JPH07175475A - Electronic musical instrument - Google Patents

Abstract

PURPOSE:To express the natural resonance of strings and to realize the reverberation of natural musical tones similar to the reverberation of a musical instrument by calculating the vibration mode when a damper of a piano maintains contact with a string as a resonance of two loops having two variable delay devices. CONSTITUTION:A control section instructs various kinds of control data, such as sound production channel data CHNO, sound production information KON, pitch information H and dynamic information VEL, to the sound production channels to 688 in a sound production channel group when information related to musical sounds is outputted from an input section. The sound production channel group is composed of sound production channels 601 to 688, driving waveform generating sections 701 to 788, string models 801 to 888 and adders 520 to 522. The driving waveform data are inputted to the adders of the loop section of string models 801 to 888 corresponding to pitches and the waveform data outputted from the adders are passed through the variable delay device and a multiplier and the loop to be inputted to the adders, by which resonance sounds are formed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic musical instrument having the same natural sounding effect as a natural musical instrument when playing.

[0002]

2. Description of the Related Art In recent years, the quality of synthetic musical tones of electronic musical instruments has dramatically improved due to the progress of digital technology.
Among them, an electronic musical instrument which synthesizes a musical sound by an arithmetic circuit using a delay means has been proposed as a method excellent in playing effect.

As a conventional electronic musical instrument, for example, Japanese Patent Laid-Open No.
No. 3-267999, Japanese Patent Application Laid-Open No. 4-204599, Japanese Patent Application Laid-Open No. 4-343394, or Japanese Patent Application No. 3-201427 filed by the present applicant.

A conventional electronic musical instrument will be described below. FIG. 13 is a block diagram of this conventional electronic musical instrument. In FIG. 13, 200 is an input unit, 201 is a control unit, 202 is a soundboard filter, 300 is a sound generation channel group, 301 to 388 are sound generation channels TG, 501 is an accumulator, and 553 is a multiplier.

The operation of the electronic musical instrument constructed as described above will be described below. First, the input unit 200 is instructed about the pitch of a musical tone to be output and the timing of the pronunciation. The input unit 200 includes, for example, a keyboard, a damper pedal,
Or, in general, MIDI (Musical Instrument Digital
It receives performance information input using a communication means known as "Interface" and outputs the performance information to the control unit 201.

The input unit 200 uses the position of the pressed key when the input form of the pitch instruction is a keyboard, the combination pattern of the pressed keys when the input device is a wind instrument shape, and the input pattern of a guitar shape. In this case, the pitch (so-called pitch name) of the musical tone to be output is determined by the position (fret) of the pressed string, and the pitch data is output to the control unit 201. Here, it is assumed that the input form is a keyboard consisting of 88 keys, and the pitch data 1, 2, ...
.., 87, 88, etc., and output pitch data consisting of 7 bits. Each pronunciation channel 301-388
Are pitch data 1, 2, ..., 87, 88, respectively.
It corresponds to.

Further, the input unit 200 is for pressing and releasing the key when the input form of the pronunciation instruction is a keyboard, for starting and stopping the exhalation in the case of a wind instrument shape, and for a guitar shape. In this case, as the pronunciation information KON regarding the on / off of the musical sound to be output with respect to the start / stop of the vibration of the strings, that is, as the information regarding the start / stop of the generation of the musical sound, KON = 1 when the sound is on, and KON = 1 when the sound is off. KON = 0, and strength information VEL = 0 to 1 depending on the strength of key depression or expiration.
27 (weak to strong) is output to the control unit 201 as damper information DON = 127 to 0 (a state where the pedal is depressed to a state where the pedal is not depressed) relating to the on / off state of the damper pedal.

Each tone generation channel TG301 to 388 is
For example, it is configured as shown in FIG. In FIG. 14, 1
00 is a loop, 101 is a drive waveform generator, 103 is a variable delay device, 104 is a filter, 502-503 are adders,
552 is a multiplier.

Various control data output from the control unit 201 is output in combination with the pitch data H so as to be control data effective for a specific sounding channel. When there are not the same number of tone generation channels as the pitch data H, the control unit 201 functions as a so-called channel assigner to output the tone generation channel data CHNO that specifies the tone generation channel and to output the other tone control data. It shall be used in combination.

Further, the drive waveform generator 101 can be constructed as shown in FIG. In FIG. 15, 107
Is a memory that stores drive waveform data in advance, 108 is a counter that reads the drive waveform data stored in the memory 107, and 551 is a multiplier.

In FIG. 15, when the pronunciation information KON reaches the value "1" (sound on), the counter 108 is reset and the count value (address in the memory 107) becomes the value "0". After that, at the timing of the system clock CK equal to the so-called sampling frequency, the counter 1
08 counts up the count value, and the drive waveform data stored in the address corresponding to the count value of the counter 108 is output from the memory 107. Here, the counter 108 addresses the entire address area of the memory 107 from the value “0”, and then the memory 107
When the final address is reached, the count value is held, that is, the counting operation is stopped. Such a counting operation of the counter 108 is suitable for synthesizing a transient musical sound such as a piano or a guitar, and in order to synthesize a quasi-stationary musical sound such as a violin or a wind instrument, the driving waveform data is repeated. Output it. That is, the drive waveform data corresponding to the rising portion of the musical sound is stored in the memory 10.
After being read from 7, the count operation of the counter 108 may be controlled so that the drive waveform data corresponding to the steady part of the musical sound is repeatedly read from the memory 107.
The drive waveform data output from the memory 107 is applied to the multiplier 5
In 51, the scaling is performed according to the strength by the strength information VEL, and then output from the drive waveform generating unit 101. It is assumed that the memory 107 stores the waveform data corresponding to the acceleration given to the strings for playing the guitar strings and the drive waveform given to the piano strings from the piano hammer.

