JP2001067089A - Reverberation effect device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a reverberation sound that has not an excessively flat characteristic, is tasteful, and is somewhat peculiar, by controlling a mixing ratio when a signal outputted from a fist all-pass filter is added to a signal outputted from a second all-pass filter. SOLUTION: A delay coefficient determining circuit 61, a multiplification coefficient determining circuit 62, and a mixing ratio determining circuit 63 of a filter control circuit 6 creates filter coefficient DL1-DL4, a1-a4, and AMP 1-AMP2 from input parameters DTY, DSIP, ROUGH. When a signal outputted from a first all-pass filter is added to a signal outputted from a second all-pass filter, for example, setting mixing coefficients of the mixing ratio AMP1=0.5, AMP2=0.5, a frequency that has a peak and a frequency that has a dip occur due to a phase delay characteristic of the all-pass filters. Thus, an interesting effect different from a usual filter can be provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a parameter setting device and a reverberation effect device capable of appropriately setting parameters of a reverberation filter.

[0002]

2. Description of the Related Art Conventionally, a technique of using a filter including a delay circuit to artificially form a reverberation sound in an electronic musical instrument or the like has been generally known. For example, Tokuhei 1
JP-A-57799 and JP-A-2-287973 disclose a reverberation adding apparatus that obtains natural reverberation characteristics by using a plurality of filters including delay circuits and using a reverberation filter in which the filters are connected in parallel or in series. ing.

[0003] In such a reverberation adding apparatus, in order to obtain a more natural reverberation sound, it is desirable that a filter for providing reverberation has a frequency characteristic as flat as possible without a peak at a specific frequency. . For this purpose, in the prior art described in the above-mentioned publication, the number of delay stages of the delay circuits of several filters to be combined is made to be relatively prime. As a result, the combined filters do not have a common peak due to the number of delay stages,
Since reverberation is generated based on delay times that do not overlap each other, a reverberation filter having a flat frequency characteristic as a whole can be obtained.

[0004]

In the formation of reverberation using a reverberation filter in which a plurality of filters are combined, it is necessary for an operator to set (or correct) reverberation parameters in parallel. Or set (modify) all parameters of each filter connected in series
There is a need to. However, since the number of these parameters is very large as a whole, it has been very troublesome and difficult for the operator to appropriately set and correct all of them. In addition, if the parameters are improperly set, unnecessary frequency peaks may occur, and natural reverberation may not be obtained.

[0005] In an electronic musical instrument, an operator can select an effect program in order to impart various sound effects, and appropriate parameters are provided to a reverberation filter in a so-called factory preset according to the selected effect program. Some of them are, but in such a case, you can only select an effect program from a predetermined range, and if the operator wants a fine reverberation effect, it can not realize it There was a problem.

The present invention has been made in view of the above-mentioned problems in the prior art, and when a reverberation effect device is created using an all-pass filter, the reverberation effect is reduced to some extent so that the reverberation effect does not become too flat to produce a dull reverberation effect. It is a first object of the present invention to provide a reverberation effect device that can be attached.

It is a second object of the present invention to provide a reverberation effect device capable of setting parameters so that the reverberation filter does not have an inappropriate bias.

[0008]

To achieve this object, the present invention provides a first all-pass filter to which a signal to which reverberation is to be applied is inputted, and an output signal of the first all-pass filter being inputted. A second all-pass filter, adding means for adding a signal output from the first all-pass filter and a signal obtained from the second all-pass filter, and a mixing ratio of the signal added by the adding means. Controlling means for controlling the frequency characteristic of the output signal from the adding means, to generate peaks and gaps according to the characteristics of the first and second all-pass filters and the control of the mixing ratio by the control means. It is characterized by the following.

Further, the first and second all-pass filters each have a delay circuit, and the number of delay stages in the delay circuit of the first all-pass filter and the number of delay stages in the delay circuit of the second all-pass filter are mutually different. It is characterized by comprising a setting means for setting so as to be the same.

