JPH0689097A - Effector - Google Patents

Abstract

PURPOSE:To efficiently control reflected sounds, such as resonance sounds generated by the body of the strings, which are dense and are large in quantity by integrally adjusting envelopes and/or distributions. CONSTITUTION:A digital signal processor(DSP) 20 uses a RAM 21 as a shift register and is capable of taking an output out of any stage of this shift register. This DSP 20 takes musical tone data out of prescribed plural addresses according to the data given from a CPU 10 and multiplies respective sets of the data by a prescribed coefft. Plural pieces of the musical tone data multiplied by this coefft. are thereafter added (convolution computation). The musical tone data subjected to the convolution computation is outputted as a resonant musical tone signal. The musical tone data are taken out of extremely many taps in order to simulate the resonance of the musical instrument, but the addresses of the many taps of this electronic musical instrument are determined overall so that the coefft. to be multiplied is determined overall by the envelope.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an effect imparting device for electronically forming an initial reflection of a musical tone signal to impart a resonance effect.

[0002]

2. Description of the Related Art When a musical tone of a stringed instrument such as a violin or a guitar is produced by an electronic musical instrument, it is necessary to simulate resonance caused by the body of the musical instrument. Resonance by the body of the instrument is caused by the formation of many reflected sounds in the body in a short time, so it is necessary to generate more reflected sounds with a long delay line in order to simulate this. .

In the case of changing the resonance characteristic with such a configuration, two parameters of a tap position (address: timing of reflected sound generation) and a coefficient (amount of reflected sound) of the delay line are changed for each reflected sound. There is a need to. However, it is virtually impossible to individually change the parameters of many reflected sounds, for example, hundreds of reflected sounds, so a plurality of parameter patterns for realizing the resonance characteristics of the musical instrument are set in advance and stored in the memory. , And any one of them is selected by a selection switch or the like.

[0004]

However, such a method has a drawback that only preset parameter patterns can be used and the degree of freedom in setting resonance characteristics is limited. Although the flexibility will increase if a large number of preset patterns are prepared, it will be difficult to find a desired pattern from many preset patterns because the cost will increase due to an increase in memory capacity.

Further, even if a pattern close to the desired one is found, it is not possible to obtain an ideal parameter pattern (resonance characteristic) because fine adjustment cannot be performed.

On the other hand, it is conceivable to open all tap positions and coefficient parameters so that the user can freely adjust them, but as described above, it is practically impossible to edit because the number is huge. there were.

It is an object of the present invention to provide an effect imparting device capable of editing many reflected sound parameters by a simple method.

[0008]

According to the invention of claim 1 of this application, in an effect imparting device for forming a plurality of continuous reflected sounds with respect to an inputted musical tone signal, an envelope forming means for forming an envelope, Coefficient controlling means for controlling the magnitudes of the plurality of continuous reflected sounds in accordance with the shape of the envelope formed by the envelope forming means.

According to the invention of claim 2 of this application, in the effect imparting device for forming a plurality of continuous reflected sounds with respect to the inputted musical tone signal, the function generating means for generating the distribution function and the function generating means are generated. The distribution control means for controlling the distribution of each reflected sound by the function described above is provided.

[0010]

The effect imparting apparatus of the present invention forms a plurality of continuous reflected sounds with respect to the inputted tone signal. The reflected sound imparts an effect such as resonance by the body of the stringed instrument. The envelope forming means forms an envelope, and this envelope controls the change in the volume of each reflected sound. The envelope is a time function that represents the change in size. This makes it possible to control the entire sound even if there are many reflected sounds. Further, the function generating means generates a distribution function,
The distribution of the timing of each reflected sound is controlled by this distribution function. Even if there are many reflected sounds due to this, it becomes possible to perform timing control as a whole.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An electronic musical instrument incorporating an effect imparting device according to an embodiment of the present invention will be described with reference to the drawings. In FIG. 1, a CPU 10 as a control unit includes a ROM 12, a RAM 13, a random number generator 14, a panel interface 15, a keyboard interface 17, and a sound source 18 via a bus 11.
And a DSP (digital signal processor) 20 is connected. The ROM 12 is a program and parameter table (FIG. 4) described later, and a distribution function table (FIG. 6).
Etc. are remembered. The RAM 13 stores setting data and performance data. The random number generator 14 is a circuit that generates a random number for randomizing the delay time due to reflection. The operation panel 1 on the panel interface 15
6 is connected. The operation panel 16 has a structure as shown in FIG. A keyboard 18 is connected to the keyboard interface 17. The keyboard 18 has a tone range of about 5 octaves. The sound source 19 may be any sound source capable of electronically forming a tone signal. Generally, an FM synthesis sound source, a sampling sound source (waveform memory sound source) or the like can be used. The DSP 20 is a circuit that gives various effects to the tone signal generated by the sound source 19. The type and degree of the effect to be given are input from the CPU 10. In this electronic musical instrument, the DSP 20 is used to impart a resonance effect due to the initial reflection occurring in the body of the stringed instrument. A RAM 21 is connected to the DSP 20. The RAM 21 can function as a shift register under the addressing control of the DSP 20. The tone signal to which the effect is added by the DSP 20 is output to the DAC 22. The DAC 22 converts this musical tone signal into an analog signal and outputs it to the sound system 23. The sound system amplifies this analog musical sound signal and outputs it as sound from an output device such as a speaker.