The drive waveform data output from the drive waveform generator 101 is added to the feedback data (FB) from the soundboard filter 202 in the adder 502, and then added to the waveform output from the multiplier 552 in the adder 503. After being added to the data, it is input to the variable delay unit 103.
The variable delay unit 103 delays the input waveform data by a delay time corresponding to the pitch H of the musical sound to be output and outputs the delayed waveform data to the filter 104. The waveform data input to the filter 104 has its frequency spectrum changed, and the multiplier 552
Is output to the adder 503 and is then added to the newly input drive waveform data. That is, a synthetic musical tone is formed while the waveform data circulates in the loop 100 composed of the adder 503, the variable delay device 103, the filter 104, and the multiplier 552. The waveform data thus synthesized is output as the sound output channels 301 to 388 (CH output). The value G multiplied by the multiplier 552 is
It is supplied from the control unit 201 as a value G obtained by multiplying the damping multiplier S peculiar to the string forming each pitch of the piano and the dump multiplier D determined from the depression amount of the damper pedal and the pronunciation information KON. The damper multiplier D is set to D = 1 when the sounding information KON is on, and is 1 when the sounding information KON is off according to the magnitude of the depression amount of the damper pedal.
It shall change from a positive decimal less than to a value of 0.

In the loop 100 described above, the variable delay device 103 having the delay time corresponding to the pitch H forms waveform data having the basic period corresponding to the pitch H. The pitch H is 50 when the clock CK is 40 [kHz].
If 0 [Hz], the corresponding delay amount corresponds to 80 unit delays. Output pitch H and clock CK
When the delay amount determined from the above is not an integer, the decimal delay amount can be realized by linear interpolation or filter processing. Further, since the filter 104 passes through each time the waveform data circulates through the loop 100, the filter 10
By setting 4 as a low-pass filter, the higher frequency components included in the waveform data are attenuated more rapidly with the passage of time, and the higher frequency components sound disappears sequentially with the passage of time like a piano or a guitar. A musical tone that sounds like a go is formed. Further, since the multiplier 552 passes each time the waveform data circulates in the loop 100, the multiplier G or the multiplier S for determining the multiplier G is set to 1
By setting it to a decimal number less than, it is possible to prevent oscillation due to a loop and realize an attenuation system envelope suitable for pianos and guitars.

As described above, each sound generation channel 301
The waveform data (CH output) output from ˜388 are cumulatively added in the accumulator 501 by the number of all sound generation channels, and the soundboard filter 202 is provided with the characteristics of the resonator, that is, the piano soundboard or the body of the guitar. As a result, musical tone waveform data is output. Further, the output of the soundboard 202 is multiplied by a multiplier 553 by a multiplier corresponding to the depression amount of the damper pedal. This multiplier is assumed to be a value close to 0 when the damper pedal is not depressed, and close to a value 1 when the damper pedal is depressed.

In the above, the tone generation channel TG
Includes a drive waveform generator 101, but a time delay corresponding to at least the pitch to be output, such as an effect circuit which generates a sound by resonance described in Japanese Patent Laid-Open No. 63-267999. It is assumed that the loop 100 includes the variable delay device 103 having In this case, the input to the tone generation channel TG is a mixture of musical tone waveform data of a plurality of pitches.

[0016]

However, in the above-mentioned conventional structure, the resonance of the sound generation channels 301 to 388 corresponding to the pitches of 88 pianos is simply realized in accordance with the depression amount of the damper pedal. However, when the damper pedal is not depressed, there is a problem that the natural sound of the sound cannot be reproduced because the resonance sound is not generated in the sound generation channel corresponding to the pitch that is not keyed.

Further, even in the tone generation channel corresponding to the pitch of a keystroke, there is a problem in that it is not possible to reproduce the manner in which the tone color changes due to the damper hitting the string when the key is released.

Further, there is a problem that the spatial resonance of the string resonance sound cannot be reproduced. The present invention solves the above-mentioned conventional problems, and an object of the present invention is to provide an electronic musical instrument capable of realizing a natural musical sound similar to that of a musical instrument according to a performance.

According to the first aspect of the present invention, when a key of a certain pitch of a piano is tapped in a state where the damper pedal is not depressed, the strings corresponding to the pitch other than the string of the pitch that is tapped resonate while being affected by the damper. The purpose is to synthesize the states and reproduce the natural sound.

In the second and third aspects of the invention, when the keyboard of a pitch with a piano is tapped in a state where the damper pedal is not depressed, the strings corresponding to the pitches other than the string of the pitch being tapped are affected by the damper. While resonating, the purpose is to reproduce the sound with natural timbre.

A fourth object of the present invention is to realize a natural change in the tone of a tone color in accordance with the amount of depression of the damper pedal in an electronic musical instrument which can reproduce a natural tone in a state where the damper pedal is depressed and a state where the damper pedal is not depressed. To do.

A fifth aspect of the present invention realizes a state in which when the pressed keyboard is released, that is, when the key is released, the damper hits the resonating string, so that the volume decreases and the timbre changes. The purpose is to do.

An object of the sixth invention is to effectively realize the second to fifth inventions with a simple structure.

[0024]

To achieve this object, the electronic musical instrument of the first invention comprises a loop portion formed of an adder, a variable delay device and a multiplier for multiplying a predetermined multiplier. A plurality of string models, and a coefficient generation unit that outputs a multiplication coefficient G to the multiplier.

The electronic musical instrument of the second aspect of the invention has the first adder, the first variable delay device, and the first multiplier for multiplying by a predetermined multiplier.
Second multiplier for multiplying the first loop unit, the second adder, the second variable delay device, and a predetermined multiplier.
And a second loop section formed by a multiplier of the above and a plurality of string models.

