JP3413890B2 - Music synthesizer - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone synthesizer suitable for simulating a musical tone of a natural musical instrument, particularly a wind instrument. [0002] Conventionally, there has been known a musical sound synthesizer for synthesizing musical tones by operating a model simulating a sounding mechanism of a natural musical instrument. Among the tone synthesizers of this type, a tone synthesizer that synthesizes a wind instrument is configured by connecting an excitation circuit that simulates the operation of a mouthpiece unit and a resonance circuit that simulates the resonance characteristics of a resonance tube. For example, wind instruments such as saxophones, trumpets and clarinets have a conical resonance tube.
This type of resonance tube can be regarded as a series connection of cylindrical tubes of different thicknesses, so that a waveguide (bidirectional transmission circuit) including a delay circuit and a junction (coupling circuit) are cascaded in multiple stages. Was simulated by the circuit. A tone synthesis circuit using such a waveguide and a junction is disclosed in, for example, JP-A-63-40199. Incidentally, in the waveguide, a multiplier and a low-pass filter are introduced together with a delay circuit in order to simulate the reflection including loss at the end of the tube, and these are multiplied and cut as reflection coefficients corresponding to the timbre. The off-frequency is supplied. In such a configuration, when a performance input (for example, breath pressure, etc.) for producing a musical tone is supplied, the performance rises rapidly (a so-called sharp attack), and when the supply of the performance input is stopped, If a musical tone that slowly falls (see FIG. 10 (a)) is desired, the absolute value of the reflection coefficient γ is set to approximately “1”. Conversely, when a performance input is supplied, the tone slowly rises. At the same time, when the supply of the performance input is stopped, if a musical tone that rapidly falls (see FIG. 10B) is desired,
The absolute value of the reflection coefficient γ was set small. [0006] However, with the increasing demand for generating musical tones that cannot be obtained with existing natural musical instruments, the conventional configuration in which the reflection coefficient is always constant during supply of a performance input has been increased. However, there is a problem that it is not possible to create a musical tone whose rise and fall are both rapid (or slow). The present invention has been made in view of the above-described problems, and an object of the present invention is to be able to faithfully reproduce a musical tone of a wind instrument having a conical tube, and to generate a novel musical tone that has never been seen before. It is an object of the present invention to provide a musical sound synthesizer capable of performing such operations. [0007] In order to solve the above-mentioned problem, an excitation signal generated in response to a sound generation start instruction and a sound generation end instruction is circulated through a loop including a multiplier and a delay means. A tone synthesizer for synthesizing a musical tone, a delay time control means for controlling a delay time of the delay means in accordance with a pitch of a musical tone to be generated, a pitch of a musical tone to be generated and a delay time of the delay means . accordingly met multiplier coefficient generating means for generating a multiplication coefficient of the multiplier
At least when instructing the end of the pronunciation, the multiplication coefficient is changed.
And further comprising: According to the above arrangement, the delay time of the delay means is determined according to the pitch of a musical tone to be generated. Further, the multiplication coefficient of the multiplier is determined in consideration of the delay time of the delay means and the pitch of a musical tone to be generated. This makes it possible to make the attenuation of the musical sound due to the circulation of the loop constant with respect to the pitch. If the multiplier coefficient of the multiplier is different between the start and end of the tone generation, it is possible to synthesize a musical tone as never before. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS <Basic Configuration> First, a basic configuration of the present invention will be described with reference to FIG. As shown in this figure, the mouthpiece section of the wind instrument is simulated, and a wave is supplied to an excitation section 10 (not shown in this figure) for supplying a performance input such as a breath pressure signal corresponding to the start / end of a musical sound. A guide W1 is connected, and parameters are supplied to each of these parts. The waveguide W1 simulates the input acoustic impedance of a cylindrical tube such as a clarinet.
9 and is circulated. The waveguide W1 includes a delay element 31 for simulating a delay when a sound wave reciprocates in the cylindrical tube, a low-pass filter 33 having a filter coefficient (cutoff frequency) α for simulating an acoustic loss in the cylindrical tube, and multiplication. It is constituted by a loop circuit comprising a multiplier 35 having a coefficient γ and simulating the reflection at the terminal end in the cylindrical tube. Here, the delay element 31 is configured by a shift register having the number of stages L driven by, for example, the sampling frequency Fs. The number L of stages of the delay element changes according to the pitch of a musical tone to be generated. This is due to the fact that the wind instrument changes the pitch of the generated musical tone depending on the length of the tube (resonance frequency). Next, the multiplication coefficient γ will be described. First, the amplitude of the breath pressure signal supplied from the excitation unit 10 is “−2”.
