Next, in the subtractor 51, the blowing pressure signal PRESS at that time point is subtracted from this new pressure signal q, and the subtraction result is the new pressure difference signal Δ.
It is output as q and supplied to the read filter 53 and the Graham function table 54 through the averaging filter 52.

As described above, various signals propagate through the loop composed of the non-linear portion 101, the junction 102 and the linear portion 103, so that the sounding mechanism in the natural wind instrument is simulated. The traveling wave pressure signal q ₀ is then supplied to the sound system 75 and sounded as a musical tone.

[0011]

There are various possible uses of the electronic musical instrument using the above-described musical tone synthesizer, and as one example, it is expected to be used for practicing the performance of a natural musical instrument. . For example, playing a natural wind instrument that emits a large volume, such as a trumpet or a saxophone, in a densely populated residential area creates various problems. Is desirable. An electronic musical instrument used for such an application is not sufficient to simply generate a musical tone that is faithful to a natural musical instrument, and needs to be able to be played in the same manner as a natural musical instrument. That is, in the case of a wind instrument, it is necessary to control the musical sound based on the player's breath pressure, lips posture, tension, and the like.

In order to make this possible, a technique for measuring various parameters in a playing state and converting the measured parameters into the above-mentioned parameters Q, fc, C, emb is required. Here, the outline will be briefly described with reference to FIGS. 4 and 5. First, FIG. 4 shows an input section simulating a mouthpiece, and the pressure p at which the lips 80, 80 of the performer bite the mouthpiece 81, the lips 80 and the leads 82.
Of the contact area S _L , the aperture apr, and the distance x from the tip of the lead 82 to the position where the lip 80 is in contact (a sensor for measurement is not shown). When these measurement results are supplied to the arithmetic circuit shown in FIG.
Each parameter Q, fc, C, em required in
b is output.

By the way, in the circuit shown in FIG. 3, there is a delay-free loop. That is, as indicated by the broken line A in FIG.
It is a loop that returns to the non-linear unit 101 again via 2. Therefore, there is a risk that parasitic oscillation will occur in the loop. Although the averaging filter 52 is provided in order to prevent this, parasitic oscillation may occur nevertheless. For example, when an attempt is made to synthesize a hard sound by raising the feedback gain to some extent, it is extremely difficult to synthesize the original musical sound because parasitic oscillation occurs.

As a method of eliminating such a defect, it is considered to prevent the parasitic oscillation by increasing the sampling frequency and improving the accuracy of the simulation. However, in order to obtain a practically sufficient sampling frequency, an expensive processor is required, and it is practically extremely difficult to use it in consumer equipment. Further, as shown in FIGS. 4 and 5, in order to enable the performance by the same playing method as that of the natural musical instrument, various sensors and complicated arithmetic circuits are required, which makes the electronic musical instrument more expensive. . The present invention has been made in view of the above-mentioned circumstances, and an object thereof is to provide a musical tone synthesizer which can be manufactured at low cost and has high fidelity.

[0015]

In order to solve the above problems, according to the structure of claim 1, a digital linear circuit for delaying and outputting a supplied digital signal, and an output signal of the digital linear circuit. Converting an analog signal into an analog signal, an analog nonlinear circuit for giving frequency characteristics to the analog signal output from the first converting means and outputting the analog signal, and an output signal of the analog nonlinear circuit is digitalized. Second conversion means for converting into a signal and supplying it to the digital linear circuit is provided.

Further, in the structure according to claim 2,
A mouthpiece having a lead, a flow velocity sensor for detecting an air flow velocity in the mouthpiece, a linear circuit for amplifying and delaying an output signal of the flow velocity sensor, and outputting the amplified signal, and the linear circuit provided in the mouthpiece. And a vibrating means that vibrates based on the output signal of 1.

[0017]

When the digital signal is supplied to the digital linear circuit, the digital signal is delayed and then output. The output digital signal is converted into an analog signal via the first conversion means,
Supplied to an analog nonlinear circuit. The signal supplied to the analog non-linear circuit has a second characteristic after the frequency characteristic is added.
Is converted into a digital signal through the conversion means and is supplied to the digital linear circuit. As described above, since the linear circuit is composed of the digital circuit, the delay circuit is composed at a low cost, and since the nonlinear circuit is composed of the analog circuit, abnormal oscillation does not occur and the stability is high.