The electronic musical instrument of the third aspect of the invention has the first adder, the first variable delay device and the first multiplier for multiplying by a predetermined multiplier.
Second multiplier for multiplying the first loop unit, the second adder, the second variable delay device, and a predetermined multiplier.
A second loop section formed by a second loop section, a third multiplier for multiplying the waveform data in the first loop section by a predetermined multiplier, and an output of the third multiplier by the second loop section. A third adder for adding to the third loop, a fourth multiplier for multiplying the waveform data in the second loop unit by a predetermined multiplier, and a fourth multiplier for adding the output of the fourth multiplier to the first loop unit. And a plurality of string models composed of an adder of

An electronic musical instrument according to a fourth aspect of the present invention includes a first adder, a first variable delay unit and a first multiplier, and a second adder. A second loop unit formed of a second variable delay unit and a second multiplier, a third multiplier that receives the waveform data in the first loop unit as an input, and an output of the third multiplier To the second loop unit, a fourth multiplier that receives the waveform data in the second loop unit as an input, and the output of the fourth multiplier to the first loop unit And a second adder for the second loop section
Of a plurality of string models each including a fifth multiplier for controlling the input to the adder of the first string, and a plurality of multiplication coefficients that change corresponding to the amount of depression of the damper pedal.
To a fifth multiplier, the coefficient generating section outputs the coefficient to the fifth multiplier.

An electronic musical instrument according to a fifth aspect of the present invention includes a first adder, a first variable delay device and a first multiplier, and a second adder. A second loop unit formed of a second variable delay unit and a second multiplier, a third multiplier that receives the waveform data in the first loop unit as an input, and an output of the third multiplier To the second loop unit, a fourth multiplier that receives the waveform data in the second loop unit as an input, and the output of the fourth multiplier to the first loop unit And a second adder for the second loop section
Of a plurality of string models each including a fifth multiplier for controlling an input to the adder of the string and a plurality of predetermined multiplication coefficients corresponding to a key depression. And a coefficient generation unit for changing a plurality of multiplication coefficients output to the multiplier and output to the first to fifth multipliers in response to key release.

The electronic musical instrument of the sixth invention comprises a first adder, a first multiplier which receives the output of the first adder, and a first multiplier which delays the output of the first multiplier. A variable delay unit, a second multiplier that receives the output of the first adder, a second adder that adds the output of the first variable delay unit and an output of the second multiplier, and a second multiplier A plurality of string models each including a second variable delay device that delays an adder output and a third multiplier that receives the second variable delay device output as an input and outputs a multiplication result to the first adder are provided. Have

[0030]

According to the first aspect of the present invention, the drive waveform data corresponding to the pitch H to be output is input to the adder of the loop portion of the string model corresponding to the pitch other than the pitch H in the first invention. The waveform data output from the adder passes through a variable delay unit and a multiplier, and a loop is input to the adder to form a resonance sound. Assuming that the sum of the delay amounts in the variable delay device is almost the same as the time period of the fundamental frequency of the corresponding pitch of the string model, the coefficient generator generates a multiplication coefficient whose value exceeds 0 even when the damper and the string are in contact. Since the resonance sound having an effective volume is obtained by outputting the resonance sound to the multiplier, a natural sound is synthesized.

In the second invention, the drive waveform data corresponding to the pitch H to be output corresponds to pitches other than the pitch H, and the first and second loop parts of the string model. 2 is input to the adder. The waveform data output from the first adder is input to the first adder via the first variable delay unit and the first multiplier, and the tone color of the resonance tone is formed by the first loop. Similarly, the waveform data output from the second adder is input to the second adder via the second variable delay unit and the second multiplier, and the tone color of the resonance tone is formed by the second loop. To be done. The sum of the delay amounts of the first and second variable delay devices is set to be substantially the same as the time period of the fundamental frequency of the corresponding pitch of the string model, and the ratio of the delay amounts of the first and second variable delay devices is corresponding. By keeping the same ratio of the distance from the part of the pitch string where the damper hits to the fixed ends of both ends of the string, the resonance of the string divided into two parts by the damper in the first and second loop parts Since it is formed, the state of the string that vibrates in a divided manner is reproduced by the contact with the damper.

According to a third aspect of the present invention, the drive waveform data corresponding to the pitch H to be output is the first and second loop parts of the string model corresponding to the pitch other than the pitch H. 2 is input to the adder. The waveform data output from the first adder is input to the first adder via the first variable delay unit and the first multiplier, and the tone color of the resonance tone is formed by the first loop. Similarly, the waveform data output from the second adder is input to the second adder via the second variable delay unit and the second multiplier, and the tone color of the resonance tone is formed by the second loop. To be done. The sum of the delay amounts of the first and second variable delay devices is set to be substantially the same as the time period of the fundamental frequency of the corresponding pitch of the string model, and the ratio of the delay amounts of the first and second variable delay devices is corresponding. By keeping the same ratio of the distance from the part of the pitch string where the damper hits to the fixed ends of both ends of the string, the resonance of the string divided into two parts by the damper in the first and second loop parts It is formed. Further, since the damper and the string do not form a completely fixed end, the waveform data formed by the first and second loop portions are the third multiplier and the third adder or the fourth multiplier, respectively. Since it is input to the second and first loop units via the fourth adder, it is possible to reproduce a state in which two string vibrations that are divided and vibrated by the contact of the damper interfere with each other.

According to a fourth aspect of the invention, when the damper pedal is not depressed, the coefficient generating section outputs a plurality of corresponding multiplication coefficients to the first to fifth multipliers.
Due to the resonance sound formed separately by the first and second loop portions,
The appearance of a string that vibrates separately due to the contact of the damper is reproduced. Further, as the amount of depression of the damper pedal gradually increases, the coefficient generation unit gradually decreases the multiplication coefficients of the first, second and fifth multipliers and gradually increases the multiplication coefficients of the third and fourth multipliers. By doing, the first
Of the first and second variable delay units causes resonance due to one loop portion formed of the adder, the first and second variable delay units, and the third and fourth multiplier units. By making the sum of the quantities equal to the time period of the fundamental frequency of the corresponding pitch of the string model, the resonance of the time period equal to the fundamental frequency is realized.

