JP2720627B2 - Music synthesizer - Google Patents

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 for synthesizing natural musical instrument sounds.

[0002]

2. Description of the Related Art A tone synthesizer which synthesizes a tone by creating a model simulating a resonance part of a natural musical instrument such as a resonance tube of a wind instrument or a string of a stringed instrument, and inputting an excitation signal to the model. It has been known. Here, an FIR (finite impulse response) filter is often used as a model of the resonance unit. In creating a model of a resonance unit using an FIR filter, first, an impulse response of a resonance unit of a natural musical instrument to be subjected to musical sound synthesis is observed. The following is an example. In the case of a wind instrument having a cylindrical tube having a length L as shown in FIG. 11, when an impulse is input from a mouthpiece, a reflected wave pulse is observed in the mouthpiece after a lapse of time T as shown in FIG. . Here, the time T is given by
= 2L / v. In the case of a wind instrument having a conical tube as shown in FIG. 13, an impulse response as shown in FIG. 14 is obtained. In the case of a bowed musical instrument as shown in FIG. 15, an impulse response obtained by superimposing the impulse responses shown in FIGS. 16A and 16B can be obtained by inputting an impulse to a bowed point. Here, the times T ₁ and T ₂ until the reflected pulse appears are determined by the distance L ₁ from the bowed point to the bridge and the distance L ₂ from the bowed point to the nut, and T ₁ = 2L ₁ / v, T ₂ = 2L
₂ / v. Further, in the case of a piano as shown in FIG. 17, since the strings are thick, the elastic characteristics of the strings affect the propagation of vibration. Therefore, an impulse response illustrated in FIG. 18 is obtained. The impulse response thus obtained is sampled on the time axis at a predetermined sampling cycle interval, and each sample value is set as a filter calculation coefficient in the FIR filter. By doing so, a filter having exactly the same impulse response as the resonance part of the natural musical instrument to be created is created. Then, by using the filter thus created as a filter for tone color adjustment, it is possible to synthesize a tone having a tone color close to that of a natural musical instrument.

[0003]

By the way, the timbre of a natural musical instrument changes with the pitch and the passage of time. Further, the timbre changes according to various performance conditions, not limited to the pitch and the passage of time. Therefore, in the above-described conventional tone synthesizer, it is preferable to be able to change the filter calculation coefficient of the FIR filter for tone adjustment, since a performance close to an actual natural musical instrument can be performed. However, in the case where the filter calculation coefficient can be changed in this way, if the filter calculation coefficient is changed while a certain musical tone is being generated, there is a problem that annoying click noise is generated. SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a tone synthesizer capable of performing tone change without generating click noise even during sounding.

[0004]

Musical tone synthesizing apparatus according to the present invention SUMMARY OF THE INVENTION comprises a parameter generating means for generating a filter calculation coefficient corresponding to the tone color of a musical tone to be sounded, a plurality of FIR filters each input is commonly connected , And these FIs
A loop means including a mixing circuit for mixing and outputting each output of the R filter; and generating an excitation signal,
Exciting means to be input, and when a change in timbre is instructed, a filter calculation coefficient corresponding to the new timbre is output to the FIR filter whose output is not output from the mixing circuit at the time when the timbre change is instructed. At the same time as the setting, the output of the FIR filter is controlled to increase with time, while the output of the FIR filter whose output is output from the mixing circuit at the time when the change of the tone is instructed decreases with time. Control means for controlling so that only the output from the FIR filter in which the filter calculation coefficient corresponding to the new tone color is finally set is output from the mixing circuit. It is characterized by.

[0005]

According to the above arrangement, when the timbre is changed, the form of the filter operation is continuously changed on the time axis, and finally the filter operation corresponding to the new timbre is performed.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. [First Embodiment] FIG. 1 is a block diagram showing a configuration of a musical sound synthesizer according to a first embodiment of the present invention. This tone synthesizer simulates a wind instrument having a cylindrical tube as shown in FIG. In FIG. 1, 1
Is an excitation unit that simulates the mouthpiece of a wind instrument, 2
Is a coefficient interpolation type FIR filter simulating a cylindrical tube. Reference numeral 3 denotes a junction simulating the connection between the mouthpiece and the cylindrical tube, and is composed of adders 3a and 3b. The output of the excitation unit 1 is input to one input terminal of an adder 3a, and the output of the adder 3a is a coefficient interpolation type FIR
The signal is input to the filter 2 and to one input terminal of the adder 3b. The output of the coefficient interpolation filter 2 is input to the other input terminal of the adder 3b and the other input terminal of the adder 3a, and the output of the adder 3b is input to the excitation unit 1.