[0010]

When the signal output from the first all-pass filter and the signal obtained from the second all-pass filter are added, the mixing ratio is controlled. As a result, with respect to the frequency characteristics of the output signal from the adding means, peaks and gaps are generated according to the characteristics of the first and second all-pass filters and the control of the mixing ratio by the control means.

Each of the first and second all-pass filters has a delay circuit, and the number of delay stages in the delay circuit of the first all-pass filter and the number of delay stages in the delay circuit of the second all-pass filter are mutually prime. Set as follows. This realizes a remaining shared filter having a small density bias.

[0012]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a block configuration of an electronic musical instrument to which a reverberation filter parameter setting device according to an embodiment of the present invention is applied. This electronic musical instrument includes a performance information generating circuit 1, a musical tone synthesizing circuit 2, a filter circuit 3, various operators 4,
A control circuit 5 and a filter control circuit 6 are provided.

The performance information generating circuit 1 generates performance information such as a key code, a key on / off signal, and touch information according to the performance of a keyboard (not shown) by the operator. These pieces of performance information are input to the tone synthesis circuit 2. The various controls 4 are switches for setting a tone color and setting parameters relating to reverberation. The operation event of the operation element 4 is detected by the control circuit 5. When there is an operation event of the operator 4 relating to the timbre, the control circuit 5 provides the tone synthesis circuit 2 with timbre information according to the operation. The tone synthesis circuit 2 synthesizes a tone waveform with a tone corresponding to the input tone color information based on the input performance information, and outputs the synthesized tone waveform to the filter circuit 3.

Further, the operator uses the predetermined operator 4 to
Three parameters related to reverberation (the details will be described later)
Can be set respectively. When there is an operation event of a parameter related to reverberation, the control circuit 5 gives the filter control coefficients (three) to the filter control circuit 6 according to the operation. The filter control circuit 6 generates a filter coefficient based on the input filter control coefficient and supplies the filter coefficient to the filter circuit 3. The filter circuit 3 filters and outputs a tone signal according to the input filter coefficient.
Thereby, a reverberation sound is added to the musical sound signal.

Next, the configurations of the filter circuit 3 and the filter control circuit 6 will be described in detail.

FIG. 2 shows a detailed block configuration of the filter circuit 3 of FIG. The filter circuit 3 includes four all-pass filters 21, 22, 23, 24 and two multipliers 3.
1 and 32 and an adder 41. Multiplier 3
The multiplication coefficients of 1 and 32 are AMP1 and AMP2, respectively. The multiplication coefficients AMP1 and AMP2 are parameters provided from the filter control circuit 6 in FIG. 1, that is, some of the filter coefficients.

Four all-pass filters 21, 22, 2
3, 24 are connected in series in this order. The tone waveform signal input to the filter circuit 3 is input to the all-pass filter 21. An input terminal of the multiplier 31 is connected to a connection point between the all-pass filter 22 and the all-pass filter 23. The input terminal of the multiplier 32 is connected to the output terminal of the all-pass filter 24. The outputs of the multipliers 31 and 32 are input to the adder 41, and the addition result is output from the filter circuit 3.

FIG. 3 shows a detailed block configuration of the all-pass filter 21 of FIG. All-pass filter 21
Represents a delay circuit 51, adders 52 and 53, and a multiplier 5
4,55. The number of delay stages of the delay circuit 51 is DL
1. The multiplier coefficient of the multipliers 54 and 55 is a1. The number of delay stages DL1 and the multiplication coefficient a1 are parameters provided from the filter control circuit 6 in FIG. 1, that is, some of the filter coefficients.

The tone waveform signal input to the all-pass filter 21 is input to an adder 52 and a multiplier 54. The result of the multiplication by the multiplier 55 is input to the adder 52 as another input. The result of the addition by the adder 52 is
To enter. The output of the delay circuit 51 is input to the adder 53. The adder 53 has a multiplier 54 as another input.
Is input after reversing the sign. The addition result of the adder 53 is input to the multiplier 53 and becomes the output of the all-pass filter 21.