According to a fifth aspect of the present invention, the coefficient generating section outputs a predetermined multiplication coefficient to the first to fifth multipliers of the string model corresponding to the pitch of the key being pressed, and the sound to be output substantially. One loop corresponding to the high pitch is formed, and pitches to be output are synthesized based on the input drive waveform data. When the key is released, the coefficient generator gradually increases the multiplication coefficients of the first, second, and fifth multipliers with time, and gradually increases the multiplication coefficients of the third and fourth multipliers with time. By making it small, the first and second loop portions are effectively made effective, and the divided vibration of the strings caused by the damper coming into contact with the strings is synthesized, and the timbre change due to the contact of the damper is realized. Become.

Further, in the sixth invention, when the damper and the string are not in contact with each other, the multiplication coefficient of the first multiplier of the string model is set to 1 and the multiplication coefficient of the second multiplier is set to 0. The loop portion formed by the first adder, the first multiplier, the first variable delay device, the second adder, the second variable delay device, and the third multiplier by placing the string Resonant sounds of the corresponding pitch of the model are synthesized. When the damper and the string are in close contact with each other, the multiplication coefficient of the first multiplier of the string model is set to 0, and the multiplication coefficient of the second multiplier is set to 1 to make the first addition. From the contact portion of the string divided into two by the contact of the damper, by the loop portion formed by the damper, the second multiplier, the second adder, the second variable delay device, and the third multiplier. The resonance sound of the pitch corresponding to any one string up to the fixed end of the string is synthesized. Further, by setting one of the multiplication coefficients of the first and second multipliers to be large and the other to be small according to the degree of contact between the damper and the strings, it is possible to continuously combine the materials according to the contact state.

[0036]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the arrangement of the electronic musical instrument according to the first embodiment. In FIG. 1, 11 is a coefficient generator, 203 is a soundboard filter, 400 is a sound channel group, and 510 is an adder. In addition, 200 is an input part,
Reference numeral 201 is a control unit, 202 is a soundboard filter, and 553 is a multiplier, which have the same configurations as those of the conventional example. The soundboard filter 202 and the soundboard filter 203 are the same.

The operation of the electronic musical instrument of the first embodiment constructed as above will be described below.

When the information about the musical sound to be output is output from the input section 200, the control section 201 performs the sound generation channels in the sound generation channel group 400 (see FIG.
601 to 688), pronunciation channel data CHNO, pronunciation information KON, pitch information H, strength information V
Instruct various control data such as EL.

The tone generation channel group 400 is constructed as shown in FIG. In FIG. 2, 601 to 688 are sound generation channels, 701 to 788 are drive waveform generators, 801 to 8
Reference numeral 88 is a string model, and reference numerals 520 to 522 are adders. The drive waveform generators 701 to 788 are the same as the drive waveform generator 101 of the conventional example.

The string models 801 to 888 are constructed as shown in FIG. In FIG. 3, 411 and 412 are variable delay devices, 421 and 422 are filters, and 555 to 564.
Is a multiplier and 580 to 585 are adders. The filters 421 to 422 are the same as the filter 104 of the conventional example.

First, the relationship between the damper and the strings will be described with reference to FIG. In FIG. 4, 900 is a soundboard, 901 is a string, 902 is a piece, 903 is a fixed end, 904.
Is a damper, and 905 is a damper felt forming a part of the damper 904.

The damper 904 is normally in contact with the string, and the damper felt 905 functions to damp the vibration of the string 901.

When a keyboard (not shown) of a piano is tapped, the damper 904 of the string 901 corresponding to the key is mechanically lifted, and the damper felt 905 is separated from the string 901. When the piano hammer collides with the position of the string 901 shown in FIG.
The part indicated by starts to vibrate, and the vibration of the string 901 is the string 901.
From the soundboard 900 directly or through the bridge 902 as a sound wave. Further, when the finger is released from the keyboard (key release), the lifted state of the damper 904 is canceled, and as a result, the damper 904 descends due to its own weight, and the damper felt 905 and the string 901 come into contact again. Then, the vibration of the string 901 is rapidly attenuated and the sound is silenced. The fixed end 90 of the string 901
It is assumed that the fundamental frequency of the vibration of the portion indicated by L between 3 and the piece 902 corresponds to the pitch of the output musical sound.

When the damper pedal (not shown) is depressed, the dampers 904 of all the strings are mechanically lifted by a height corresponding to the depression amount of the damper pedal, and the string vibrating at this point, or After that, the vibration of the string that is struck and vibrated causes the other strings that were not vibrating to vibrate, and these vibrations are also transmitted through the piano housing (not shown) or the sound board 900 or the piece 902. Propagating to 901, the part indicated by L of the string 901 vibrates, and the vibration of the string 901 is emitted in the same manner as when the key is pressed. When the damper pedal is released, the damper 904 comes into contact with the string 901 again.
The vibration of 01 is rapidly attenuated and the sound is silenced.

Regarding the vertical movement of the damper 904 depending on the operation state of the damper pedal and the keyboard, the keyboard is prioritized when a key is pressed (key pressed), and the damper pedal is prioritized when the key is released.

When the damper pedal is not depressed, that is, when the damper felt 905 and the string 901 are in contact with each other, when the keyboard corresponding to another string is hit, the piano hammer hits the other string. The striking vibrations and the vibrations of other strings propagate to the strings 901 through the piano housing, the damper 904, the damper felt 905, the soundboard 900 and the piece 902, so that the part indicated by L1 or L2 of the strings 901 vibrates. It will be. The level of these vibrations usually indicates a level of about -40 to -50 dB (decibels) than the vibration of a string that is tapped. Note that, in reality, since the vibration is in a state where the damper felt 905 is still in contact with the damper felt 905, the damping of the vibration by the damper felt 905 is large, and the vibrations of the portions L1 and L2 are the same as the damper felt 905 and the strings 901. Propagate through the contact site of. Further, since vibration is suppressed in the string 901 of the part L3 ′ where the damper felt 905 and the string 901 are in contact with each other,
Substantially, the portions L1 ′ and L2 ′ of the string 901 vibrate as effective portions for vibration.