The input signal to the excitation unit 1 corresponds to the pressure due to the contribution of the backward wave of the air vibration wave that is fed back toward the lead in the mouthpiece of the wind instrument.
The pressure of the traveling wave is added to the actual pressure by the adder 3b, and is input to the subtractor 101 in the excitation unit 1. Then, the value P corresponding to the blowing pressure is subtracted from the output of the adder 3b by the subtracter 101, and a signal corresponding to the pressure in the mouthpiece is output from the subtractor 101. The output signal of the subtracter 101 is attenuated by a high-frequency component by passing through a filter 102 for phase correction, and a filter 103 (usually a low-pass filter) which simulates a response characteristic of a lead to a pressure change in a mouthpiece and a mouse The flow rate of the air flow in the piece is input to a nonlinear circuit 104 that simulates a saturation characteristic of the flow rate of the air flow in the mouthpiece. Here, the selectivity Q, cutoff frequency fc, and gain G of the filter 103 are adjusted by control means (not shown) corresponding to the tone color to be obtained. Also,
The nonlinear circuit 104 and a nonlinear circuit 106 described below
For example, an RO that stores a table of a target nonlinear function
It can be realized by M (read only memory). The output of the filter 103 is input to the adder 105, and the embouchure signal E related to the pressure at which the player holds the mouthpiece is added. Then, a signal corresponding to the pressure applied to the lead is output from the adder 105, and a nonlinear circuit 106 simulating the change in the cross-sectional area of the gap between the lead and the mouthpiece with respect to the change in the pressure of the lead.
Is input to Then, the multiplier 107 multiplies the output signal of the non-linear circuit 106 by the output signal of the non-linear circuit 104, and outputs a signal corresponding to the volume flow rate of the airflow passing through the gap between the mouthpiece and the lead. The multiplier 108 multiplies the output signal of the multiplier 107 by a value Z corresponding to the impedance to the airflow in the mouthpiece. Then, a signal corresponding to the pressure change generated in the mouthpiece is output from the multiplier 108. The output signal of the multiplier 108 is added to the adder 3a.
Is input to the coefficient interpolation type FIR filter through

[0008] coefficient interpolation type FIR filter 2, multipliers M0～M _n delay circuits D ₁ to D _n for sequentially delaying the input signal, the coefficient multiplication process of the outputs of the input signal and the delay circuits D ₁ to D _n perform these multipliers M ₀ ~M _n adder SUM adds the multiplication results, and the filter operation coefficient a ₀ ~a _n supplied from the control means (not shown) to interpolate the time axis, each multiplier M ₀ each consisting of supplying interpolators I ₀ ~I _n in ~M _n as a multiplication factor. Here, the interpolator I _i (i
= 0 to n) can be realized, for example, by a first-order low-pass filter using an IIR (infinite impulse response) filter shown in FIG. Further, by control means (not shown), the time constant τ _{i (i} = 0~n) is interpolator I _{i (i} = 0~
n), and a filter operation coefficient a in each interpolator
_The response characteristic to the change of _i is adjusted. When using the IIR filter shown in FIG. 2 as interpolator Ii, by providing a constant tau _i as a multiplication coefficient to the multiplier MH time,
The response characteristics can be adjusted.

The operation of the tone synthesizer will be described below. Each multiplier coefficient interpolation filter 2, the coefficient a ₀ ~a _n filter operation in accordance with the tone color which has been set is set by control means not shown. When an instruction to generate sound is given to a control unit (not shown) by operating an operation unit such as a keyboard, for example, signals P and E are supplied to the excitation unit 1 from the control unit. A signal corresponding to the pressure change in the piece is output from the excitation unit 1. The output signal of the excitation unit 1 is input to the coefficient interpolation type FIR filter 2 via the junction 3, and the output of the coefficient interpolation type filter 2 is output via the junction 3 to the excitation unit 1
Will be returned to Then, the excitation unit 1 causes the coefficient interpolation type F
The signal to be input to the IR filter 2 is output again. Thereafter, the same operation is repeated, and a signal circulating in the tone synthesizer is extracted and output as a tone signal.