0dt (1/10) "is considered.
During time dt, the signal causes the previously described loop circuit to go through (F
s · dt) Advance by samples. That is, the loop circuit is operated during the time dt. It will go around only once. Here, the signal amplitude becomes γ times per round of the loop circuit. Therefore, the loop is traversed the number of times shown in (Equation 1), and this
20 dB ”, It is. Since the end of the cylindrical tube discussed here is an opening, if the multiplication coefficient γ is negative, from equation (2), It becomes. When the pipe to be simulated has a conical pipe shape such as a saxophone, it can be regarded that two pipes are connected. Therefore, a waveguide simulating a conical tube is shown in FIG.
As shown in (1), a waveguide W2 obtained by simulating a conical tube is connected in parallel to the waveguide W1. In this case, the total number of delay stages is the sum of the number of delay stages of the two waveguides W1 and W2, and the signal becomes γ _S · γ _L times for each round of the loop circuit. Therefore, the following equation is established in the same manner as equation (3). (Equation 4) In this equation (4), especially when γ _S = γ _L , Therefore, the multiplication coefficients γ _S and γ _L are given by the following equations, respectively. (Equation 6) Here, the time dt is fixed, the number of delay stages is determined in accordance with the pitch of a musical tone to be generated, and the number of delay stages is substituted into the equation (3) (in the case shown in FIG. 1) to obtain a multiplication coefficient indicating a reflection coefficient. If γ is determined, the attenuation rate of the tone signal after the supply of the performance input is stopped can be made constant at each pitch. Next, each embodiment in which this basic configuration is applied to a musical sound synthesizer for an electronic musical instrument will be described. <First Embodiment> FIG. 4 is a block diagram showing a schematic configuration of this electronic musical instrument. In this figure, reference numeral 41 denotes a keyboard composed of a plurality of keys, a key-on KON indicating a key press / release, and a pitch PIT indicating a pitch of a pressed key.
CH and touch information indicating the strength of key depression are supplied to the control unit 42. On the other hand, reference numeral 43 denotes a tone color setting operator which selects a tone color of a musical tone to be generated, sets parameters defining various envelope waveforms described later, and supplies these information to the control unit 42. The control unit 42 generates various parameters based on the supplied information and supplies the generated parameters to the musical sound generation unit 44. Next, details of the tone generator 44 will be described with reference to FIG.
It will be described with reference to FIG. As shown in FIG.
4 is a waveguide W1 (see FIG. 1), an excitation unit 10,
And various components supplied from the control unit 42 to these components in accordance with the lapse of time after key-on / off. Note that the tone generator 44
Is supplied to another external processing device (not shown) (for example, the sound is externally generated by a sound system including an amplifier and a speaker). The output signal from the waveguide W1 is supplied to one input terminal of an adder 51 and is also supplied to one input terminal of an adder 52. The output of the adder 51 is supplied to the addition input terminal of the adder 53 as a signal corresponding to the pressure of the air vibration wave fed back toward the reed in the mouthpiece of the wind instrument. And the adder 53
As a result, the breath pressure signal P corresponding to the blowing pressure is subtracted from the output of the adder 51, and a signal corresponding to the pressure in the mouthpiece is output. Next, the output signal of the adder 53 is a reed filter 54 simulating the response characteristics of the reed to the pressure change in the mouthpiece, and the saturation characteristic of the air flow rate in the mouthpiece with respect to the air pressure in the mouthpiece. Is input to the nonlinear circuit 55 that simulates Note that the coefficients RECEF (cutoff frequency, selectivity, etc.) of the lead filter 54 are controlled according to the set timbre. Thereafter, a signal EMB corresponding to the pressure applied by the player to the mouthpiece is added to the output signal of the lead filter 54 by the adder 56. And
A signal corresponding to the pressure applied to the lead is output from the adder 56 and input to a nonlinear circuit 57 that simulates a change in the cross-sectional area of the gap between the lead and the mouthpiece with respect to a change in the pressure of the lead. The output signals of the non-linear circuit 55 and the non-linear circuit 57 are multiplied by the multiplier 58 to become a signal corresponding to a volume change of the airflow passing through the gap between the lead and the mouthpiece. Is supplied to the input terminal of Then, this signal and the output signal of the waveguide W1 are added by the adder 52 and supplied to the other input terminal of the adder 51, and input to the waveguide W1 to circulate through the waveguide W1. It has become. The loop gain envelope generator 60 (hereinafter, abbreviated as “loop gain EG”) calculates a reflection coefficient γ based on a set parameter.