The operation of the tone generation channel instructed to turn on KON when the tone data H is instructed to be sounded by keystroke will be described below.

In accordance with various control data output from the controller 201, the drive waveform generators 701 to 788 of the tone generation channels 601 to 688 form the instructed pitch data H by the same operation as the conventional example. The drive waveform data of is output. The drive waveform data output from the drive waveform generation unit i is input to the adder 580 of the corresponding string model i (any one of 801 to 888).

Corresponding to the control data output from the control unit 201, the coefficient generating unit 11 operates in a manner to be described later, and a specific tone generation channel CHNO, K here.
The multiplication coefficients DP1 to DP4 and LP1 to LP4 are output to the tone generation channel i whose ON is turned on. The attenuation multiplier peculiar to the string is S (= 0.9). DP2 for simplicity
And DP4, LP1 and LP2 have a value of 0, and DP1 has a value of 0.0.
1, the value of DP3 is 1, and the values of LP3 and LP4 are the square roots of 0.994, the string model of FIG. 3 is substantially the adder 584, the variable delay unit 411, the multiplier 563, and the adder 5
83, the variable delay unit 412, and the filter 422 form a loop, and if the delay amount by the variable delay units 411 and 412 is the delay amount corresponding to the same pitch data H as the variable delay unit 103, the conventional string model 100. It has almost the same structure as. If the multiplication coefficient LP3 is considered to be the value 1, a circuit equivalent to the string model 100 is obtained if the value of LP4 is the coefficient G output from the control unit 201. Therefore, the value of the multiplication coefficient LP4 is output from the control unit 201. String model 1 if coefficient G
Since the circuit is equivalent to 00, if the multiplication coefficient LP4 is a positive number less than or equal to 1, the drive waveform data input to the adder 580 is synthesized with the pitch tone to be output by the same operation as in the conventional example. It

Next, the damper pedal is depressed and no sound is instructed (KON = OFF).
Describe the operation of the pronunciation channel.

When KON is off, the drive waveform data is not output from the drive waveform generator i (when 0 value is output).
Therefore, the drive waveform data output from another drive waveform generating unit is input to the string model i corresponding to the drive waveform generating unit i via the adder 520. The soundboard filter 20
Feedback from 2,203, ie, multiplier 5
The 53 output is also added by the adder 520. The output of the adder 520 is input to the multiplier 555 of the string model i.

Corresponding to the control data output from the control unit 201, the coefficient generation unit 11 performs multiplication operations DP1 to DP4 and LP1 for all the tone generation channels 601 to 688 by an operation described later. ~ Output LP4. The attenuation multiplier peculiar to the string is S (= 0.9). For simplification, DP2, DP4, LP1 and LP2 are set to the value 0, DP1
Is 0.01, DP3 is 1, and LP3 and LP4 are S
Assuming the square root of 0.949, the string model of FIG. 3 is substantially a loop formed by the adder 584, the variable delay unit 411, the multiplier 563, the adder 583, the variable delay unit 412, and the filter 422. Variable delay device 411, 412
Assuming that the delay amount due to the variable delay device 103 corresponds to the same pitch data H as the variable delay device 103, the delay amount is substantially the same as that of the conventional string model 100. Further, the multiplication coefficient LP3 is set to the value 1, and LP4
Note that if the value of is the coefficient G output from the control unit 201, a circuit equivalent to the string model 100 is obtained.
The waveform data input to the multiplier 555 is added via the adder 580 by the same operation as in the conventional example, and the pitch tone corresponding to each sound generation channel is synthesized as a sound effect.

The tone generation channel in which KON is turned on when the damper pedal is depressed and the tone generation channel in which KON is turned off and the damper pedal is depressed have the multiplication coefficient supplied from the coefficient generator 11 as described above. Since they are the same, the same operation is performed on the waveform data output from the adder 580. For the sounding channel i being sounded, the drive waveform data output from the drive waveform generating unit i is directly input to the adder 580 of the string model i and the string model via the adder 520 and the multiplier 555. Although there are two paths, i.e., the path input to the adder 580 of i, the value of the multiplication coefficient DP1 is 0.0.
Since it is about 1, the component of the drive waveform data generated by the drive waveform generation unit i output from the adder 580 of the string model i is 1.01 times, but there is no problem in sound quality. Further, in order to prevent such a situation, the drive waveform data output from the drive waveform generation unit i of the tone generation channel i whose KON is turned on is prohibited from being input to the adder 520, or the coefficient generation unit 11 is used. It suffices to appropriately adjust the value of the multiplication coefficient supplied from.

Next, the operation of the sound generation channel in which the damper pedal is not depressed and sound generation is not instructed (KON = OFF) will be described.

When KON is off, drive waveform data is not output from the drive waveform generating section i (when 0 value is output).
Therefore, the drive waveform data output from another drive waveform generating unit is input to the string model i corresponding to the drive waveform generating unit i via the adder 520. The soundboard filter 20
Feedback from 2,203, ie, multiplier 5
The 53 output is also added by the adder 520. The output of the adder 520 is input to the multiplier 555 of the string model i.