Further, in the structure according to claim 2,
When the performer blows into the mouthpiece, the flow velocity of the airflow in the mouthpiece caused by the breath is detected by the flow velocity sensor. The output signal of the flow velocity sensor is amplified through a linear circuit and then delayed to vibrate the vibrating means.
When the vibrating means vibrates, a compression wave of air is generated around the vibration means, and the compression wave becomes a pressure reflection wave and propagates in the mouthpiece to reach the lead and vibrates the lead. When the reed vibrates, a compression wave of air is generated around the reed, and the compression wave propagates in the mouthpiece as a pressure traveling wave. Then, a local change in the flow velocity of the air occurs due to the pressure traveling wave and the pressure reflected wave, and this change is detected via the flow velocity sensor.

[0019]

EXAMPLES A. Prerequisites First, the present invention's embodiment, in order to solve the problems described above, have attempted to configure the musical tone synthesizing apparatus by an analog circuit. The resulting circuit is shown in FIG. In the figure, the musical sound synthesizer includes a non-linear unit 201, a linear unit 203,
It is composed of a sound system 75.

In the non-linear section 201, 151 is an analog subtractor, and when both the pressure signal q and the blowing pressure signal p are supplied as voltage signals, the latter is subtracted from the former,
The subtraction result is output as a pressure difference signal Δq which is a voltage signal. Next, reference numeral 153 is a read filter, which is a second-order low-pass filter capable of designating the voltage increase ratio Q and the resonance frequency fc. Note that the read filter 153
Is preferably realized by a biquad filter. The pressure difference signal Δq ′ filtered by the reed filter 153 is supplied to the adder circuit 156 where it is added to the voltage signal e indicating the embouchure. And
When the addition result is supplied to the slit function calculation circuit 157, the aperture area signal S _R is output.

Next, 154 is a Graham function operation circuit, and when the pressure difference signal Δq is supplied, the flow velocity signal v is output as a voltage signal based on this. Next, when the opening area signal S _R and the flow velocity signal v are supplied to the analog multiplier 158, they are multiplied and the volume flow velocity signal f that is the multiplication result is output as a voltage signal.

Next, 160 is a voltage / current converter,
The volume flow velocity signal f is converted into a current signal and output. This current signal is multiplied by the constant z in the analog multiplier 159, and the multiplication result zf is supplied to the linear unit 203 as a current signal. For convenience of explanation, the voltage / current converter 16
Although 0 and the analog multiplier 159 are separated, it goes without saying that both may be integrated. next,
The linear portion 203 is composed of a ladder circuit including a plurality of capacitors and coils.

Therefore, when the current signal zf is supplied to the one end 203a of the linear portion 203, the one end 203a changes to the other end 203a.
A voltage traveling wave is generated toward b. Here, the other end 203
In b, since the circuit is short-circuited, the other end 2
The voltage traveling wave reaching 03b is reflected and propagated toward the one end 203a. Then, the voltage appearing at the one end 203a is supplied to the analog subtractor 151 as the voltage signal q, and the blowing pressure signal p is subtracted therein. Thus, it can be seen that the linear unit 203 functions as a delay circuit.

In the above configuration, the analog subtractor 151
When the pressure signal q indicating the air pressure in the mouthpiece and the blowing pressure signal p indicating the blowing pressure of the performer are supplied to, the latter is subtracted from the former, and the pressure difference signal Δq which is the result is obtained. The pressure difference signal Δq is supplied to the reed filter 153.
And the Graham function calculation circuit 154, respectively. In the reed filter 153, the pressure difference signal Δ
For q, the voltage rise ratio Q and the resonance frequency fc are calculated based on the characteristics of the lead portion of the wind instrument to be simulated.
Is given and the result is output as a pressure difference signal Δq ′. On the other hand, in the Graham function operation circuit 154,
The flow velocity signal v is output based on the pressure difference signal Δq.

Next, the pressure difference signal Δq 'is added to the adding circuit 156.
Is supplied to the pressure difference signal Δq ′, the voltage signal e indicating the embouchure is added, and the addition result is supplied to the slit function calculation circuit 157. by this,
The aperture function signal S _R is output from the slit function calculation circuit 157.

