JPH04355795A - Distortion circuit - Google Patents

Abstract

PURPOSE:To outputs a musical sound which is rich in timbre and to output a sound which is close to a live sound when the level of an input musical sound signal is low by varying the mixing ratio of a signal generated from an input musical sound signal through distortion processing and the input musical sound signal according to the level of the input musical sound signal. CONSTITUTION:A level detecting means 20 detects the level of the input musical sound signal, a distortion generating means 21 adds distortion to the input musical sound signal, and a filtering means 22 performs a filtering process. A mixing means 23 mixes the output of the filtering means 22 and the input musical sound signal while setting the mixing ratio according to the level of the input musical sound signal detected by the level detecting means 20 and outputs the mixed signal as an output musical sound signal.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distortion circuit, and more particularly, to a distortion circuit which distorts an input musical sound waveform to include many overtone components to obtain an output musical tone signal with rich tone quality.

0002

[Prior Art] In devices that handle sound, such as electronic stringed instruments (for example, guitars, etc.), electronic musical instruments such as electronic organs and synthesizers, and musical sound output devices that process and output sounds generated by other sound sources, An important issue is how to produce musical tones with rich tones. As a means to produce such rich sounds, distortion processing has been conventionally performed to include many overtone components in the electrical signal of the sound generated by a sound source or the like.

Such a conventional distortion circuit includes, for example, a distortion circuit 1, a filter circuit 2, a distortion generation circuit 3, and a low-pass filter circuit 4, as shown in FIG. This conventional distortion circuit 1 removes noise contained in an input musical tone signal by a filter circuit 2 provided before a distortion generating circuit 3, and clips the waveform of the input musical tone signal at a predetermined signal level by the distortion generating circuit 3. and distort it. The musical tone signal distorted by this distortion generation circuit 3 is output to a low-pass filter circuit 4 provided after the distortion generation circuit 3, and the low-pass filter circuit 4
removes to some extent unnecessary harmonic components of the distorted musical sound waveform input from the distortion circuit 3, adds roundness to the sound, and outputs it as an output musical sound signal. Therefore, by removing unnecessary harmonic components from the output musical tone signal using the low-pass filter circuit 4, it is possible to output the musical tone signal as a soft and warm sound from the musical tone generator or the like.

0004

[Problem to be Solved by the Invention] However, in such conventional distortion circuits, a low-pass filter circuit is provided after the distortion generation circuit, so it is difficult to reduce the volume of musical sounds such as electronic stringed instruments and make them close to live sounds. If you want to play the input musical sound signal in the same state, the distortion generation circuit outputs the input musical sound signal without adding distortion because the signal level is low, but filtering processing is performed by the low-pass filter circuit installed after the distortion generation circuit. Therefore, a sound that is different from the live sound, for example, a sound that is more muffled than the live sound, is played. As a result, there was a problem in that it was not possible to perform in a state close to the real sound.

[0005] Therefore, the inventions of the present application enable a musical performance that is rich in musical accuracy using a musical sound signal that applies distortion to an input musical sound signal and undergoes filtering processing, and when the volume of the input musical sound signal is reduced, it is possible to perform a musical sound that is similar to a live sound when the volume of the input musical sound signal is reduced. The purpose is to make it possible to play similar sounds.

[0006]

[Means for solving the problem] The invention according to claim 1 includes:
A distortion circuit includes a level detecting means for detecting the signal level of an input musical tone signal, a distortion generating means for distorting the waveform of the input musical tone signal, and a filtering device for performing a predetermined filtering process on the musical tone signal output from the distortion generating means. means, mixing means for mixing and outputting the input musical tone signal and the musical tone signal output from the filtering means at an arbitrary mixing ratio, and setting the mixing ratio by the mixing means based on the detection result of the level detecting means. The invention according to claim 2 is characterized in that the distortion circuit comprises a level detection means for detecting the signal level of the input musical tone signal, and a mixing ratio setting means for distorting the waveform of the input musical tone signal. filtering means that performs a predetermined filtering process on the musical tone signal outputted by the distortion generating means; and a filtering means that performs predetermined filtering processing on the musical tone signal outputted by the distortion generating means; The present invention is characterized by comprising a selection means for selecting and outputting one of the following.

[0007]

In the first aspect of the invention, the input musical tone signal is input to the level detecting means, and the level detecting means detects the signal level of the input musical tone signal. Further, the input musical tone signal is input to the distortion generating means, and the distortion generating means distorts the waveform of the input musical tone signal and outputs the distorted waveform to the filtering means. The filtering means performs a predetermined filtering process on the musical tone signal output from the distortion generating means and outputs it to the mixing means. The mixing means mixes the input musical tone signal and the musical tone signal output from the filtering means at an arbitrary mixing ratio and outputs the mixture as an output musical tone signal. The mixing ratio by this mixing means is set by a mixing ratio setting means based on the detection result of the level detecting means. Therefore, it is possible to include distortion in the input musical tone signal and to perform filtering processing, thereby making it possible to perform musical tones with rich musicality.
In addition, when the level of the input musical sound signal is narrowed down, by increasing the mixing ratio of the input musical sound signal based on the level of the input musical sound signal, it is possible to perform a sound close to a real sound. In the invention as set forth in claim 2, the input musical tone signal is input to the level detecting means and the selecting means, and the level detecting means detects the signal level of the input musical tone signal. Further, the input musical tone signal is input to the distortion generating means, and the distortion generating means distorts the waveform of the input musical tone signal and outputs the distorted waveform to the filtering means. The filtering means performs a predetermined filtering process on the musical tone signal output from the distortion generating means and outputs it to the selection means. The selection means selects either the input musical tone signal or the musical tone signal output from the filtering means, based on the detection result of the level detection means, and outputs the selected one as an output musical tone signal. Therefore, when the input musical tone signal exceeds a predetermined level, the waveform of the input musical tone signal is distorted and filtering processing is performed to enable musical tones with rich musicality to be played, and the level of the input musical tone signal is reduced. When the input musical tone signal reaches a predetermined level or lower, the input musical tone signal is output as is as an output musical tone signal, making it possible to perform live tones.