The all-pass filters 22, 23, and 24 have exactly the same configuration as the all-pass filter 21 shown in FIG. However, the parameters given to the all-pass filter 22 from the filter control circuit 6 in FIG. 1 are the number of delay stages DL2 and the multiplication coefficient a2, and the parameters given to the all-pass filter 23 are the number of delay stages DL3 and the multiplication coefficient a3. The number of delay stages is DL4 and the multiplication factor is a4.

As described above, the filter coefficient given from filter control circuit 6 to filter circuit 3 in FIG.
1 to DL4, multiplication coefficients a1 to a4, and multiplication coefficient AM
There are 10 P1 and AMP2.

Next, how the ten filter coefficients given to the filter circuit 3 are determined will be described in detail.

As described above, the operator can set three parameters related to reverberation using the predetermined operator 4. From an operator's point of view, these three parameters are parameters that determine the reverberation effect of the filter circuit 3. The control circuit 5 detects the operation of the operating element 4 and outputs the values of these three parameters to the filter control circuit 6 as filter control coefficients. The three parameters are density DTY, dispersion DISP, and roughness RO
UGH. Hereinafter, symbols DTY, DISP, ROUG
H indicates a parameter and its value, respectively. The filter control circuit 6 performs the above-described 1 based on these three parameters DTY, DISP, and ROUGH.
Generate and output zero filter coefficients.

FIG. 4 shows a detailed block configuration of the filter control circuit 6. The filter control circuit 6 includes a delay stage number determination circuit 61, a multiplication coefficient determination circuit 62, and a mixture ratio determination circuit 63. The delay stage number determination circuit 61 generates and outputs the delay stage numbers DL1 to DL4 based on the input density DTY. The multiplication coefficient determination circuit 62 is configured to control the number of delay stages DL1 to DL1 output from the distributed DISP and delay stage number determination circuit 61.
Based on DL4, multiplication coefficients a1 to a4 are generated and output. The mixture ratio determination circuit 63 generates and outputs multiplication coefficients AMP1 and AMP2 based on the roughness ROUGH.

Here, the principle of determining these filter coefficients will be described.

In FIG. 3, since the number of delay stages of the delay circuit 51 is DL1, the delay time is DL1 / FS (FS is a sampling frequency). Similarly, the delay times of the delay circuits of the other all-pass filters are DL2 / F, respectively.
S, DL3 / FS, and DL4 / FS. Further, the closer the value of the multiplication coefficient a1 (a2 to a4) is to “1”, the longer the reverberation time of the all-pass filter.

Such filter coefficients DL1 to DL4
When determining a1 to a4, when one parameter is set, the range of another parameter may be determined to some extent. For example, when the delay time of one filter is set, the delay time of another filter needs to be at least different from the delay time of the previously set filter. Ideally,
As described in the section of the prior art, all (here, 4
First, it is necessary to set the filter delay times (the number of delay stages) relatively prime. Thereby, the bias of the reverberation density distribution related to the delay time can be reduced.

In this embodiment, in order to set the number of delay stages DL1 to DL4 of the four all-pass filters to be relatively prime, the parameters are determined so as to satisfy the following conditions. The number of delay stages DL1 to DL4 is a positive integer. The number of delay stages DL1 to DL4 are relatively prime,
Each is a prime number. It is assumed that DL4 <DL2 <DL3 <DL1. The prime number is obtained by referring to a predetermined prime number table.

A prime number table for obtaining a prime delay stage number may be created, for example, as follows. First, the maximum number of delay stages that can be prepared by hardware is obtained. In this embodiment, it is assumed that there are 1024 stages. The range of the variable DTY (parameter given to the filter control circuit 6) for controlling the delay length is set to 128 levels from "0" to "127". The above-mentioned 1024 steps are logarithmically divided into 128 steps, and the prime number closest to the division position is extracted. The prime number table thus obtained is shown below.