Corresponding to the control data output from the control unit 201, the coefficient generation unit 11 operates by the operation described later, and when the KON of the tone generation channel i is off, the multiplication coefficients DP1 to DP4 and LP1 to LP1. Output LP4. Let the value S be the damping multiplier specific to the string. For simplicity, DP1 has a value of 0.002, DP2 has a value of 1, and DP3 and DP4 have a value of 0.
5, the value of LP1 and LP2 is smaller than the square root of S of 0.949, 0.6, and the value of LP3 and LP4 is 0.2, the string model of FIG. Becomes That is, the first loop includes the adder 581, the variable delay unit 411, the filter 421, and the multiplier 561, and the second loop includes the adder 582, the variable delay unit 412, the filter 422, and the multiplier 562. The first and second loops are connected to each other by a multiplier 563 and an adder 583, or a multiplier 564 and an adder 584, respectively. Here, by setting the delay amounts of the variable delay device 411 and the variable delay device 412 respectively as the time delay amounts corresponding to the fundamental frequencies of the vibrations of the parts indicated by L1 and L2 of the string 901 in FIG. 4, the damper felt 905 9 in a state where the string 901 and the string 901 are in contact with each other
The vibrations of L1 and L2 of 01 will be simulated by the first and second loops. The multiplication in the multiplier 561 and the multiplier 562 is performed on the reflection coefficient of the L1 vibration of the string 901 on the bridge 902 and the damper felt 905, or the reflection coefficient of the L2 vibration of the string 901 on the fixed end 903 and the damper felt 905. Corresponding LP1 and LP
It is multiplied by the value of 2, but the damper felt 904 has a small reflectance and absorbs the vibration of the string 901.
The value of 2 is a relatively small value of 0.6. Further, the vibration of the first and second loops also vibrates the damper 904 itself via the damper felt 905, so that the vibrations in the respective loops are not only reflected in the loops but also transmitted to the respective loops. Will be done. In this transmission of vibrations, the transmission from the first loop to the second loop is performed by the multiplier 563 and the adder 583, and the transmission from the second loop to the first loop is performed by the multiplier 564. Adder 5
It will be simulated by 84 and. Further, the time delay amount of the variable delay devices 411 and 412 is set to the string 901 of FIG.
By making the effective lengths L1 ′ and L2 ′ of the vibration of the same equal to the actual piano, a resonance state closer to an actual piano is simulated.

The multiplication coefficient output from the coefficient generator 11 corresponds to the ON / OFF signal KON for sounding and the OFF / ON (contact / non-contact) DMP signal of the damper 402 determined by the signal DON for turning the damper pedal on / off. A value like this is output. The DMP signal is KON and DO.
It is determined as the logical product of N. In addition, here, each multiplication coefficient is assumed to take two values, but it is better to change it while interpolating with time so as to smoothly change between the two values so as to prevent noise generation. Good.

As described above, the sound generation channels 601-
Multipliers 559-56 of 688 string models 801-888
The waveform data output from 0 is for each string model 801 to 8
After being added by the adder 585 of 88, the multipliers 557 and 558 multiply the multiplication coefficients GL and GR by
It is output as the L output (left output) and the R output (right output) of the string model. Here, the multiplication coefficients GL and GR
Is a value peculiar to each sound generation channel 601 to 688, and is a weighting factor for determining the distribution of stereo to the left and right two channels, and is usually a value from 0 to 1 and the square of GL and GR. The sum is determined to be equal to 1.

The L and R outputs output from the tone generation channels 601 to 688 are added by adders 521 and 522, respectively.
After the sum of the L output and the sum of the R output are calculated in (1), they are input to the soundboard filters 202 and 203, respectively. Here, the soundboard filters 202 and 203 output the musical tone waveform data L and the musical tone waveform data R by the same operation as in the conventional example. The tone waveform data L and R obtained as described above are generated by the sound generation channels 601 to 601.
The two soundboard filters 202 and 203 having the same characteristics perform left-right independent filtering on the waveform data to which the stereo signal distribution unique to 688 is respectively performed, so that the influence of the stereo signal distribution is left as it is. Musical sound waveform data L showing good stereo characteristics,
R will be obtained.

Further, the tone waveform data L and R output from the soundboard filters 202 and 203 are added by the adder 510, and then multiplied by a predetermined feedback multiplier in the multiplier 553 and output as feedback data FB. Then, it is input to the adder 520. Here, the feedback multiplier has a function of determining the mixing ratio between the feedback data FB in the adder 520 and the drive waveform data output from the drive waveform generators 701 to 788, and this value is usually 0. About 0.002 to 0.01 is preferable.

As described above, each string model 801-8
The stereo outputs of the sound generation channels 601 to 688, which are distributed to the L output and the R output at 88, are independently filtered by the soundboard filters 202 and 203 and output as the musical tone waveform data L and R.
By adding at 0, the tone generation channel 601 is output as monaural waveform data via the multiplier 553 and the adder 520.
To 688, the stereo-distributed waveform data output from each sound generation channel 601 to 688 can give a uniform influence to each sound generation channel 601 to 688 regardless of the stereo distribution amount. it can.

As described above, according to the first embodiment, the damper 904 is in contact with the strings, and one string having two vibrating parts is connected to the adder and the variable delay device, respectively. Since the modeling is performed by two loops formed by the filter and the other strings and the vibrations from the sound board are input from the adder 520, a state in which the two loops independently resonate, and a multiplier are also provided. The states in which they interfere with each other are combined via 563 and 564, and a resonance tone having a natural tone color can be obtained.

The second embodiment of the present invention will be described below. The second embodiment has the same configuration as the first embodiment except for the coefficient generator 11. The coefficient generator 11 is configured as shown in FIG. In FIG. 6, 401 to 408 are coefficient memories, 410 is an address generator, 450 is a multiplier, 4
Reference numeral 51 is an adder. Similar to the conventional example, the control unit 20
1 outputs an on / off signal KON of the keyboard and a depression signal D that takes a value from 0 to 1 corresponding to the depression amount of the damper pedal. The coefficient generator 11 applies multiplication coefficients LP1 to LP4, DP1 to the tone generation channels TG601 to 688.
~ DP4 is output by the operation described below.

The damper pedal depression signal D is supplied to the adder 4
At 51, it is converted into a DINV signal by the calculation of (1-D). In the multiplier 450, the keyboard on / off signal KON and DIN
The V signal is multiplied and the DMP signal is output. The address generator 410 outputs address values 0 to 127 by scaling the value of DMP. The coefficient memories 401 to 408 have addresses 0 to 127.
Since the multiplication coefficient is stored in the address generator 410,
The multipliers LP1 to LP4 and DP1 to DP4 as shown in FIG. 7 are output according to the address value output from the.

The multiplication coefficient output from the coefficient generator 11 is supplied to each of the tone generation channels TG601 to 688, and the musical tone is synthesized by the same operation as in the first embodiment.