[0008]

[Example] The following is a detailed explanation based on an example. 1 to 15 are diagrams showing an embodiment of a distortion circuit according to the invention according to claim 1.

FIG. 1 is a circuit configuration diagram of a distortion circuit 10 according to a first aspect of the invention. The distortion circuit 10 includes an A/D converter 11, a DSP (Digital Signal
rocessor) 12, D/A converter 13, CPU (
Central Processing Unit)1
4. ROM (Read Only Memory) 15 and RAM (Random Access Memory)
) 16 mag.

[0010] The A/D converter 11 receives an analog input musical sound signal obtained by sampling a musical sound waveform output from an electronic stringed instrument such as a guitar at a predetermined sampling interval. The input musical tone signal is converted into digital signal and the digital input musical tone signal is sent to DSP1.
Output to 2. Therefore, an input musical tone signal sampled at predetermined sampling intervals and digitally converted is input to the DSP 12. The DSP 12 subjects the input musical tone signal inputted from the A/D converter 11 to the distortion processing of the present invention and outputs it to the D/A converter 13 .

The D/A converter 13 converts the musical tone signal from the DSP 12 into an analog signal and outputs it as an output musical tone signal to a musical tone generating device (not shown) or the like. The ROM 15 stores programs, especially programs for the distortion circuit of the present invention, and other necessary data and coefficients.
RAM 16 is used as a work area.

The CPU 14 transfers the program in the ROM 15 to the DSP 12, causes the DSP 12 to perform distortion processing, controls each section of the distortion circuit 10, and executes processing as the distortion circuit 10. The DSP 12 performs level detection processing (envelope extraction processing), distortion generation processing (clip processing), filtering processing, and mixing processing, and the configuration of a pseudo processing block of the DSP 12 can be shown as shown in FIG. That is, the DSP 12, as its pseudo processing block,
It includes a level detection processing section (envelope extraction processing section) 20, a distortion generation processing section (clip processing section) 21, a filtering processing section 22, and a mixing processing section 23.

Level detection processing section (level detection means) 20
, the input musical tone signal WIN is input, and the level detection processing section 20 includes gates 31 and 32, a multiplier 33, and an adder 34.
, a delay element 35, and a code detector 36. The input musical tone signal WIN is input to the gate 31, and the gate 31 operates under the control of the code detector 36 to open and close the gate. The multiplier 33 multiplies the output of the gate 31 or 32 by a predetermined attenuation coefficient PDEC and outputs the result as an envelope output signal WENV to the mixing processing unit 23 and to the delay element 35 . The delay element 35 delays the output of the multiplier 33 by one input musical tone signal WIN and outputs it to the adder 34, and the adder 34 receives the signal data from this delay element 35 and the input musical tone signal WIN.
is entered. The adder 34 adds the data from the delay element 35 and the input musical tone signal WIN and outputs the result to the sign detector 36. However, since the input musical tone signal WIN is now a negative input, the adder 34 adds the input musical tone signal WIN to the input musical tone signal WIN. musical tone signal WI
N and the data from the delay element 35 are subtracted and output to the sign detector 36. The sign detector 36 is the adder 3
It is detected whether the calculation result of step 4 is positive or negative, and the opening/closing of the gates 31 and 32 is controlled based on the detection result. That is, when the calculation result of the adder 34 is positive (that is, when the input from the delay element 35 is larger than the input musical tone signal WIN), the sign detector 36
When the gate 32 is opened and the calculation result of the adder 34 is negative (
That is, when the input musical tone signal WIN is larger than the input from the delay element 35), the gate 31 is opened.

The distortion generation processing section (distortion generation means) 21 overflows the maximum bit of the register of the DSP 12 by performing addition processing and shift processing on the input musical tone signal WIN, and converts the input musical tone signal WIN into the input musical tone signal WIN using the maximum bit of the register.
is clipped to add distortion to the input musical tone signal WIN. That is, the distortion generation processing section 21 includes adders 41 and 42.
, 43, 44, 45, 46, multipliers 47, 48, 49,
It is equipped with 51. Adders 41, 42, 43, 44, 4
5 and 46 add the input signals, and a multiplier 47,
48, 49, and 51 multiply the input signal by "2" input as a multiplication coefficient. Therefore, the distortion generation processing section 21
are adders 41, 42, 43, 44, 45 and multiplier 4
After repeating addition processing and multiplication processing in steps 7, 48, 49, and 51, the adder 46 finally performs addition processing to multiply the input musical tone signal WIN by 2048. Normally, when the input musical tone signal WIN is multiplied by 2048, the register in the DSP 12 overflows, and the input musical tone signal WIN can be clipped. In this way, the input musical tone signal WIN
Clip processing is performed on the input musical tone signal WIN by increasing the value and overflowing the register of the DSP 12. If the input musical tone signal WIN is generated from a guitar or the like, the waveform of the first input musical tone signal WIN will be extremely , and subsequent waveforms become very small. Therefore, unless the input musical tone signal WIN is increased by nearly 1000 to 2000 times, distortion processing will only be applied to the first waveform.
That is, this is because no distortion is added. In this way, when distortion processing is performed by multiplying the input musical tone signal WIN by 2048 and overflowing the register of the DSP 12, the input musical tone signal WIN input first and the input musical tone input thereafter are Distortion processing, that is, distortion can be applied appropriately even to the input musical tone signal WIN which has a large difference from the signal WIN. The distortion generation processing section 21 receives the input musical tone signal WI.
Distortion is added to N and output to the filtering processing section 22 as a clip signal WDST.