FIG. 2 is a diagram showing a partial configuration of the operation panel. The display 30 is a liquid crystal matrix display. In the figure, a screen for editing the envelope of the entire state of initial reflection is displayed on the screen. A cursor key 31 and a data key 32 are provided below the display 30. The cursor key 31 is a key for moving the cursor displayed on the display 30. The data key 32 is a key switch for increasing / decreasing the parameter displayed at the position of the cursor, and selecting forward / reverse.

Here, the resonance effect provided by the DSP 20 is a simulation of resonance caused by initial reflection on the body of a stringed instrument, the soundboard of a piano, or the like. Resonance of various musical instruments can be reproduced by changing the settings such as the magnitude and duration of the initial reflection, and it is also possible to give the musical sound a unique resonance not found in natural musical instruments.

The DSP 20 uses the RAM 21 as a shift register and can take out an output from any stage (tap: address) of the shift register. DSP2
0 takes out musical tone data from a plurality of predetermined addresses in accordance with the data given from the CPU 10, multiplies each by a predetermined coefficient, and then adds the plurality of musical sound data multiplied by this coefficient (convolution operation). The musical tone data that has been subjected to the convolution calculation is output as a resonating musical tone signal.

To simulate the resonance of a musical instrument, musical tone data is taken out from a very large number of taps, but in this electronic musical instrument, the address of each tap is comprehensively determined by a function, and the coefficient to be multiplied is comprehensively included by an envelope. I make a decision.

FIG. 3 is a diagram showing the envelope shape.
FIG. 4 is a diagram showing an envelope parameter table. The envelope is expressed as a time-amplitude function divided into an attack portion and a decay portion.

The attack portion is a function that rises linearly as a proportional function, and is expressed by y = AR · t. Here, y is the amplitude. t is a parameter indicating time (sampling clock: 1/48000)
Seconds). AR is an attack rate calculated from an attack parameter (FIG. 4).
(See (A)). Attack parameter (0 to 99)
Is a parameter indicating the time until the peak (= 1: normalized maximum value) is reached. That is, when the attack parameter is 0, 0 to 1 is obtained with one sampling clock.
When the attack parameter = 99, the sampling clock having a maximum value of 8192 (= 2 ¹³ ) gradually rises to 0-1.

The decay portion is a function that decays logarithmically from the peak to 0, and is expressed by y = DR ^(t-AT) . Here, AT is an attack time (att
(ack time: time to reach 0 → 1 in the attack portion), and DR is a decay rate calculated from the decay parameter (decay rate: see FIG. 4B). The decay parameter (0 to 99) is a parameter indicating the time until the peak decays to 0.
That is, when the decay parameter = 0, the attenuation is 1 to 0 in one sampling clock, and when the decay parameter is 99, the maximum value is 16384 (= 2 ¹⁴ ) in the sampling clock. When operating at a clock of 48 kHz, one clock is about 0.000021 seconds, 8192 clocks is about 0.17 seconds, and 16384 clocks is about 0.34 seconds.