When the operation of changing the tone or the operation of changing the timbre by operating the timbre manipulator or the like is performed during the period in which the tone generation operation is being performed, the control means causes the coefficient interpolation type FIR filter 2 to operate. The filter calculation coefficients a _i (i = 0 to n) supplied thereto are changed. However, these filter operation coefficients a _i ( _{i to} n) are calculated by the interpolator I _i (i = 0).
To n), the signals are interpolated and supplied to the respective multipliers M _i (i = 0 to n) as exemplified by the symbol a _i ′ in FIG. As described above, since each multiplication coefficient of each multiplier M _i (i = 0 to n) gradually changes to a coefficient value corresponding to a new timbre, the timbre can be changed without generating click noise during sounding. Done.

[Second Embodiment] FIG. 4 is a block diagram showing a configuration of a musical sound synthesizer according to a second embodiment of the present invention.
This tone synthesizer simulates a wind instrument having a cylindrical resonance tube, as in the first embodiment.
In this figure, 2a and 2b are F
An IR filter, has a configuration obtained by removing the interpolator I ₀ ~I _n in the coefficient interpolation type FIR filter 2 shown in FIG. The multiplier M ₀ ~M _n in each of these FIR filters 2a and 2b filter calculation coefficient to be supplied is set directly from the control means. The signal supplied from the excitation unit 1 via the junction 3 is input to both the FIR filters 2a and 2b. Each output of the FIR filters 2a and 2b is multiplied by multipliers r and 1-r by multipliers 4a and 4b, respectively, and each multiplication result is added by an adder 4c. And adder 4
The addition result at c is fed back to the excitation unit 1 via the junction 3.

The operation of this embodiment will be described below. When the timbre is not changed, the value of the multiplication coefficient r is set to 0 or 1 by the control means. Here, assuming that r = 0 is set, the output signal of the FIR filter 2b is fed back to the excitation unit 1. When the operation of changing the timbre is performed, the filter calculation coefficient corresponding to the new timbre is set in the FIR filter 2a by the control means. After this setting, the value r of the multiplication coefficient is gradually changed from 0 to 1 by a gradual temporal change. As a result, of the signals fed back to the excitation unit 1, the output signal of the FIR filter 2b gradually decreases, and the output signal of the FIR filter 2a gradually increases. Finally, when r = 1, only the output signal of the FIR filter 2a is fed back to the excitation unit 1, and the timbre change is completed. When a new timbre change operation is performed, a filter calculation coefficient corresponding to the new timbre is set in the FIR filter 2b,
The value of r is gradually changed from 1 to 0. FIG. 5 shows that the multipliers 4a and 4b in FIG.
a and 2b have been moved to the input sections. Also in this configuration, an operation completely equivalent to that in FIG. 4 is performed.

[Third Embodiment] FIG. 6 is a block diagram showing a configuration of a tone synthesizer according to a third embodiment of the present invention.
This tone synthesizer simulates a wind instrument having a conical tube as shown in FIG. FIG.
The input acoustic impedance Z of the conical tube as shown in FIG. 3 can be expressed by the following equation (1).

(Equation 1) Here, ρ is the medium density (g / cm ³ ), and c is the sound velocity (cm
/ Sec), X is the throat length (cm), and L is the length of the conical tube (cm). K is a wave constant (rad / cm)
When the wavelength of the sound is λ (cm), k = 2π /
λ. S is the initial area of the conical tube. Now,
In the first and second terms of the denominator in Equation 1,

(Equation 2)

(Equation 3) far. Where ω is the angular frequency of the sound (rad / sec)
And ω = c · k. Also, M = ρ · X / S. By this replacement operation, the following Expression 4 is obtained.

(Equation 4) That is, the input acoustic impedance Z of the conical tube can be regarded as the impedance Zx and ZL connected in parallel. Here, since the input impedance seen from the transmitting side when the receiving side is short-circuited in the lossless parallel line having the length L is Zi = j · tan (k · L), the impedance Z _L is reduced in thickness. It is possible to simulate with a constant cylindrical tube. In FIG. 13, when the throat length X is small, it is possible to approximate the impedance Zx expressed by the above equation 2 by the following approximate equation 5.