The filtering processing section (filtering means) 22 is a normal first-order low-pass filter, and includes multipliers 61 and 62, an adder 63, and a delay element 64. The multiplier 61 multiplies the clip signal WDST from the distortion generation processing unit 21 by the filter coefficient PFI0 and outputs the result to the adder 63, and the adder 63 multiplies the signal from the multiplier 61 by the data from the multiplier 62. Then, the filter signal W
It is output as FIL to the mixing processing section 23 and also to the delay element 64. The delay element 64 delays the input filter signal WFIL by one input musical tone signal WIN and outputs the delayed signal to the multiplier 62. A filter coefficient PFI1 is further input to the multiplier 62, and the multiplier 62
2 is a filter coefficient PFI1 applied to a filter signal WFIL delayed by one input musical tone signal WIN by a delay element 64.
is multiplied and output to the adder 63. The adder 63 adds the output of the multiplier 62 and the output of the multiplier 61 and outputs the result as a filter signal WFIL.

The mixing processing section (mixing means) 23 has the following formula (1
) is performed to perform the mixing process, and the mixing result is output as the output signal output musical tone signal WOUT. WOUT=WENV×WFIL×Pa
+(1-WENV)×WIN×Pb... (1) That is, the mixing processing section 23 has multipliers 71, 72, 73, 7
4 and adders 75 and 76. The adder 75 receives the envelope output signal W from the level detection processing section 20.
ENV is input, and "
1'' is input, and the adder 75 adds an addition constant 1 to the envelope output signal WENV and outputs the result to the multiplier 73. Since the envelope output signal WENV is now input as a negative input to the adder 75, the adder 75 changes the envelope output signal W from 1 as a result.
ENV is subtracted and output to the multiplier 73. The filter constant Pb is input to the multiplier 73, and the multiplier 73 inputs the filter constant Pb to the output of the adder 75.
is multiplied and output to the multiplier 74. As this filter constant Pb, "1" is currently adopted. The input musical tone signal WIN is input to the multiplier 74 , and the multiplier 74 multiplies the output of the multiplier 73 by the input musical tone signal WIN and outputs the result to the adder 76 .

On the other hand, the multiplier 71 receives the envelope output signal WENV from the level detection processing section 20 and also receives the filter constant Pa. and outputs the multiplication result to the multiplier 72. The multiplier 72 further includes a filtering processing section 22.
The multiplier 72 inputs the filter signal WFIL from the multiplier 71 to the filter signal WFIL.
It is multiplied by L and output to the adder 76.

Adder 76 adds the multiplication result of multiplier 72 and the multiplication result of multiplier 74, and outputs the addition result as output musical tone signal WOUT. Therefore, the mixing processing section 23 performs the mixing process according to the above equation (1), mixes the input musical tone signal WIN and the filter signal WFIL according to the envelope value, and outputs the output musical tone signal WOUT.
Output as .

FIG. 3 is a specific circuit configuration diagram of the DSP 12, which includes a program memory 81, a control circuit 82, an input register (PI) 83, and a coefficient memory (P).
84, work memory (W) 85, flag register (SF
0) 86, a multiplication section 87, an addition/subtraction section 88, an output register (OR) 89, and the like.

The program memory 81 stores a program as a distortion circuit of each invention of the present application.
This program is written by the CPU 14 shown in FIG. An address counter (not shown) is connected to the program memory 81.
The program contents are sequentially supplied to the control circuit 82 by addressing the address counter.

The control circuit 82 controls each part of the DSP 12 according to the program in the program memory 81 to execute the distortion processing according to each invention of the present application, and the details of the processing will be described later.

The input register (PI) 83 receives the input musical tone signal WIN, and the input register (PI) 83 receives the input musical tone signal WIN.
After temporarily storing this input musical tone signal WIN, the internal bus 9
0 to each part of the DSP 12. This input musical tone signal WIN is a signal input from the A/D converter 11 of FIG. 1, and is sampled at predetermined sampling intervals.

The coefficient memory (P) 84 is a register for storing various coefficients necessary for the DSP 12 to perform distortion processing. These various coefficients are
It is stored in the ROM 15 in FIG.
The coefficients are read from the OM 15 and written into the coefficient memory (P) 84. As shown in FIG. 4 as a memory map of the coefficient memory (P) 84, the coefficients set in the coefficient memory (P) 84 have a data name PDEC at address 0.
The envelope attenuation coefficient (0.5) is stored as the address 1, and the filter coefficient (0.5) is stored as the data name PFI0.
4000H) has data name PFI1 at its address 2.
The filter coefficient (4000H) is stored as address 3, and the filter constant (a=1
), and a filter constant (b=1) is further set at address 4 as data name Pb.

The work memory (W) 85 stores the input musical tone signal WIN inputted through the input register (PI) 83, the data of the calculation results in the multiplication section 87 and the addition/subtraction section 88, which will be described later, and the output musical tone signal WOUT. This is work memory for temporary storage. This work memory (W) 8
For example, as shown in the memory map of the work memory (W) 85 in FIG. The envelope output signal is sent to its address 2 with the data name WDS
The clip signal is stored as T, the filter signal is stored as data name WFIL in address 3, the output musical tone signal is stored as data name WOUT in address 4, and the constant (7FFFH) is stored as data name WONE in address 5.
=1) is stored.

The flag register (SF0) 86 is set with a flag F (AR) from the addition/subtraction unit 88, which will be described later.
The flag F (AR) to be set is set in work memory (W) 8.
Output to 5. Based on this flag register (SF0) 86, writing of data to the work memory (W) 85 is inhibited and canceled as will be described later. That is, when the flag F (AR) of "0" is set in the flag register (SF0) 86, the flag register (SF0) 86 is set to the work memory (W) 85.
Prohibits writing data to.