7, 11, 13, 17, 19,
23, 29, 31, 37, 41, 43, 4
7, 53, 59, 61, 67, 71, 7
3, 79, 83, 89, 97, 101, 10
3,107,109,113,127,131,13
7,139,149,151,157,163,16
7,173,179,181,191,193,19
7, 199, 211, 223, 227, 229, 23
3,239,241,251,257,263,26
9,271,277,281,283,293,30
7,311,313,317,331,337,34
7,349,353,359,367,373,37
9,383,389,397,401,409,41
9,421,431,433,439,443,44
9,457,461,463,467,479,48
7,491,499,503,509,521,52
3,541,547,557,563,569,57
1,577,587,593,599,601,60
7,613,617,619,631,641,64
3,647,653,661,691,709,74
3,773,797,839,863,907,94
1,983,1021

Hereinafter, the leading data "7" of the prime number table is the 0th data, the next "11" is the first data, the next "13" is the second data, ...
Shall be called. The final data “1021” is the twelfth
This is the seventh data.

In processing, as shown in a flowchart described later, a parameter DTY (“0”) for determining a delay time is set.
, The DTY-th data of the prime number table is read based on
The number of delay stages DL3, DL2, DL4 is sequentially obtained by returning a predetermined number of data from the read position.
The details of the method of obtaining will be described later in detail. Since all elements of the prime number table are prime numbers, if the value read from the prime number table is set as the number of delay stages, the number of delay stages will be relatively prime.

The number of delay stages set in this way corresponds to the density of the impulse response of each filter. Conversely, the parameter name is DTY (Density) indicating the density (actually, the reciprocal of the density).
That is, when the parameter DTY is small, the number of delay stages is small, and the reverberation output (impulse response) when one impulse is input to the filter is frequently output and has a high density. Conversely, when the parameter DTY is large, the number of delay stages becomes large, and the impulse response becomes low in density.

Next, the principle of obtaining the multiplication coefficients a1 to a4 (represented by the symbol a) will be described. When the multiplication coefficient a is set to "1", the impulse input to the all-pass filter is output forever. Conversely, if the multiplication coefficient a is set to "0", the impulse is simply delayed and output only once. That is, the parameter a is a parameter related to the duration of the impulse response. Equation (1) below shows the impulse response of the all-pass filter of this embodiment. In addition,
Similarly to the case where the multiplication coefficients a1 to a4 are represented by the symbol a, the number of delay stages DL1 to DL4 is represented by the symbol DL.

[0036]

(Equation 1)

Thus, it can be seen that the impulse response of the all-pass filter attenuates at the equal ratio a from the time when the time DL / FS elapses after the input of the impulse. Here, as a condition for quantitatively determining the decay time, consider the decay time DT until the amplitude is attenuated by −20 dB. The relationship between the decay time DT and the multiplication coefficient a is defined by the following equation (2).

[0038]

(Equation 2)

In this embodiment, the parameter a is determined assuming that the attenuation time DT in each of the four all-pass filters 21 to 24 is equal. Therefore, the decay time D
When T is determined (because the number of delay stages DL has already been obtained as described above), the multiplication coefficient a (a1 to a4) is obtained. This decay time D as a parameter
Although T may be directly given, the range of the decay time DT is large and requires a large number of bits in processing.
In this embodiment, it is converted into another parameter (distributed DISP). The conversion formula between the time DT and the dispersion DISP is shown in the following formula (3).

[0040]

(Equation 3)

In the equation (3), when the dispersion DISP is “0”, the time DT is 16 ms, and the dispersion DISP is “1”.
27 ", the time DT is 16 seconds. The distributed DISP takes a positive integer from “0” to “127”. Since DISP indicates the degree of dispersion (time length) of the impulse response, the parameter name was set to Dispersion.