FIGS. 5A, 5B and 5C are views for explaining the distribution of taps in the DSP 20. FIG. 6 is a diagram showing a distribution function. The tap distribution mode includes a uniform mode (MODE = 0) and a non-uniform mode (MODE = 1). In the case of the equal mode, taps are provided at substantially equal intervals as shown in FIG. 5A so that musical tone data can be taken out. However, if they are exactly equal, frequency characteristics are generated in the cycle of the extraction intervals, and the resonance sound has a feeling of frequency. Therefore, in the case of the equal mode, the address of each tap is randomized for a minute time so as not to have a specific frequency characteristic. A random number generator 14 is used for this randomization. On the other hand, in the unequal mode, the function selection parameter selec
One of the functions shown in FIG. 6 is selected by t (S = 0 to 9). This function is expressed by f (x) = x ^6-s S ≦ 5 ... (a) f (x) = x ^{(1 / (s-5))} S> 5 .... (b) . This function may be stored in the ROM as a table or may be calculated each time.

When S ≦ 5, according to the equation (A), the tap positions are dense at the beginning and become sparse as time passes (FIG. 5B).
Be of the type. Further, in the case of S> 5, according to the formula (B), the type becomes as sparse as the beginning and becomes dense later, as shown in FIG. 5C. The state of distribution differs depending on the value of S.

7 to 12 are flowcharts showing the operation of the electronic musical instrument.

FIG. 7 shows the main routine. When the power is turned on, first, the initial setting operation is executed (n1). The initial setting operation is an operation such as register reset. After this, the ON of the cursor key and data key is detected (n
2, n4). When it is detected that the cursor key is turned on, the cursor in the display is moved corresponding to the key (n3). When the data key is turned on, the numerical value of the parameter at which the cursor is located is changed (n5), and the display content of the parameter is also changed (n5).
6). After that, parameter processing is executed (n7). In addition, keyboard processing such as formation of a tone signal accompanying keyboard operation (n
8) and other processing such as master volume control are executed (n9). N2 while the electronic musical instrument is powered on
The operations from to n9 are repeatedly executed.

8 to 12 are flowcharts showing the parameter processing routine. First, FIG. 8 shows a time setting operation. A parameter for determining the envelope of the initial reflected sound is set from the setting contents (n10). The parameters set by this operation are: attack rate, a
tack time, decay rate, dec
ay time and NTAP. Where att
ack rate, attack time, deca
The y rate and the decay time are the parameters described in FIG. NTAP is data indicating the number of taps (the number of reflected sounds) extracted from the shift register. Next, attack time and decay ti
Add me and the total time of all envelopes
The image is calculated (n11). The total time is compared with NTAP (n12).

If the total time is large, to
The tal time and MAX DELAY are compared (n13). MAX DELAY is a value indicating the maximum number of shift register stages that can be set in the RAM 21. to
If the total time is longer than this, it is not possible to delay that length, so MAX D is added to the total time.
The value of ELAY is set (n14). After this mod
Judge 1/0 of e (n15) and proceed to the corresponding processing.

FIG. 9 shows the all taps setting operation. This operation is a process when NTAP is equal to or more than the total time, and is an operation of extracting musical tone data from all taps within the envelope time. The address LD (i) has i
Set the same value as. That is, one address is advanced (n21, n24). Attack time att
Up to ack time (n23), LC (i) = AR × i (AR: attack rat
The coefficient LC (i) is calculated in (e) (n21). Further, in the decay section, LC (i) = DR ^(i-AT) (DR: decay rat ⁾
e) calculates the coefficient LC (i) (n24). Of the NTAP, only the address is set for the portion protruding from the total time (n27 to n29).

In FIG. 10, NTAP is the total time.
The processing when mode = 0 (equal mode) when less than e is shown. First, Ins DLY for calculating the provisional tap interval Ins DLY is calculated by Ins DLY = (total time) / (MAX TAP). As a general rule, taps are set at this interval. However, this tap position is randomized because it does not have a specific frequency characteristic. First, i is set to 1 (n31), and a random number is fetched from the random number generator 14 (n3
2). The random number value is set in the register rnd. This random value is a value in the range of 0 or more and 1 or less (0 ≦ rnd ≦ 1). Based on this, the address LD (i) is calculated.

LD (i) ← Ins DLY × i + Ins DLY × (rnd-0.5) It is determined whether this address is an attack part or a decay part (n34). In the case of the attack part, LC (i) = AR × LD (i) (AR: attack r
ate) to calculate the coefficient LC (i) (n35). In the case of the decay section, LC (i) = DR ^{(LD (i) -AT)} (DR: decay rate, AT: attac
The coefficient LC (i) is calculated by (k time) (n36).