(Equation 5) In other words, by applying the equation based on the equation (5), a conical tube can be approximated by a configuration in which two cylindrical tubes are connected in parallel. The coefficient interpolation type filters 2c and 2d in FIG. 6 simulate two cylindrical tubes approximating the conical tube in this way. These coefficient interpolation filters 2c and 2d have the same configuration as that used in the first embodiment (FIG. 1), and filter control coefficients corresponding to the tone colors to be generated are set by the control means. Is done. 5 is a junction that mediates transfer of signals between each of the exciter 1, coefficient interpolation filter 2c and 2d, the multiplier 5M ₁ ~5M _3,
Consisting subtractor 5S ₁ ~5S ₃ and the adder 5A. After the output signal of the exciter 1 through the junction 3 and one sample period delay circuit 6, a subtracter one input to the negative input terminal of the 5S _1, multiplier 5M adder coefficient C ₁ is multiplied by ₁ 5A. Here, the one-sample period delay circuit 6 is connected to the adder 3 at the junction 3.
a and loop passing through the subtractor 5S ₁ at junction 5 is provided in order to avoid a delay-free loop containing no delay element. The output signal of the coefficient interpolation type FIR filter 2c is a subtractor 5S
₂ is input to the minus input terminal, and is multiplied by the coefficient C ₂ by the multiplier 5M ₂ and input to the adder 5A.
The output signal of the coefficient interpolation type FIR filter 2d, while inputted to the negative input of the subtractor 5S _3, the coefficient C ₃ are input to the adder 5A are multiplied by the multiplier 5M _3. Here, each coefficient C _{1 to} C ₃ is C ₁ + C ₂ + C ₃ ≦
Within the range satisfying 2, a value corresponding to the shape of the conical tube to be simulated is used. Addition result in the adder 5A is input to the positive input of each of the subtractor 5S ₁ ~5S _3. The output of the subtractor 5S ₁ is fed back to the exciter 1 through the junction 3, a subtractor 5S ₂ and 5
Each output of the S ₃ are respectively inputted to the coefficient interpolation filter 2c and 2d. Also in this configuration, as in the first embodiment, interpolation of filter operation coefficients is performed by interpolators in the coefficient interpolation type FIR filters 2c and 2d. Therefore, even when the filter operation coefficient is switched by tone change during tone generation, the tone is changed to a new tone without generating click noise.

[Fourth Embodiment] FIG. 7 is a block diagram showing a configuration of a musical tone synthesizer according to a fourth embodiment of the present invention.
This musical sound synthesizer has FIR filters 2ca and 2cb and a FIR filter 2da having the same configuration as the FIR filter 2a shown in FIG. 4, instead of the coefficient interpolation filters 2c and 2d in the third embodiment (FIG. 6). And 2db. Also, the FIR filter 2ca
Multiplier 5M ₂ a that multiplies the output of by a multiplication coefficient r
And the multiplication coefficient 1− with respect to the output of the FIR filter 2cb.
r multiplier 5M ₂ b and these multipliers 5M ₂ a
And mixing unit consisting of an adder 5A ₂ for supplying to the multipliers 5M ₂ by adding the multiplication results of 5M ₂ b is provided. Similarly, the outputs of FIR filters 2da and 2db are mixed by a mixing unit including multipliers 5M ₃ a (multiplication coefficient r), 5M ₃ b (multiplication coefficient 1-r) and adder 5A _3, and ₃ to be supplied. Further, the FIR filter 2ca, 2cb, the input of each of 2da and 2 db, subtractor 5S ₂ a for outputting subtracts the output from the addition result of the adder 5A of the FIR filter, 5S ₂ b, 5S ₃ a and Each output terminal of 5S ₃ b is connected.

In this tone synthesizing apparatus, tone color is changed by the same control as in the second embodiment. That is, when the timbre is changed while the sound is being generated in the state of r = 0, the filter calculation coefficient corresponding to the new timbre is set for each of the FIR filters 2ca and 2da, and then r Is gradually changed from 0 to 1. Further, when the tone color is changed while the sound is being generated in the state of r = 1, the FIR filters 2cb and 2db
After the filter calculation coefficients corresponding to the new timbres are set for each of, the value of r is gradually changed from 1 to 0. The multipliers 5M ₂ a and 5M ₂ b are connected to the subtractor 5
Provided before each of S ₂ a and 5S ₂ b, the multiplier 5M ₃
a and 5M ₃ b may be provided before each of the subtractors 5S ₃ a and 5S ₃ b. In this configuration, an operation completely equivalent to the above configuration is performed.