Next, in the analog multiplier 158,
The opening area signal S _R is multiplied by the flow velocity signal v, and the multiplication result is output as the volume flow velocity signal f. The volume flow velocity signal f is converted into a current signal in the voltage / current converter 160, then multiplied by the constant z in the analog multiplier 159, and the multiplication result is supplied to the linear unit 203 as the current signal zf. In the linear portion 203, a voltage traveling wave is generated from one end 203a toward the other end 203b based on the supplied current signal. When this voltage traveling wave is reflected at the other end 203b, it propagates from the other end 203b toward the one end 203a. The voltage appearing at the one end 203a is converted into a voltage signal q by the analog subtractor 1
Returned to 51. Further, the voltage signal q is sounded as a musical sound through the sound system 75.

According to the above configuration, since the entire circuit is realized as an analog circuit, the sampling frequency can be regarded as infinity and can be equated with the digital circuit, and the parasitic frequency is based on the finite sampling frequency of the digital circuit. It is possible to prevent oscillation. However, the above-described configuration has a drawback that a huge amount of capacitors and coils are required to realize the linear portion 203, and the configuration is still complicated. On the other hand, it is possible to use a BBD element or the like as a substitute for the ladder circuit, but there is a problem that the device becomes expensive. As described above, it was found that the intended purpose cannot be sufficiently achieved only by replacing the conventional digital circuit with the analog circuit. The above is the premise of the embodiment described later.

B. First Embodiment Next, a first embodiment of the present invention will be described with reference to FIG. In this embodiment, a non-linear portion 204 mainly composed of an analog circuit, a linear portion 104 composed of a digital circuit, and a junction 102 interposed between the two.
And a sound system 75. The configuration of the non-linear unit 204 is similar to that of the non-linear unit 201 described with reference to FIG. However, in the analog subtractor 151, the D / A converter 171 is connected to the input end of the pressure signal q.
Is provided, and A is provided after the analog multiplier 158.
A / D converter 170 and a digital multiplier 172 are provided. Therefore, the nonlinear unit 204 is the nonlinear unit 201.
It operates in the same manner as, except that the input / output signals are digital signals.

Next, the junction 102 and the linear portion 104 are the same as those described in FIG. The delay circuits 70 and 71 in the linear unit 103 are replaced with one delay circuit 74 in the linear unit 104, but both are characteristically the same. In this embodiment, the non-linear unit 204 is composed of an analog circuit,
The junction 102 and the linear section 104 are composed of digital circuits. Therefore, in the non-linear unit 204, parasitic oscillation based on the sampling frequency peculiar to the digital circuit can be prevented, while the linear unit 104 can be configured by using an inexpensive flip-flop or the like. That is, it can be seen that the musical sound synthesizer according to the present embodiment can be manufactured at low cost but has extremely high fidelity by skillfully combining the digital technology and the analog technology.

C. Second Embodiment Next, a second embodiment of the present invention will be described with reference to FIG. In this embodiment, the non-linear unit 205, the linear unit 104,
The junction 102 and the linear part 104 are provided between the junction 102 and the linear part 104.
The configuration is similar to that of the first embodiment.

Next, the non-linear portion 205 is constructed by using a mouthpiece 180 for a natural wind instrument, and a lead 181 is provided at one end thereof. 182 is a flow velocity sensor 182 that measures the flow velocity in the mouthpiece 180, and is provided near the center of the mouthpiece 180. As the flow velocity sensor 182, it is preferable to use a heat ray velocity meter or the like. The measurement result of the flow velocity sensor 182 is amplified by the amplifier 183 and supplied to the junction 102 via the A / D converter 170 and the digital multiplier 172.

Reference numeral 185 is a diaphragm composed of a speaker or the like, and a voice signal output through the D / A converter 171 and amplified by the amplifier 184 is sounded. Also, 186 is a padding such as cotton, which is provided at the end portion 180b of the mouthpiece 180 to suppress air vibration.
Is packed in. Next, the operation of this embodiment will be described.
First, when the performer blows from one end 180a of the mouthpiece 180, the flow velocity sensor 182 detects the change in the flow velocity caused by the breath. In addition, the flow velocity sensor 18
2 is the mouthpiece 18
Since the cross-sectional area of 0 is constant, the air flow velocity is uniquely determined with respect to the air flow rate, and the volume flow velocity signal f, which is a voltage signal indicating the air flow rate, is output from the amplifier 183.