It is supplied sequentially according to the key-on KON.
Here, an example of the configuration of the loop gain EG is shown in FIG.
In this figure, the EG state control unit 61 counts the sampling clock φ in response to the key-on KON, and changes its output signal every time the count result reaches a parameter that defines the time direction of the envelope waveform. The data selector 62 receiving this output signal at the control input terminal SEL sequentially switches the data supplied to each input terminal to obtain a reflection coefficient γ for simulating the reflection at the opening end, that is, of the multiplier 35 in FIG. Output as multiplication coefficient. If necessary, the output interpolator 63 (usually a low-pass filter) is used as the reflection coefficient γ.
May be used. Referring back to FIG. 5, the delay length control unit 64
Controls the number of delay stages L of the delay element 31 based on the pitch PITCH indicating the pitch and the filter coefficient (cutoff frequency) CUTOFF of the LPF 33. Number of delay stages L
Is determined together with the signal PITCH and the filter coefficient C
The reason why UTOFF is reflected is that when the cutoff frequency of the LPF 33 changes, the delay amount of the LPF 33 changes, and the delay amount of the entire loop by the waveguide W1 also changes. LPF33 filter coefficient CUT
By considering OFF, the delay amount of the entire loop by the waveguide W1 can correspond to the pitch PITCH. In the delay length control unit 64, for example, a two-dimensional table is created in advance, and the number L of delay stages corresponding to the input pitch PITCH and filter coefficient CUTOFF is read. The release coefficient supply unit 65 includes a delay stage number L controlled by the delay length control unit 64, a damping release rate DRRi (i = 1, 2, 3) for defining a period for supplying a reflection coefficient immediately after key-off. From the coefficient γ _i
Is calculated. That is, the release coefficient supply unit 65
The coefficients γ ₁ to γ ₃ corresponding to the damping release rate DDRi are calculated by the following equation in which dt in the equation (3) is replaced with DRRi. (Equation 7) Next, the breath pressure signal generator 66
Is generated according to the key-on KON. For this reason, various parameters for defining the breath pressure signal P are supplied to the breath pressure signal generation unit 66. Similarly, the filter EG6
7 sequentially supplies the filter coefficient CUTOFF of the LPF 33 in accordance with the key-on KON. FIGS. 7A to 7C show the correspondence between the parameters supplied to the breath pressure signal generator 66, the loop gain EG60, and the filter EG67, and the envelope waveform defined by these parameters. Each is shown. Each of these envelopes produces a musical tone
Key-on KON indicating the start (key depression / key release) ((d) in FIG.
Reference). That is, in these envelopes, the attack is caused by key-on, and the release is caused by key-off. FIG. 7A shows an envelope waveform of the breath pressure signal P by the breath pressure signal generator 66. In the rising portion (attack) of the envelope, the period is defined by the attack rates AR1 and AR2 and the decay rate DR, and the amplitude level is defined by the attack levels AL1 and AL2 and the sustain level SL. In the falling part (release) of the envelope, the period is defined by the release rate RR. A constant value is defined between the attack part and the release part by the sustain level SL. The parameters defining the envelope can be arbitrarily set by the tone color setting operator 43 in FIG. FIG. 7B shows a loop gain EG6.
The state of supply of the reflection coefficient γ by 0 indicates a stepwise change corresponding to the envelope of the breath pressure signal P. That is, immediately after key-on, γ _A1 , γ _A2 and γ _D are sequentially supplied immediately after the key-on in accordance with the periods defined by the attack rates AR1 and AR2 and the decay rate DR. Γ ₁ , γ _2, and γ ₃ are sequentially supplied corresponding to the periods defined by the release rates DRR1, DRR2, DRR3. The reflection coefficients γ _{1 to} γ ₃ are obtained based on the aforementioned equation (7). FIG. 7C shows the supply state of the filter coefficient CUTOFF by the filter EG67.
Changes stepwise according to the envelope. That is, in the attack of the envelope, the period is defined by the attack rates AR1 and AR2 and the decay rate DR, and the amplitude level is defined by the filter attack levels FAL1 and FAL2 and the filter sustain level FSL.