The multiplier 87 includes gates 91 and 92, registers (M0) 93 and (M1) 94, a gate 95, and a multiplier 9.
6 and register (MR) 97, and gate 9
Outputs from the coefficient memory (P) 84, work memory (W) 85, and input register (PI) 83 are input to 1 and 92. The gates 91 and 92 are connected to the control circuit 8.
2 controls its operation, and controls which input data is output to register (M0) 93 and register (M1) 94. Register (M0) 93 temporarily stores data input through gate 91, outputs it to multiplier 96, and feeds it back to gate 91. Register (M1) 94 temporarily stores data input through gate 92, outputs it to multiplier 96 through gate 95, and feeds it back to gate 92. The gate 95 also receives data from an adder/subtracter 88, which will be described later, and the gate 95 also receives data from the control circuit 82.
selects data from register (M1) 94 and addition/subtraction section 88 and outputs it to multiplier 96.

Multiplier 96 multiplies data input from register (M0) 93 and register (M1) 94, and outputs the result of the operation to register (MR) 97. The register (MR) 97 temporarily stores the multiplication result of the multiplier 96 and then outputs it to the gate 92 and the addition/subtraction section 88 . The addition/subtraction unit 88 includes gates 101 and 102, a register (A0) 103, a register (A1) 104, gates 105 and 106, an adder/subtractor 107, a register (AR) 108, a clipper 109, a register (SR) 110, and the like. Gate 101, 1
02 receives the outputs from the coefficient memory (P) 84, work memory (W) 85, and input register (PI) 83. The operation of the gates 101 and 102 is controlled by the control circuit 82, and the input data is transferred to the register (A0) 103 and the register (A1).
104 is controlled. The register (A0) 103 temporarily stores the data input through the gate 101, outputs it to the gate 105, and feeds it back to the gate 101. Register (A1) 104 temporarily stores data input through gate 102 , outputs it to gate 106 , and feeds it back to gate 102 . The gate 105 has a register (MR
)97 is also input, and gate 105
operates under the control of the control circuit 82 and registers (A0
) 103 and the data from the multiplier 87 are selected and output to the adder/subtracter 107 . In addition to the data from the register (A1) 104, the gate 106 also has an adder/subtractor 1.
The data of the operation result of 07 is stored in the register (AR) 108.
The gate 106 operates under the control of the control circuit 82 to select the input data and output it to the adder/subtractor 107 .

The adder/subtractor 107 performs addition or subtraction processing on the input data, outputs the result of the operation to the clipper 109 and the gate 106, and converts the maximum bit of the result of the operation into flag data F(
AR) to the flag register (SF0) 86. The clipper 109 is for preventing data overflow, and the data that has passed through the clipper 109 is supplied to a register (SR) 110. Furthermore, this clipper 109 has a built-in bit shifter, and can shift the input data by one bit to obtain a value obtained by processing the data twice. Register (S
The output of R) 110 is output to the gate 95 of the multiplier 87, and is also supplied to the work memory (W) 85 via the internal bus 90 as the result of processing for a certain note.

The calculation results in the multiplication section 87 and the addition/subtraction section 88 are outputted from the addition/subtraction section 88 to the work memory (W) 85 via the bus 90, and the data for which all the calculation processing has been completed are sent to the work memory (W) 85. ) 85 to the output register (OR) 89;
outputs the input data to a musical tone output device or the like.

Next, the operation will be explained. The invention according to claim 1 outputs a signal obtained by subjecting the input musical tone signal WIN to distortion processing by changing the mixing ratio of the input musical tone signal WIN and the input musical tone signal WIN in accordance with the signal level of the input musical tone signal WIN. Its feature is that the musical sound signal WIN is output as a sound close to a real sound.

The distortion circuit 10 has its DSP
As shown in FIG. 6, the input musical tone signal WIN is inputted (step S1), the envelope is extracted (step S2), and the input musical tone signal WIN is clipped (distorted) (step S3). , filtering processing (step S4), mixing processing (step S5)
and output processing (step S6).

The distortion circuit 10 includes a DSP 12
In order to cause the DSP 12 to perform these processes, the CPU 14 first performs program setting processing and various data setting processing on the DSP 12. That is, claim 1 is determined by the DSP 12.
In order to perform the distortion processing described, first,
It is necessary to set programs necessary to perform the intended distortion processing and data used to execute the programs in the program memory 81, coefficient memory (P) 84, and work memory (W) 85.

Therefore, the CPU 14 reads the necessary programs and data from the ROM 15 and stores them in the program memory 81, coefficient memory (P) 84 and work memory (W).
Set to 85. As shown in FIG. 4, the coefficient memory (P) 84 stores the damping coefficient PDEC and the filter coefficient PFI.
0, filter coefficient PFI1, filter constant Pa
and filter constant Pb are set, and the work memory (
W) 85 is set to a constant WONE, as shown in FIG.

Each process performed by the DSP 12 will be explained below. Input Processing Input processing is a process of capturing the input musical tone signal WIN into the work memory (W) 85 of the DSP 12, and the input musical tone signal WIN is captured into the work memory (W) 85 via the input register (PI) 83. . That is, as shown in FIG.
(Step S10)
1). As described above, this input musical tone signal WIN is obtained by sampling an analog tone signal at a predetermined sampling timing and converting it into a digital signal by the A/D converter 11 of FIG.