Next, the volume flow velocity signal f is converted into a digital signal in the A / D converter 170 and is multiplied by the constant z through the digital multiplier 172, so that the digital multiplier 172 outputs the signal zf indicating the air pressure. Is output.
This signal zf is supplied to the linear unit 104 via the junction 102 as the traveling wave pressure signal q ₀ . In the linear unit 104, the traveling wave pressure signal q ₀ is delayed and filtered, and after a predetermined time has passed, it is output as a reflected wave pressure signal q _i . This reflected wave pressure signal q _i
Is supplied to the D / A converter 171 as the pressure signal q via the junction 102. The pressure signal q is
After being converted into an analog signal in the D / A converter 171, amplified by the amplifier 184, the diaphragm 185
Is pronounced through.

When the diaphragm 185 produces a sound, the compressional wave of the air generated by the production of the sound is generated by the end portion 1 of the mouthpiece 180.
Propagating from 80b toward one end 180a, leads 181
Reach By the way, since the player is still breathing into the mouthpiece 180 even at this point, a pressure corresponding to the sum of the blowing pressure p generated thereby and the pressure q generated by the diaphragm 185 is applied to the lead 181. This pressure causes the lead 181 to vibrate.

When the lead 181 vibrates, a compressional wave of air is generated, and this compressional wave propagates as a pressure traveling wave in the mouthpiece 180 from one end 180a toward the terminal end 180b. Here, the compression wave generated by the lead 181 and the diaphragm 185 locally causes a change in the flow velocity of air, and thus the change in the flow velocity is detected via the flow velocity sensor 182. Then, the detection result of the flow velocity sensor 182 is converted into a traveling wave pressure signal q ₀ via the amplifier 183, the A / D converter 170, and the junction 102,
It is supplied to the linear unit 104 again and is sounded via the sound system 75.

As described above, the nonlinear section 205 and the linear section 10
Signals are exchanged between 4 and junction 10
2 by the pressure signal q output from the mouthpiece 18
The physical state inside 0 changes. Therefore, the lead 181
Resonates in a cycle corresponding to the delay time in the linear section 104, and thereby the frequencies of the traveling wave pressure signal q ₀ and the reflected wave pressure signal q _i , that is, the pitch of the musical sound is determined.

As described above, according to the present embodiment, since the mouthpiece 180 of the actual wind instrument is used in the non-linear section 205, the circuit configuration of the non-linear section can be extremely simplified and can be manufactured at a low cost. It turns out that the fidelity is high. Further, in the present embodiment, the diaphragm 18
Of the sounds generated by 5, the sound that leaks out of the mouthpiece is weak, and headphones or the like can be freely adopted as the sound system 75. It is suitable.

As described above, according to the tone synthesis apparatus of the first aspect, the linear circuit is composed of the digital circuit and the non-linear circuit is composed of the analog circuit. It is possible to increase fidelity. Further, in the structure according to the second aspect, since the musical sound is synthesized by utilizing the physical behavior of the mouthpiece having the lead, the electric structure can be simplified and the musical sound with high fidelity can be obtained. Can occur.

[Brief description of drawings]

FIG. 1 is a block diagram of a first embodiment of the present invention.

FIG. 2 is a block diagram of a second embodiment of the present invention.

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

FIG. 4 is an operation explanatory diagram of a conventional musical sound synthesizer.

FIG. 5 is a block diagram of a main part of a conventional musical sound synthesizer.

FIG. 6 is a block diagram of a circuit on which the present invention is based.

[Explanation of symbols]

 104 linear part (digital linear circuit, linear circuit) 170 A / D converter (second conversion means) 171 D / A converter (first conversion means) 180 mouthpiece 182 flow rate sensor 185 diaphragm (vibration means) 204 non-linear Section (analog nonlinear circuit)

Claims (2)

[Claims] 1. A digital linear circuit that delays and outputs a supplied digital signal, a first conversion unit that converts an output signal of the digital linear circuit into an analog signal, and an output from the first conversion unit. An analog non-linear circuit for giving frequency characteristics to the analog signal and outputting the analog signal; and a second conversion means for converting an output signal of the analog non-linear circuit into a digital signal and supplying the digital signal to the digital linear circuit. A tone synthesizer characterized by. 2. A mouthpiece having a lead, a flow velocity sensor for detecting an air flow velocity in the mouthpiece, a linear circuit for amplifying and delaying an output signal of the flow velocity sensor, and outputting the amplified signal, the linear circuit being provided in the mouthpiece. And a vibrating means for vibrating based on the output signal of the linear circuit.