Also, in the release of this envelope, the period is defined by the filter release rate FRR,
The amplitude level is defined by the filter release level FSL and is kept constant thereafter. The filter sustain level FSL between the attack part and the release part is defined as a constant value. In the embodiment having such a configuration, the kind of timbre to be generated prior to the key pressing operation and the various parameters defining the envelope waveform described above are selected by the timbre setting operator 43, and the control unit 42 Supplied to In this state, when a key is pressed on the keyboard 41, the key-on KON goes high to indicate that, and the pitch PIT indicating the pitch of the key.
The CH and touch information indicating the strength of the key press are transmitted to the control unit 4.
2 is supplied. The control unit 42 considers (multiplies) the key information KON and the signal PITCH indicating the pitch with respect to the parameter defining the amplitude direction of the envelope waveform of the breath pressure signal P in consideration of the touch information. The other parts are supplied to the tone generator 42 as they are while being supplied to the tone generator. Thus, in the tone generator 44, the breath pressure signal P whose envelope corresponds to the period defined by the attack rates AR1 and AR2 and the decay rate DR immediately after key-on and whose envelope is shown in FIG. Generated
The breath pressure signal becomes a signal corresponding to a change in volume of the airflow passing through the gap between the lead and the mouthpiece by the excitation unit 10, and circulates through the waveguide W1,
It is returned to the excitation unit 10. At this time, the wave guide W1
, The multiplication coefficient of the multiplier 35, that is, the reflection coefficient γ indicating the reflection simulated by the waveguide W1, is changed by the loop gain EG60 as shown in FIG. 7B to the attack rates AR1, AR2 and the decay rate DR. , The filter coefficient CUTOFF of the LPF 33 changes stepwise.
Is changed by the filter EG67 as shown in FIG. By such an operation, the tone generation unit 4
The envelope of the output tone of No. 4 rises in the attack as shown in FIG. 7 (f). At this time, if the reflection coefficient is intentionally set to a value larger than "1", the attack portion of the output musical tone can be more rapidly activated. When the pressed key is released (key-off), first, the key-on KON goes low. In the tone generator 44, the envelope of the breath pressure signal P is changed by the breath pressure signal generator 66 to the sustain level S.
L changes from the steady state defined by L to “0” corresponding to the period defined by the release rate RR (FIG. 7).
(A)). Similarly, L is determined by the filter EG67.
The filter coefficient CUTOFF of the PF 33 changes from the steady state defined by the sustain level FSL to the release level F corresponding to the period defined by the release rate FRR.
It changes to the value specified by RL (see FIG. 7C). On the other hand, the reflection coefficient immediately after key-off includes γ ₁ , γ ₂ and γ _{3 due} to the loop gain EG60, and damping release rates DRR1, DRR2 and DRR.
The power is supplied in a stepwise manner corresponding to the period specified in 3. By such an operation, the envelope of the output musical tone of the musical tone generating section 44 falls at its release section as shown in FIG. 7 (f). As described above, the number of delay stages L of the delay element is determined so that the amount of delay of the waveguide W1 corresponds to the pitch. Further, based on the number of delay stages L, the reflection coefficients γ _{1 to} γ ₃ are calculated by the formula ( ₁ ). 7), the attenuation of the envelope of the output musical sound at the release portion can be constant regardless of the pitch. <Second Embodiment> Next, a second embodiment of the present invention will be described with reference to FIG. The difference from the first embodiment is that the waveguides W1 and W2 in the tone generator 44 are connected in parallel via adders 71 and 72. The signals input to these waveguides are output signals of the drive waveform generator 88 multiplied by a predetermined envelope ENV.
2 are supplied to each of the other input terminals. The drive waveform generator 88 generates a desired signal by a signal WAVE indicating a tone color and a key-on KON by a known method (for example, a waveform memory method). The output signal of the tone generator 44 is obtained by adding the output signals of the waveguides W1 and W2 to each other by the adder 73. In the tone generator of this embodiment, the waveguide is divided into two parts, so that the attenuation of the output tone in the release section is constant regardless of the pitch, as in the first embodiment. In order to make the pitch PITCH
At the same time, the number of delay stages of both delay elements 31 and 32 is determined in consideration of the respective cutoff frequencies in waveguides W1 and W2, and a multiplication coefficient indicating a reflection coefficient must be determined from the number of delay stages. No. For this reason, the delay length control unit 84 determines the filter coefficients COEF2, L of the pitches PITCH and LPF33.
The number of delay stages DLY2 of the delay element 31 and the number of delay stages D of the delay element 32 are obtained from the filter coefficient COEF1 of the PF 34 and the delay ratio DLYRATIO indicating the ratio of the delay amount.