Next, mixing coefficients AMP1, AMP2
The principle of the determination will be described. Mixing coefficient AMP1, A
MP2 is determined based on the roughness ROUGH.
In FIG. 2, for example, when AMP1 = 0 and AMP2 = 1, the four all-pass filters 21 to 24 are in series, and the frequency characteristics are theoretically flat. On the other hand, A
When MP1 = 0.5 and AMP2 = 0.5, a peak frequency and a dip frequency appear due to the phase delay characteristic of the all-pass filter. Thereby, an interesting effect different from a normal filter can be obtained. The reason why the parameter name of ROUGH is roughness is that it is a parameter for obtaining such an effect.

Mixing coefficient A from roughness ROUGH
The conversion equation for calculating MP1 and AMP2 is the following equation (4).
Shown in

[0044]

(Equation 4)

The delay stage number determination circuit 61, the multiplication coefficient determination circuit 62, and the mixture ratio determination circuit 63 of the filter control circuit 6 shown in FIG. 4 are based on the above-described principle, and have the input parameters DTY, DISP, ROUGH. From the filter coefficients DL1 to DL4, a1 to a4, AMP1, AMP
2 is generated and output.

FIG. 5 is a flowchart for mainly explaining the procedure for determining the filter coefficients in the filter control circuit 6. First, when the parameter DTY (a positive integer in the range of "0" to "127" as described above) is set by the operation element 4 in step S1, the prime number table is referred to by the set value DTY in step S2. , And the obtained value is set as the number of delay stages DL1. For example, if DTY = 10 is set, the tenth data "4
"3" is set as the number of delay stages DL1.

Next, at step S3, a predetermined value D is obtained from DTY.
The result of subtracting the LR (the fixed value “3” in this embodiment) is set as DLP3, and it is determined in step S4 whether DLP3 is smaller than “0”. If DLP3 is smaller than "0", DL1-2 is calculated in step S6, the result is set in the number of delay stages DL3, and the process proceeds to step S7. Otherwise, in step S5, the above-mentioned prime number table is referred to by DLP3, the obtained value is set to the number of delay stages DL3, and the process proceeds to step S7.

Next, the result obtained by subtracting the predetermined value DLR = 3 from DLP3 in step S7 is defined as DLP2,
At 8, it is determined whether or not DLP2 is smaller than "0". DL
If P2 is smaller than "0", DL is determined in step S10.
3-2 is calculated, the result is set to the number of delay stages DL2, and the process proceeds to step S11. Otherwise, in step S9, the above-mentioned prime number table is referred by DLP2,
The obtained value is set in the number of delay stages DL2, and the process proceeds to step S11.

Next, in step S11, the result obtained by subtracting the predetermined value DLR = 3 from DLP2 is set as DLP4, and in step S12, it is determined whether DLP4 is smaller than "0".
If DLP4 is smaller than "0", DL2-2 is calculated in step S14, the result is set in the number of delay stages DL4, and the process proceeds to step S15. If not, in step S13, the above-mentioned prime number table is referred to by DLP4, and the obtained value is set in the number of delay stages DL4.
Go to 5.

In step S15, the number of delay stages DL1 to DL4
Is output to the filter circuit 3. Next, the parameter DISP input in step S16 and the number of delay stages D of each filter are set.
The multiplication coefficients a1 to a4 are calculated from L1 to DL4 and output to the filter circuit 3. Further, in step S17, mixing coefficients AMP1 and AMP2 are calculated from the parameter ROUGH and output to the filter circuit 3. Parameters a1 to a
4 and AMP1 and AMP2 are calculated by the above equation (1).
(4).

In the above procedure, first, the parameter DT
The number of delay stages DL1 is obtained by referring to the prime number table by Y, and the number of delay stages DL3, DL2, DL4 is sequentially obtained by returning three (DLR = 3) from the position on the table.
The value of DLR is not limited to “3” and may be any other appropriate value. The operator may be allowed to set the DLR. Further, in the above procedure, all-pass filters 21, 23, 2
The number of delay stages DL1, DL3, D in the order of 2, 24
Although L2 and DL4 are obtained, the order in which they are obtained is not limited to this.