Envelope Extraction Process When the input musical tone signal WIN is stored in the work memory (W) 85, an envelope extraction process is performed on the input musical tone signal WIN. That is, the input musical tone signal WIN is read from the work memory (W) 85, transferred to the register (A0) 103 via the gate 101, and set (step S201). Then, the envelope output signal WE
NV is read from the work memory (W) 85 and set in the register (A1) 104 via the gate 102 (step S201). When data is set in register (A0) 103 and register (A1) 104, the data is set in register (A0) through gates 105 and 106.
The input musical tone signal WIN and the envelope output signal WENV of the register (A1) 103 and register (A1) 104 are transferred to the adder/subtractor 107, and a subtraction process (WENV) is performed to subtract the input musical tone signal WIN from the envelope output signal WENV.
-WIN), and the result of the subtraction is stored in the register (AR
) 108 (step S202). next,
The most significant bit of this register (AR) 108, that is, the data indicating the code, is transferred and set as the sign flag F (AR) to the flag register (SF0) 86, and the input set in the register (A0) 103 in step S201 is set. The musical tone signal WIN is passed through the adder/subtractor 107 and transferred to the register (AR) 108 (step S203).
). Furthermore, the input musical tone signal WIN of this register (AR) 108 is set in the register (SR) 110 (step S204).

Next, in step S203, the code flag F(AR) set in the flag register (SF0) 86 becomes 1.
In other words, whether it is negative or positive (negative when the sign flag F (AR) is 1, positive when the sign flag F (AR) is 0) (step S205), ) is 1, that is, when the subtraction result in step S202 is positive, the input musical tone signal WIN set in the register (SR) 110 in step S204 is set as the envelope output signal WENV at address 1 of the work memory (W) 85. (Step S206
). On the other hand, in step S205, the flag register (SF
0) When the sign flag F(AR) set to 86 is not 1, that is, the subtraction result (WEN
When the input waveform (V - WIN) is negative, that is, the input waveform is smaller than the envelope waveform, the register (S
R) 110 to the work memory (W), and the process directly advances to step S207.

In step S207, the coefficient memory (
P) 84 damping coefficient PDEC in register (M0) 93
The envelope output signal WENV of the work memory (W) 85 is transferred to the register (M1) 94 and set. Next, the attenuation coefficient PDEC of this register (M0) 93 and the envelope output signal WENV of the register (M1) are transferred to the multiplier 96 for multiplication processing (PDEC × WENV), and the multiplication result is sent to the register (MR). 97 to the register (A
0) is set to 103 (step S208).

Next, the data of this register (A0) 103 is transferred to the register (AR) 108 and the register (SR).
) 110 as the envelope output signal WENV at address 1 of the work memory (W) (steps S209, S210, S211). Through the above processing, the DSP for one sampling input musical tone signal WIN is
This means that the envelope extraction process in step 12, that is, the level detection process has been completed, and by sequentially repeating this process, envelope extraction processes can be performed for the input musical tone signals WIN that are sequentially input at sampling timings. That is, when the input waveform is smaller than the envelope waveform, that is, (WENV −
WIN) < 0, the envelope output signal WEN
The envelope output signal WENV is attenuated by multiplying V by an attenuation coefficient PDEC. Therefore, as shown in FIG. 9, even when the input musical tone signal WIN gradually attenuates, an appropriate envelope output can be ensured.

Clip Processing This clip processing is distortion generation processing in the distortion generation processing section 21 in FIG. By repeating this process, the maximum bit of the register of the DSP 12 is overflowed, and the input musical tone signal WIN is clipped by the maximum bit of the register. That is, specifically, as shown in FIG.
(step S301), and register (A0) 1
03 and the input musical tone signal WIN set in the register (A1) 104, the addition/subtraction unit 88 repeats the multiplication process five times (steps S302, S30).
3, S304, S305, S306).

This multiplication process is performed by register (A0) 103 and register (A1) 104, as shown in FIG.
The input musical tone signal WIN set to
7, and the input musical tone signal WIN is transferred to the adder/subtractor 107.
Add them together and store the addition result in register (AR) 10.
8 (step S310). This register (
The added data of AR) 108 is transferred to the clipper 109, which doubles the added data by bit shifting and transfers the doubled data to the register (SR) 110 (step S
311). The data that has been subjected to addition and multiplication processing in this way is transferred again to register (A0) 103 and register (A1) 104, and the same processing is repeated five times to perform the multiplication processing shown in FIG. Do this 5 times. When the multiplication process is completed in this way, Figure 1
0, register (A0) 103 and register (
A1) The data set in 104 is added to/subtracted by adder/subtractor 107.
The input musical tone signal WIN is multiplied by 2048 as described above, and the addition result is transferred to the register (AR) 108 (step S3).
07). The input musical tone signal WIN multiplied by 2048 is transferred from the register (AR) 108 to the work memory (W) 85 via the register (SR) 110.
W) Set address 2 of 85 as a clip signal WDST (steps S308, S309).

The clip signal WDST set in the work memory (W) 85 overflows the register of the DSP 12 and becomes a signal clipped at the maximum bit of the register. Therefore, by clipping the input musical tone signal WIN by a predetermined value (maximum bit of the register), distortion can be added to the input musical tone signal WIN. By sequentially repeating the above processing, clip processing (distortion generation processing) can be performed on the input musical tone signal WIN input at each sampling timing.

Filtering Process This filtering process is performed by the filtering processing unit 22 in FIG. 2, and specifically, is executed by the DSP 12. That is, as shown in FIG. 12, the DSP 12 first transfers the clip signal WDST in the work memory (W) 85 to the register (M1) of the multiplication unit 87, and
The filter coefficient PFI0 in the coefficient memory (P) 84 is transferred to the register (M0) 93 (step S401). The data in register (M0) 93 and register (M1) 94 are transferred to multiplier 96, and multiplication processing (
WDST×PFI0) and transfers the multiplication result to the register (A0) 103 of the addition/subtraction unit 88 (step S402). This multiplication result register (A0) 103
Transfer from the register (MR) 97 of the multiplication section 87 to the gate 105 of the addition/subtraction section 88;
The signal is transferred to the register (A0) via the adder/subtractor 107, the register (AR) 108, the clipper 109, the register (SR) 110, and the gate 101. When the transfer of this multiplication result to the register (A0) 103 is completed, the filter signal WFI of the work memory (W) 85 is
L is read out and transferred to the register (M1), and the filter coefficient PFI1 of the coefficient memory (P) 84 is read out and transferred to the register (M0) 93 (step S4).
02).