[Fifth Embodiment] FIG. 8 is a block diagram showing a configuration of a musical sound synthesizer according to a fifth embodiment of the present invention.
This tone synthesizer simulates a stringed instrument such as a piano. A loop circuit 100 including a coefficient interpolation type FIR filter 7, an adder 8, a coefficient interpolation type FIR filter 9 and an adder 10 in FIG. Corresponds to strings. More specifically, the coefficient interpolation type filter 7 simulates the impulse response of the portion from the string striking point of the piano to the bridge, and the coefficient interpolation type filter 9 simulates the impulse response from the string striking point of the piano. This simulates the impulse response up to the capo dust. These coefficient interpolation type FIR filters 7 and 9
Has exactly the same configuration as that used in the first embodiment, and a filter operation coefficient is set by control means (not shown). The outputs of the coefficient interpolation filters 7 and 9 are added by the adder 11 and output.

The excitation imparted to the string by the hammer is simulated by an exciter 200 described below.
First, the adder 201 has a hammer with a string ST at one input end.
A hammer speed signal VEL0 corresponding to the speed colliding with R is input. The other input terminal of the adder 201 receives the integrated value of the integrator 202. This integral value corresponds to a speed change generated in the hammer due to the interaction between the hammer and the strings. The calculation of the speed change will be described later. From the adder 201, a signal obtained by correcting the hammer speed signal VEL0 by the change in speed, that is, a signal corresponding to the current hammer speed is obtained. Then, the output signal of the adder 201 is input to the integrator 203 and integrated, and a hammer displacement signal HD corresponding to the displacement of the hammer is output. On the other hand, the adder 204 applies a coefficient SA to the output signal of the adder 11 by the multiplier 205.
The signal multiplied by DM and the output signal of multiplier 206 are input. Here, the output signal of the multiplier 205 corresponds to the speed of the string at the striking point, and the output signal of the multiplier 206 corresponds to the speed correction provided to the string by the hammer. Accordingly, the adder 204 outputs a signal SV corresponding to the current string speed. Then, the signal SV is integrated by the integrator 207, and a string displacement signal SD corresponding to the string displacement is obtained. Then, the string displacement signal SD is subtracted from the hammer displacement signal HD by the subtracter 208, and a relative displacement signal SHD corresponding to the amount of bite of the string with respect to the hammer is obtained.

The relative displacement signal SHD is output from the nonlinear circuit 209
And a signal F corresponding to the repulsion between the hammer and the strings is output from the nonlinear circuit 209. Nonlinear circuit 20
9 is multiplied by a multiplier 210 by a multiplication factor 1/2.
Is multiplied. As a result, the velocity components of the vibration waves propagating on both sides of the string striking point are output from the multiplier 210 and fed back to the adders 8 and 10 of the loop circuit 100. The output signal of the non-linear circuit 209 is
6 is multiplied by a predetermined multiplication coefficient FADM, and a signal corresponding to the speed change given to the string by the above-mentioned hammer is obtained from the multiplier 206. In addition, the nonlinear circuit 2
09 is output from the multiplier 211 by a multiplier coefficient −
Multiplied by 1 / M (where M is the mass of the hammer), a signal HA corresponding to the acceleration acting on the hammer is output.
Then, the signal HA is integrated by the integrator 202 to obtain a signal corresponding to the above-described change in the speed of the hammer. Also in this musical tone synthesizer, since the filter calculation coefficient given from the control means is used after being interpolated on the time axis, it is possible to change the timbre without generating click noise.