The above operation is repeated until i becomes NTAP (n37, n38).

FIG. 11 shows the unequal mode (mode = 1) operation. First, set 1 to i, and LD (1), L
The address and coefficient of the peak of the attack part at C (1) (=
Set 1). This is to ensure that musical tone data is taken out from the peak and the envelope becomes clear no matter how the taps are distributed.

Next, 1 is added to i (n42). A random number value is fetched from the random number generator, and a temporary address LD temp is obtained based on the distribution function (n43, n44).
Address L for which this LD temp has already been determined
It is determined whether or not it overlaps with D (j) (j = 1 to i-1) (n45 to n48). If there is the same one, this LD t
The emp is discarded and the random number value fetching (n43) is performed again. If there is no same thing, set LD temp to LD (i)
(N50). Since the address LD (i) has been determined, the coefficient LC (i) that matches this timing is determined (n51 to n53). That is, if it is an attack part, LC (i) = AR × LD (i) (AR: attack r
ate) determines the coefficient LC (i), and if it is the decay part, LC (i) = DR ^{(LD (i) -AT)} (DR: decay rate, AT: attac
Determine LC (i) by k time). n43 until i becomes NTAP
The following operation is repeated.

FIG. 12 shows a random inversion operation. In the operation so far, the reflected sound is extracted from all taps in the same phase (coefficient LC (i) is positive), but in natural musical instruments, the phase is often inverted due to reflection, so this is simulated. As described above, the sign of the tap coefficient LC (i) is randomly inverted to form a reflected sound of an opposite phase. The following operations are performed from i = 1 to NTAP. First, a random number value rnd (0≤rnd≤1) is fetched from the random number generator 14 (n61), and it is determined whether the random number value rnd is 0.5 or more or less (n62). rnd <0.
Only when it is 5, LC (i) is inverted (n63). After doing this for all LC (i), write all addresses LD (1) -LD (NTAP) into the address register of DSP 20 and write all coefficients LC (1) -LC (NTAP) to DS.
Write to the coefficient register of P20 (n66).

Although the envelope shape is divided into the attack portion and the decay portion and the attack portion is a straight line and the decay portion is a logarithmic curve in the above embodiment, the shape is not limited to this, and the decay portion may be a straight line. Good. Also, the number of segments is not limited to the attack part and the decay part, and may be larger.

Although the address is randomized by a random number, the amplitude may be randomized.

[0034]

According to the present invention, by adjusting the envelope and / or the distribution collectively, it is possible to collectively control many parameters of the early reflection sound. It is possible to efficiently control a large number of reflected sounds.

[Brief description of drawings]

FIG. 1 is a block diagram of an electronic musical instrument incorporating an effect imparting device according to an embodiment of the present invention.

FIG. 2 is a diagram showing an operation panel of the electronic musical instrument.

FIG. 3 is a diagram showing an envelope of early reflection in the electronic musical instrument.

FIG. 4 is a diagram showing a parameter table for determining the envelope.

FIG. 5 is a diagram showing a state in which the distribution of impulse response is variously changed.

FIG. 6 is a diagram showing a distribution function for determining the distribution of impulse responses.

FIG. 7 is a flowchart showing the operation of the electronic musical instrument (main routine).

FIG. 8 is a flowchart showing the operation of the electronic musical instrument (parameter processing: initial setting operation).

FIG. 9 is a flowchart showing the operation of the electronic musical instrument (parameter processing: all address setting operation).

FIG. 10 is a flowchart showing the operation of the electronic musical instrument (parameter processing: uniform mode operation).

FIG. 11 is a flowchart showing the operation of the electronic musical instrument (parameter processing: uneven mode operation).

FIG. 12 is a flowchart showing the operation of the electronic musical instrument (parameter processing: random inversion operation).

Claims (2)

[Claims] 1. An effect imparting device for forming a plurality of continuous reflected sounds with respect to an input musical tone signal, wherein the continuous form is formed in accordance with the envelope forming means for forming an envelope and the shape of the envelope formed by the envelope forming means. And a coefficient control means for controlling the magnitude of a plurality of reflected sounds. 2. An effect imparting apparatus for forming a plurality of continuous reflected sounds with respect to an input musical tone signal, the function generating means for generating a distribution function, and the distribution of each reflected sound by the function generated by the function generating means. And a distribution control means for controlling the effect.