As in the case of the first embodiment, in order to smooth the change of the binary values (0 and 1) that the keyboard on / off signal KON can take, the binary value is interpolated with the passage of time when changing. It is better to do so.

As described above, according to the second embodiment, according to the value of the DMP signal corresponding to the contact / non-contact of the damper 904 of the piano, the adder 584 and the variable delay unit 411 are connected in the non-contact state. A loop formed by the multiplier 563, the adder 583, the variable delay unit 412, and the filter 422 causes resonance. Further, at the time of contact, the adder 581
A first loop formed of a variable delay device 411, a filter 421 and a multiplier, an adder 582 and a variable delay device 4
12, the second loop formed by the filter 422 and the multiplier 562 resonates. The resonance of the state in which the state of contact of the damper 904 with the string 901 changes according to the amount of depression D of the damper pedal is calculated as an intermediate state between the above two states, so that a natural resonance is produced according to the amount of depression of the damper pedal. Change can be realized.

The third embodiment will be described below.
The third embodiment is also constructed in the same manner as the second embodiment, but the input section to the address generator 410 is constructed as shown in FIG. In FIG. 8, 471 is a counter, 472 is a key-off memory, and 452 is a multiplier.

When the KON signal is turned on (value 0), the counter 471 is reset and outputs the address value 0. The key-off memory 472 outputs the value 0 stored at address 0. In the multiplier 452, the KON signal value 0 is multiplied by the output value 0 of the key-off memory 472 and the value of the signal K becomes 0, and the multiplication coefficient when the KON signal is ON is output as in the second embodiment. To be done. KON signal is off (value 1)
When it changes to, the counter 471 increments the count value according to the clock CK. When the count-up from 0 to 127 is completed, the count value 127 is held until the next ON of the KON signal is input. The key-off memory 472 uses the output value of the counter 471 as an address, and stores the value 0 to 127.
Outputs a decimal value that gradually increases from 1 to 1. Since the KON signal has the value 1, the output value of the key-off memory 472 is directly output as the value of the signal K.

As described above, the damper 904 of the piano
When the KON signal changes from on to off, the signal K obtained by transforming the KON signal takes a value between on (value 0) and off (value 1). It changes sequentially with the passage of the signal K over time. The multiplication coefficients shown in FIG. 9 corresponding to the DMP signal which changes with the passage of time are set to the tone generation channels TG601 to TG68.
Supply every 8th. At the time of key-off from each sound generation channel TG, the resonance state changes with the passage of a predetermined time.

According to the third embodiment, the damper 904 contacts the vibrating string 901 when the keyboard is released and the string 901 is vibrated.
A change in the resonance state until the vibration of the damper is attenuated is represented by the contact D between the string 901 and the damper 904 at the time of key-off (key release)
Since the MP signal is changed gradually with the passage of time, it is possible to realize a natural tone color change when the key is released.

The fourth embodiment will be described. The fourth embodiment is constructed similarly to the third embodiment, but the string model is constructed as shown in FIG. In FIG. 10, 480 and 481 are adders, 485 to 487 are multipliers, and 490 and 4
Reference numeral 91 is a variable delay device, and 492 is the same filter as 421.

When the value of the DMP signal is 0, that is, KON
The operation when the signal is on or the damper pedal signal DON is on will be described below.

The waveform data output from the adder 580 circulates in a loop formed by the adder 480, the multiplier 485, the variable delay 490, the adder 481, the variable delay 491, the filter 492, and the multiplier 487. To do. At this time, as for the multiplication coefficient, DP5 has a value of 1, DP6 has a value of 0, and LP5 has a value of
Is a string-specific attenuation multiplier S of 0.9. Further, the variable delay devices 490 and 491 are assumed to have a time delay amount corresponding to the pitch of each sound generation channel 601 to 688, that is, the L vibration period of the string 901. As described above,
A loop similar to the conventional example can be formed.

When the DMP signal has the value 1, that is, when the damper 904 is in contact with the string 901, the multiplication coefficient DP5 has the value 0, DP6 has the value 1, and LP5 has the string-specific attenuation multiplier S of the value 0.9. And Further, it is assumed that the time delay amounts of the variable delay devices 490 and 491 correspond to the vibration periods of L2 and L1 of the string 901, respectively. The L1 portion of the string 901 is more likely to propagate the vibrations from the bridge 902 than the L2 portion, and resonates largely, so that the L1 portion is in a resonance state. Since the waveform data output from the adder 580 circulates in a loop formed by the adder 480, the multiplier 486, the adder 481, the variable delay device 491, the filter 492, and the multiplier 487, the L1 portion of the string 901 A resonance state corresponding to is created.

By controlling the multiplication coefficients DP5 and DP6 as shown in FIG. 11, the resonance state corresponding to the case where the DMP signal shows an intermediate value between the values 1 and 0 is calculated.

As described above, according to the fourth embodiment, the state in which the damper 904 strongly resonates when in contact with the string 901 causes the adder 480, the multiplier 486, the adder 481 and the variable delay device 491 to operate. Since the calculation is performed by circulating the waveform data in the loop formed by the filter 492 and the multiplier 487, it is possible to configure a string model capable of calculating the resonance state with a simple configuration.

As described above, the effects of the first to fourth embodiments are independent of each other, and it is obvious that a synergistic effect can be obtained by arbitrarily combining these.

The fifth embodiment will be described below.
The fifth embodiment is constructed as shown in FIG. Further, the sound generation channels of the sound generation channel group 400 are configured as shown in FIG. 14 similarly to the conventional example. The coefficient generator 11 is configured as shown in FIG.

In FIG. 12, reference numeral 409 is a coefficient memory. The address generator 410 is the same as in the second or third embodiment. In the coefficient memory 409, FIG.
It is assumed that the multiplication coefficient G supplied to the multiplier 552 is stored.