LY1 is determined. Here, the filter coefficient COEF
1 and COEF2 are supplied in the same manner as the filter EG67 in FIG.
The drive waveform EG86 that outputs NV is similar to the breath pressure signal generator in FIG. This delay ratio DLYR
ATIO is expressed by the following equation. (Equation 8) That is, the delay ratio DLYRATIO indicates a ratio of the number of delay stages DLY1 of the delay element 32 to the total number of delay stages of the delay elements 31 and 32. Note that the delay ratio DLYRATIO can be set to a desired value by the tone color setting operator 43. The total number of delay stages of the delay elements 31 and 32 is calculated by the adder 87 and supplied to the release coefficient supply unit 85. The release coefficient supply unit 85 determines the total number of delay stages and a damping release rate DRRi (i = 1,
The coefficient γ _i is calculated from the following equation. That is, the release coefficient supply unit 85 sets dt in the equation (6) to DRR
A coefficient γ _i corresponding to the damping release rate DDRi is calculated by the following equation replaced with i. (Equation 9) In the second embodiment, γ _L = γ _S is set based on equation (6). However, each of γ _L and γ _S is determined so as to satisfy equation (4). A configuration for supplying a reflection coefficient may be employed. According to the second embodiment having such a configuration, the waveforms of each part immediately after key-on and key-off are as shown in FIGS. 9 (a) and 9 (b). That is, in accordance with the key-on KON (see FIG. 9C), FIG.
As shown in FIG.
The NV is output in the same manner as the envelope waveform of the breath pressure signal P in FIG. 7A, and as shown in FIG. 9B, the reflection coefficient γ _L (= γ _S ) by the loop gain EG80.
Is output in the same manner as the reflection coefficient γ in FIG. In this embodiment, as described above,
In order that the delay amount of the waveguide W1 corresponds to the pitch,
The total number of delay stages of the delay elements 31 and 32 is determined.
Based on the total number of delay stages, the reflection coefficients γ _{1 to} γ ₃ are calculated by the equation (9).
Therefore, the attenuation of the envelope of the output musical tone at the release portion can be constant regardless of the pitch. In each of the embodiments described above, the reflection coefficient is supplied in three stages in attack and release. However, it is needless to say that the number of stages can be set to an arbitrary number by adding a parameter. According to the present invention described above, the delay time of the delay means is determined according to the pitch of a musical tone to be generated. Further, the multiplication coefficient of the multiplier is determined in consideration of the delay time of the delay means and the pitch of a musical tone to be generated.
This makes it possible to make the attenuation of the musical sound due to the circulation of the loop constant with respect to the pitch. Further, if the multiplication coefficient of the multiplier is set to a different value at the start and end of sound generation, and furthermore, the tone can be synthesized as never before.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram for explaining a basic configuration of the present invention. FIG. 2 is a block diagram for explaining a basic configuration of the present invention. FIG. 3 is a diagram for explaining a multiplication coefficient in the present invention. FIG. 4 is a block diagram showing a schematic configuration of a first embodiment of the present invention. FIG. 5 is a block diagram showing a configuration of a tone generator 44 in FIG. 4; 6 is a block diagram showing an example of a configuration of a loop gain EG60 in FIG. FIGS. 7A to 7E respectively show the state immediately after key-on,
It is a figure showing a waveform of each part immediately after key-off. FIG. 8 is a block diagram showing a configuration of a tone generator according to a second embodiment of the present invention. FIGS. 9A to 9C respectively show the state immediately after key-on,
It is a figure showing a waveform of each part immediately after key-off. FIGS. 10A and 10B are diagrams illustrating waveforms of output musical tones according to the conventional configuration. [Explanation of Signs] 31 ... Delay element (delay means), 35 ... Multiplier, 64
…… Delay length control part (delay time control means), 65 ……
Release coefficient supply unit (multiplication coefficient generation means)

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10H 1/00-7/00

Claims (1)

(57) an excitation signal generated in response to the Claims 1] start instruction and end instruction of the sound, and is circulated in a loop comprising a multiplier and delay means, musical tone synthesizing tones In the synthesizer, a delay time control means for controlling a delay time of the delay means according to a pitch of a musical tone to be generated; and a delay time of the multiplier according to a pitch of a musical tone to be generated and a delay time of the delay means . A multiplication coefficient generating means for generating a multiplication coefficient , at least at the time of instructing the end of sound generation ;
And a device for changing a multiplication coefficient .