In steps S4, S8, and S12, DLP3, DLP2, and DLP4 become smaller than "0"("0" to "12" indicating data in the table, respectively).
7 "), and in that case, in steps S6, S10 and S14," 2 "is subtracted from the number of delay stages DL1, DL3 and DL2 obtained before, and DL3 and DL3 are used for convenience. DL2 and DL4 are required,
Other methods may be used. For example, such a case is limited to a case where the initially set DLY is small.
Another prime number table to be referred to when Y is small may be provided. Further, four prime number tables for obtaining the number of delay stages from DTY may be provided, and the number of delay stages DL1 to DL4 may be obtained independently.

According to the above embodiment, if the operator specifies only three parameters, all the filter coefficients (10
Are automatically set. In addition, the set filter coefficients are selected such that the delay times of the filters are always relatively prime. Therefore, with a simple operation, appropriate parameters that do not have an inappropriate peak as a reverberation filter can be set. Also, a perfectly flat characteristic indicates that the reverberation sound has no habit at all,
If the reverberation without habit is not interesting, the roughness R
OUGH can add habits and control the degree.

In the above embodiment, the reverberation filter using only the all-pass filter has been described. However, the principle of parameter setting according to the present invention includes a plurality of filters using a delay circuit and a feedback configuration. It can be applied to circuits in general. For example, the present invention can be applied to the setting of a delay and a multiplication coefficient of a reverberation sound forming unit of Japanese Patent Publication No. 1-57999, which is cited as a conventional technique. Also, the connection of the all-pass filter is not limited to the one described in the embodiment,
It can be freely changed to another equivalent circuit. Further, the output of the all-pass filter 22 and the output of the all-pass filter 24 are mixed and controlled by the roughness ROUGH. However, the direct input and the output of the all-pass filter 24 may be mixed, or various methods may be used to obtain signals having different phase delay characteristics. Modifications are possible.

[0055]

As described above, according to the present invention, a desired amount of dips and gaps is generated,
The characteristic does not become too flat, and it is possible to generate a tasteful reverberation sound and a reverberation sound with a certain degree of habit. In addition, since the parameters of each filter are automatically generated so that the delay times of the delay circuits of the multiple filters that make up the reverberation filter are mutually prime, a reverberation filter with less bias in density is realized. Is done.

[Brief description of the drawings]

FIG. 1 is a block diagram of an electronic musical instrument to which a reverberation filter parameter setting device according to an embodiment of the present invention is applied.

FIG. 2 is a detailed block diagram of a filter circuit.

FIG. 3 is a detailed block diagram of an all-pass filter.

FIG. 4 is a detailed block diagram of a filter control circuit.

FIG. 5 is a flowchart for explaining a procedure for determining a filter coefficient;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Performance information generation circuit, 2 ... Tone synthesis circuit, 3 ... Filter circuit, 4 ... Various operators, 5 ... Control circuit, 6 ... Filter control circuit, 21, 22, 23, 24 ... All-pass filter, 31, 32 ... Multiplier, 41 Adder, 51 Delay circuit, 52, 53 Adder, 54, 55 Multiplier, 61
Delay stage number determination circuit, 62... Multiplication coefficient determination circuit, 63.

Claims (2)

[Claims] 1. A first all-pass filter to which a signal to which reverberation is to be applied is input; a second all-pass filter to which an output signal of the first all-pass filter is input; and a first all-pass filter. An adding unit that adds the output signal and a signal obtained from the second all-pass filter; and a control unit that controls a mixing ratio of the signal added by the adding unit. The output signal from the adding unit A reverberation effect device for generating peaks and gaps according to the frequency characteristics of the first and second all-pass filters and the control of the mixing ratio by the control means. 2. The first and second all-pass filters each have a delay circuit, and the number of delay stages in the delay circuit of the first all-pass filter and the number of delay stages in the delay circuit of the second all-pass filter are mutually different. 2. The reverberation effect apparatus according to claim 1, further comprising a setting unit that sets the original reverberation.