Next, the data in the register (A0) 103 is transferred to the register (AR) 108 via the gate 105 and the adder/subtractor 107, and the filter coefficient PFI1 of the register (M0) 93 and the filter signal of the register (M1) 94 are transferred. WFIL is transferred to the multiplier 96 and multiplied (PFI1×WFIL). The result of this multiplication processing is added to the register (A0
) 103 (step S403). The multiplication result of this register (A0) 103 and register (AR) 1
08 data (WDST
I1×WFIL+WDST×PFI0) and transfers the addition result to the register (AR) 108 (step S404). The data of the addition result transferred to the register (AR) 108 is transferred to the register (SR) 110.
The filter signal WFIL is set at address 3 of the work memory (W) 85 via the filter signal WFIL (step S405).
, S406). By sequentially repeating the above processing, filtering processing can be performed on the clipped signal WDST.

Mixing Process This mixing process is the process of the mixing processing section 23 shown in FIG. 2, and performs the process of equation (1) above. Specifically, DSP
12. That is, as shown in FIG.
4 and filter constant P in coefficient memory (P) 84.
a to the register (M0) 93 of the multiplier 87 (
Step S501). The filter constant Pa of the register (M0) 93 and the envelope output signal WENV of the register (M1) 94 are transferred to the multiplier 96, where they are multiplied (WENV × Pa), and the multiplication result is sent to the adder/subtracter 88. Transfer to register (A0) 103 (
Step S502). The multiplication result (WENV × Pa) of this register (A0) 103 is transferred to register (AR) 10.
8, register (M) via register (AR) 110
1) Transfer to 94 (steps S503, S504, S5
05), filter signal WFI of work memory (W) 85
L is read out and transferred to the register (M0) 93 (step S505).

The filter signal WFIL of this register (M0) 93 and the multiplication result (WE
NV × Pa ) is transferred to the multiplier 96 for multiplication processing, and the multiplication result (WENV × Pa × WFIL ) is transferred to the register (A0) 103 (step S506
). This multiplication result (WENV × Pa × WFIL)
from register (A0) 103 to register (AR) 10
8 to the register (A0) 110, and further transferred from the register (SR) to the work memory (W) 85 and written to address 4 as the output musical tone signal WOUT (step S509).

Next, the constant W is obtained from the work memory (W) 85.
Read ONE and register (A1) of addition/subtraction unit 88
104 and reads the envelope output signal WENV from the work memory (W) 85 and transfers it to the register (A0) 103 of the addition/subtraction section 88 (step S51
0). The envelope output signal WENV of this register (A0) 103 and the constant W of the register (A1) 104
ONE and is transferred to the adder/subtractor 107, and the constant WONE is
Subtract the envelope output signal WENV from (WON
E-WENV) and store the subtraction result in the register (A
R) 108 (step S511). Subtraction result of register (AR) 108 (WONE - WENV
) is transferred to the register (M1) 94 of the multiplier 87 via the register (SR) 110 (steps S512, S
513), filter constant Pb of coefficient memory (P) 84
is transferred to the register (M0) 93 of the multiplier 87 (step S513). Filter constant Pb of register (M0) 93 and subtraction result (WONE) of register (M1) 94
-WENV) is transferred to the multiplier 96, and multiplication processing (Pb
×[WONE-WENV]) and transfers the multiplication result to the register (A0) 103 of the addition/subtraction unit 88.

[0047] Now, the constant WONE is 7FFF.
Since H=1 is set, the subtraction result in step S511 is (1-WENV), and this subtraction result is multiplied by the filter constant Pb in step S514. Therefore, the multiplication process in step S514 is
(1-WENV)×Pb will be multiplied. This multiplication result ([1-WENV]×Pb) is transferred from the register (A0) 103 to the register (M1) 94 of the multiplication unit 87 via the register (AR) 108 and the register (SR) 110 (steps S515 and S516). ,
S517), the input musical tone signal W from the coefficient memory (P) 84
IN is read and transferred to the register (M0) 93 of the multiplication section 87. Input musical tone signal W of this register (M0) 93
The multiplication result of IN and register (M1) 94 is transferred to the multiplier 96 and multiplied, and the multiplication result ([1-WENV
]×Pb×WIN) in the register (A
0) 103 (step S518). When the transfer of the multiplication result to register (A0) 103 is completed,
The output signal WOUT is read from the work memory (W) 85 and transferred to the register (A1) 104 (step S518). The multiplication result ([1-WENV]×Pb×WIN) of these registers (A0) 103 and the output signal WOUT of register (A1) 104 are added to the adder/subtractor 1.
07 and performs addition processing (WOUT + ([1-WE
NV]×Pb×WIN) and transfers the addition result to the register (AR) 108 (step S5
19).

The output signal WOUT currently stored in the work memory (W) 85 is the one stored in step S509, that is, the multiplication result (WENV × Pa × WFIL) in step S506, so it is The addition result is the calculation result of the multiplier in (1) above, that is, WENV ×Pa ×WFIL + (
1-WENV)×Pb×WIN.

This multiplication result is stored in register (AR) 108.
from the work memory (
W) 85 (steps S520, S521), and writes it to address 4 as the output signal WOUT (step S521). Through the above processing, the arithmetic processing of equation (1) above can be performed, and by this arithmetic processing, the input musical tone signal WIN that has been subjected to distortion processing, that is, the input that has been subjected to clip processing and filter processing. A musical tone signal WIN, a raw input musical tone signal WIN,
Set the mixing ratio corresponding to the envelope value,
Can be mixed. By sequentially repeating this mixing process, it is possible to mix the input musical tone signals WIN sampled at predetermined sampling timings.