[Sixth Embodiment] FIG. 9 is a block diagram showing the configuration of a musical sound synthesizer for synthesizing musical sounds of a bowed instrument such as a violin according to a sixth embodiment of the present invention. In this figure, reference numerals 301a and 301b denote FIR filters simulating the impulse response of a portion from a stringed position of a violin string to a nut. These F
Each output of IR filters 301a and 301b is multiplied by multiplier coefficients r and 1-r by multipliers 302a and 302b, respectively, and each multiplication result is added by adder 303. Input to the input terminal. The output of the adder 304 is input to FIR filters 305a and 305b that simulate the impulse response of the portion from the string position to the bridge. These FIR filters 305a and 305a
05b are multiplied by multiplication coefficients r and 1-r by multipliers 306a and 306b, and the multiplication results are added by adder 307. The addition result is input to one input terminal of adder 308. Is done. This adder 3
08 is input to FIR filters 301a and 301b. The components described above constitute a loop circuit 300 corresponding to a string in a bowed musical instrument. Each output of the adders 303 and 307 is
Is added. As a result, the adder 309 outputs a signal corresponding to the speed of the string at the position of the bowed string. Then, the subtractor 401 subtracts a signal Vb corresponding to the speed of striking the string with the bow from the output signal of the adder 309, and outputs a signal corresponding to the relative speed between the bow and the string. The output signal of the subtractor 401 is input to the divider 402, and is divided by the signal Fb corresponding to the bow pressure. Then, the output signal of the divider 402 is input to a nonlinear circuit 403 having input / output characteristics illustrated in FIG. The output signal of the non-linear circuit 403 is multiplied by the signal Fb by the multiplier 404, and a signal corresponding to the speed change given to the string by the bow is output from the multiplier 404. 307 are fed back to the other input terminals. In this tone synthesizer, the tone color is changed by the same control as in the second embodiment.

As another application form of the present invention, when synthesizing a wind instrument sound, it is conceivable to realize a change in timbre during a transition period in which the open / close state of the tone hole changes by performing interpolation of filter coefficients. Can be In this case, when updating from the filter coefficient corresponding to the state of a certain tone hole to the filter coefficient corresponding to another state of the tone hole, a new state is established through the filter coefficient corresponding to the state where the tone hole is opened halfway. If the filter coefficient is changed to a filter coefficient corresponding to the tone, the reality of the timbre at the time of pitch switching is further increased, and the tone quality of the musical tone is improved. When such a tone hole is opened and closed halfway, the loss of the pressure wave in the pipe increases, and as a result, the amplitude of the musical tone becomes unstable, and the pitch becomes unstable.

[0021]

As described above, according to the tone synthesizer of the present invention, the tone can be changed without generating click noise even during sound generation.

[Brief description of the drawings]

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

FIG. 2 is a block diagram illustrating a configuration example of an interpolator in the embodiment.

FIG. 3 is a waveform chart showing an operation of the interpolator in the embodiment.

FIG. 4 is a block diagram showing a configuration of a musical sound synthesizer according to a second embodiment of the present invention.

FIG. 5 is a block diagram showing a configuration of a modification of the embodiment.

FIG. 6 is a block diagram showing a configuration of a musical sound synthesizer according to a third embodiment of the present invention.

FIG. 7 is a block diagram showing a configuration of a musical sound synthesizer according to a fourth embodiment of the present invention.

FIG. 8 is a block diagram showing a configuration of a musical sound synthesizer according to a fifth embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of a tone synthesizer according to a sixth embodiment of the present invention.

FIG. 10 is a diagram illustrating a nonlinear function used in the embodiment.

FIG. 11 is a view showing a wind instrument having a cylindrical tube.

FIG. 12 is a diagram showing an impulse response of the wind instrument.

FIG. 13 is a view showing a wind instrument having a conical tube.

FIG. 14 is a diagram showing an impulse response of the wind instrument.

FIG. 15 is a view showing a bowed musical instrument.

FIG. 16 is a diagram showing an impulse response of the bowed instrument.

FIG. 17 is a view showing a stringed musical instrument.

FIG. 18 is a diagram showing an impulse response of the stringed musical instrument.

[Explanation of symbols]

1 Exciter, 2 Coefficient interpolation type FIR filter, 4a
... Multiplier, 4b ... Multiplier, 4c ... Adder.

Claims (1)

(57) [Claims] 1. A and parameter generating means for generating a filter calculation coefficient corresponding to the tone color of a musical tone to be sounded, a plurality of FIR filters each input is connected in common, and the respective outputs of these FIR filters are mixed Means including a mixing circuit for outputting the excitation signal , and excitation means for generating an excitation signal and inputting the signal to the loop means
When a change in timbre is instructed, a filter operation coefficient corresponding to a new timbre is set in an FIR filter whose output is not output from the mixing circuit at the time when the change in timbre is instructed, The output of the FIR filter is controlled to increase as time elapses, while the output of the FIR filter output from the mixing circuit is controlled to decrease as time elapses when the change of the timbre is instructed. And control means for controlling only the output from the FIR filter in which the filter calculation coefficient corresponding to the new tone color is finally output from the mixing circuit. Musical sound synthesizer.