Therefore, when the level of the input musical tone signal WIN is large, the envelope output signal WENV becomes large in the above equation (1), and the input musical tone signal WIN subjected to distortion processing is converted into the output musical tone signal WOUT.
As the level of the input musical tone signal WIN decreases, the envelope output signal WENV in the above equation (1) decreases, and a sound close to a real tone can be set as the output musical tone signal WOUT.

Output processing Output processing outputs the output signal WOUT generated by the above mixing processing and set in the work memory (W) 85 to the DSP.
12 through the D/A converter 13 of FIG. 1 as an output musical tone signal. That is, the DSP 12 is
As shown in FIG. 14, the output signal WOUT is read from the work memory (W) 85, transferred to the register (OR) 89 (step S601), and sent from the register (OR) 89 to the D/A converter 13 at a predetermined timing. Output.

By repeating the above envelope extraction processing, clipping processing, filtering processing, and mixing processing for each input musical sound signal WIN inputted at a predetermined sampling timing, distortion processing is performed on the musical sound signal generated by a guitar or the like. In addition, the input musical sound signal WIN which has undergone clipping and filtering processing and the raw input musical sound signal WIN can be processed according to the magnitude of the envelope value of the input musical sound signal WIN.
It is possible to change the mixing ratio with the output signal WOUT and output it as the output signal WOUT. Therefore, the input musical tone signal WIN
When the envelope value of the input musical tone signal WIN is small, that is, when the level of the input musical tone signal WIN is low, the raw input musical tone signal W
It is possible to output an output signal WOUT with a high mixing ratio of IN, and it is possible to output a sound close to a real sound as the output signal WOUT.

FIGS. 15 and 16 are diagrams showing an embodiment of the distortion circuit according to the second aspect of the invention. The distortion circuit of this embodiment is applied to a distortion circuit similar to the distortion circuit 10 of the above embodiment, and the same components are given the same reference numerals and their explanations will be omitted. In particular, the same step numbers as those in the basic processing flow of the distortion circuit shown in FIG. 6 are given the same step numbers, and the explanation thereof will be omitted.

In this embodiment, the input musical tone signal is compared with a predetermined reference value (threshold level), and if the input musical tone signal is larger than the predetermined reference value, the input musical tone signal is subjected to distortion processing and is output as an output signal. , when the input musical tone signal is less than the reference value, the input musical tone signal is output as is as an output signal. Therefore, in this embodiment, the threshold level W as this reference value is
TH is stored in the ROM 15, and in processing the DSP 12, the CPU 14 reads the threshold level WTH from the ROM 15 and writes it into the work memory (W) 85 of the DSP 12.

FIG. 15 is a flowchart showing the basic processing flow of the distortion circuit 10 of this embodiment.
Steps S1 to S4 and S6 in this flowchart are the same processes as steps S1 to S4 and S6 in FIG. 6 of the above embodiment. That is, in steps S1 to S4, the input musical tone signal WIN is captured, and the envelope value of the input musical tone signal WIN is extracted, and the extracted envelope output signal WENV is stored in the work memory (W). Write to 85. Next, input musical tone signal WIN
Clip processing is performed on the clipped signal WDST, and filtering processing is performed on the clipped signal WDST.
As the filter signal WFIL, the work memory (W
)85. Therefore, work memory (W)8
5, an input musical tone signal WIN, an envelope output signal W
ENV and filter signal WFIL have been written.

Next, a switching process, which is a feature of this embodiment, is performed to determine the output signal WOUT. This switching process is performed by the DSP 12, and as shown in FIG.
The input musical tone signal WIN is read from the work memory (W) 85 and transferred to the register (A0) 103 of the addition/subtraction section 88 (step T101). Register (A0) 103
The input musical tone signal WIN transferred to is further transferred to the register (
AR) 108 and register (SR) 110 to the work memory (W) 85, and written to address 4 as the output signal WOUT (step T1).
02, T103, T104).

[0057] The input musical tone signal WIN is converted into the output signal WOUT.
When the envelope output signal WEN is written from the work memory (W) 85 to the work memory (W) 85,
Read V and transfer it to register (A0) 103,
Similarly, the threshold level WTH is read from the work memory (W) 85 and transferred to the register (A1) 104 (step T105). The envelope output signal WENV of the register (A0) 103 and the threshold level WTH of the register (A1) 104 are transferred to the adder/subtractor 107, the subtractor 107 performs subtraction processing (WTH-WENV), and the subtraction result is sent to the register ( A1) Transfer to 108 (
Step T106).

Next, the filter signal WFIL is read from the work memory (W) 85 and stored in the register (A0) 103.
(Step T107). Step T106
The subtraction result is transferred to the register (AR) 108, and the most significant bit of this subtraction result is set to the sign flag F(A
R) to the flag register (SF0) 86 (
Step T108). Also, the filter signal WFIL transferred to the register (A0) 103 in step T107 is transferred to the register (AR) 108 (step T108).
, and further transferred to the register (AR) 110 (step T109).

In this state, the code flag F(AR) is 1,
This code flag F (AR) is used to register the flag register (SF
0) Whether 86 is set to 1, i.e.
It is checked whether the subtraction result (WTH-WENV) is negative (step T110), and if it is negative, it is determined that the level of the envelope output signal WENV, that is, the input musical tone signal WIN is greater than the threshold level WTH, and the register ( The filter signal WFIL set in SR) 110, that is, the signal subjected to clipping and filtering processing, is transferred to the work memory (W) 85 and written at address 4 as the output musical tone signal WOUT (step T111). Meanwhile, in step T110, the flag register (SF0)
86 is not set to 1, it is determined that the level of the input musical tone signal WIN is lower than the threshold level WTH, and the process continues as is, that is, in step T104.
The output musical tone signal WOUT set at address 4 of the work memory (W) 85 is directly output as the musical tone signal WOU.
Set it as T.

Therefore, the envelope value of the input musical tone signal WIN is compared with a predetermined threshold level WTH, and when the envelope value is larger than the threshold level WTH, the filter signal WFIL, that is, the clipping and filtering process is performed as the output musical tone signal WOUT. The input musical tone signal WIN can be set. Furthermore, when the envelope value of the input musical tone signal WIN is smaller than the threshold level WTH, the input musical tone signal WIN can be set as the output musical tone signal WOUT. As a result, when the input musical tone signal WIN exceeds a predetermined level, the input musical tone signal WIN can be subjected to distortion processing and set as the output musical tone signal WOUT, and when the input musical tone signal WIN is below the predetermined level, the input musical tone signal WIN can be set as the output musical tone signal WOUT.

The output musical tone signal WO set in this way
UT can be output in the same way as in the above embodiment. As a result, for input musical sound signals WIN exceeding a predetermined level, appropriate distortion processing can be applied to output musical tones with rich musical accuracy, and for input musical sound signals WIN below a predetermined level, raw sounds can be output. It can be output. In each of the above embodiments, the input musical tone signal WIN and the input musical tone signal WIN subjected to distortion processing are mixed, or the input musical tone signal WIN is mixed at a predetermined level.
Input musical tone signal WI subjected to IN and distortion processing
Although we have explained the case where the input musical tone signal W
Up to a predetermined level of IN, the input musical tone signal WIN is mixed with the input musical tone signal WIN that has been subjected to distortion processing, and when it reaches a predetermined level or below, the input musical tone signal W
It is also possible to output IN as is.

[0062]

According to the invention described in claim 1, the input musical tone signal and the musical tone signal are distorted and filtered (
A signal subjected to distortion processing) can be mixed into an output musical tone signal according to the level of the musical tone signal, and when the level of the input musical tone signal is high, it is possible to perform a musical tone with rich musicality. In addition, when the level of the input musical tone signal is narrowed down, the mixing ratio of the input musical tone signal to the signal subjected to distortion processing can be increased, and an output musical tone signal close to the natural sound can be output. As a result, when the level of the input musical sound signal is low, it is possible to perform a sound close to a real sound.

According to the second aspect of the invention, the input musical tone signal and the signal obtained by applying distortion processing to the input musical tone signal can be switched and outputted depending on the level of the input musical tone signal, and the input musical tone signal can be outputted at a predetermined level. When the level exceeds the input musical tone signal, an output signal that has been subjected to distortion processing is output, so that musical tones with rich musicality can be played. By outputting an input musical tone signal as an output musical tone signal, it is possible to perform a sound close to a real sound.

[Brief explanation of the drawing]

FIG. 1 is a circuit block diagram showing an embodiment of a distortion circuit according to the invention as claimed in claim 1.

[Figure 2] Envelope value extraction processing in the DSP of Figure 1,
FIG. 3 is a configuration diagram showing a pseudo clipping process, a filtering process, and a mixing process.

FIG. 3 is a detailed circuit diagram of the DSP of FIG. 1;

FIG. 4 is a diagram showing the contents of a coefficient memory of the DSP in FIG. 3;

FIG. 5 is a diagram showing the storage contents of a work memory of the DSP in FIG. 3;

FIG. 6 is a flowchart showing basic processing in the DSP of FIG. 3;

FIG. 7 is a detailed flowchart of the input processing in FIG. 6;

FIG. 8 is a detailed flowchart of the envelope value extraction process of FIG. 6;

FIG. 9 is a diagram showing a relationship between an input musical tone signal waveform and an envelope waveform when multiplied by an attenuation coefficient.

FIG. 10 is a detailed flowchart of the clipping process shown in FIG. 6;

FIG. 11 is a detailed flowchart of the multiplication process in FIG. 10;

FIG. 12 is a detailed flowchart of the filtering process in FIG. 6;

FIG. 13 is a detailed flowchart of the mixing process of FIG. 6;

FIG. 14 is a detailed flowchart of the output processing of FIG. 6;

FIG. 15 is a flowchart showing basic processing by the DSP of the invention according to claim 2.

FIG. 16 is a detailed flowchart of the switching process in FIG. 15;

FIG. 17 is a block diagram of a conventional distortion circuit.

[Explanation of symbols]

10 Distortion circuit 11 A/D converter 12 DSP 13 D/A converter 14 CPU 15 ROM 16 RAM 20 Envelope extraction processing section 21 Clip processing section 22 Filtering processing section 23 Mixing processing section 81 Program memory 82 Control circuit 83 Input register ( PI) 84 Coefficient memory (P) 85 Work memory (W) 86 Flag register (SF0) 87 Multiplication section 88 Addition/subtraction section 89 Output register (OR)

Claims (2)

[Claims] 1. Level detection means for detecting the signal level of an input musical tone signal, distortion generation means for distorting the waveform of the input musical tone signal, and a predetermined filtering process applied to the musical tone signal outputted from the distortion generation means. filtering means; mixing means for mixing and outputting the input musical tone signal and the musical tone signal output from the filtering means at an arbitrary mixing ratio; and determining the mixing ratio by the mixing means based on the detection result of the level detecting means. A distortion circuit comprising: mixing ratio setting means for setting a mixing ratio; 2. Level detection means for detecting the signal level of an input musical tone signal, distortion generation means for distorting the waveform of the input musical tone signal, and predetermined filtering processing applied to the musical tone signal outputted from the distortion generation means. A distortion circuit comprising: filtering means; and selection means for selecting and outputting either an input musical tone signal or a musical tone signal output from the filtering means based on the detection result of the level detecting means.