DM indicating the contact state between the damper 904 and the string 901 obtained as in the second or third embodiment.
According to the P signal, the address generator 410 causes the address 0
˜127, and the coefficient memory 409 outputs the multiplication coefficient G according to the input address. Coefficient memory 409
The multiplication coefficient G stored in is stored at address 0 with a value of 0.
1, the attenuation multiplier S peculiar to the string at address 127 (value 0.9)
Is stored, and the value 0.1 is stored in addresses 1 to 126.
It is assumed that the value that gradually increases from the value to 0.9 is stored.

When the damper pedal is not operated and the keyboard is not operated, that is, when the DMP signal is 1, the multiplication coefficient G (value 0.1) is supplied to the multipliers 552 of the sound generation channels 301 to 388. It The multiplier 552 multiplies the waveform data circulating in the string model 100 by the multiplication coefficient G (value 0.1) in the same manner as in the conventional example, so that a resonance sound of silence is formed.

As described above, according to the fifth embodiment, the resonance sound can be synthesized even when the damper 904 is in contact with the string 901.

[0085]

Since the vibration mode when the damper of the piano is in contact with the string is calculated as the resonance of the loop having two variable delay devices, the natural resonance state of the string is realized. You can

Further, since the resonance sound is formed even when the damper of the piano is in contact with the strings, the resonance similar to that of the piano can be realized.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an electronic musical instrument according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an internal configuration of a sound generation channel in the first embodiment of the present invention.

FIG. 3 is a block diagram showing an internal configuration of the string model in FIG.

[Fig. 4] Structural diagram around the strings of a piano

5 is a timing diagram showing the relationship between each multiplication coefficient value and control signal in FIG.

FIG. 6 is a block diagram showing an internal configuration of a coefficient generator in the second embodiment of the present invention.

FIG. 7 is a timing chart showing a relationship between each multiplication coefficient value and a control signal in the second embodiment.

FIG. 8 is a block diagram showing an internal configuration of a coefficient generator in a third embodiment of the present invention.

FIG. 9 is a timing chart showing a relationship between each multiplication coefficient value and a control signal in the third embodiment.

FIG. 10 is a block diagram showing an internal structure of a string model in a fourth embodiment of the present invention.

FIG. 11 is a timing chart showing a relationship between each multiplication coefficient value and a control signal in the fourth embodiment.

FIG. 12 is a block diagram showing an internal configuration of a coefficient generator in a fifth embodiment of the present invention.

FIG. 13 is a block diagram showing a configuration of a conventional electronic musical instrument.

14 is a block diagram showing the internal structure of the sound generation channel in FIG.

FIG. 15 is a block diagram showing an internal configuration of a drive waveform generator in FIG.

[Explanation of symbols]

11, 701 to 788 Coefficient generation unit 100, 801 to 888 String model 101 Drive waveform generation unit 103, 411, 412, 490, 491 Variable delay unit 104, 421, 422, 492 Filter 107 Memory 108, 471 Counter 200 Input unit 201 Control unit 202,203 Soundboard filter 300,400 Sound generation channel group 301-388 Sound generation channel 401-409 Coefficient memory 410 Address generator 450,452,485-487,551-564 Multiplier 451,480-481,501-522 , 580-5
85 Adder 472 Key-off memory 601 to 688 Sounding channel 900 Soundboard 901 String 902 Piece 903 Fixed end 904 Damper 905 Damper felt

Claims (16)

[Claims] 1. A plurality of string models having a loop portion formed of an adder, a variable delay unit, and a multiplier, and a multiplication coefficient that exceeds 0 in the multiplier when the damper pedal and the keyboard are not operated. An electronic musical instrument including a coefficient generator that outputs G. 2. The electronic musical instrument according to claim 1, further comprising a coefficient generator that outputs a plurality of multiplication coefficients that change in accordance with the amount of depression of the damper pedal to the multiplier of each string model. 3. A plurality of multiplication coefficients which are output to a multiplier of each string model in response to key depression and are output to a multiplier of each string model in response to key release. The electronic musical instrument according to claim 1, further comprising a coefficient generation unit that changes the time with. 4. A first and a second adder to which waveform data is input, a first multiplier and a first variable delay device, and a first multiplier for multiplying a predetermined multiplier. And a second loop unit formed of a second adder, a second variable delay device, and a second multiplier that multiplies a predetermined multiplier. An electronic musical instrument having a plurality of electronic musical instruments. 5. Formed from first and second adders to which waveform data is input, said first adder, a first variable delay device and a first multiplier for multiplying a predetermined multiplier. A first loop unit, a second adder, a second variable delay unit, and a second multiplier that multiplies a predetermined multiplier, and A third multiplier that multiplies the waveform data of # 1 by a predetermined multiplier, a third adder that adds the output of the third multiplier to the second loop unit, and a waveform in the second loop unit. An electron having a plurality of string models including a fourth multiplier that multiplies data by a predetermined multiplier and a fourth adder that adds the output of the fourth multiplier to the first loop unit Musical instrument. 6. A first loop unit formed of first and second adders to which waveform data is input, and the first adder, the first variable delay unit and the first multiplier. A second loop unit formed of the second adder, the second variable delay unit, and the second multiplier; and a third multiplication that receives the waveform data in the first loop unit as an input. A third adder for adding the output of the third multiplier to the second loop section, a fourth multiplier for receiving the waveform data in the second loop section as an input, A fourth adder for adding the output of the fourth multiplier to the first loop section, and a fifth multiplier for controlling the input to the second adder of the second loop section. A plurality of string models to be used, and a plurality of multiplication coefficients that change corresponding to the amount of depression of the damper pedal. Electronic musical instrument having the configuration of a coefficient generator for outputting to the multiplier. 7. A first loop unit formed of first and second adders to which waveform data is input, and the first adder, the first variable delay unit, and the first multiplier. A second loop unit formed of the second adder, the second variable delay unit, and the second multiplier; and a third multiplication that receives the waveform data in the first loop unit as an input. A third adder for adding the output of the third multiplier to a second loop unit, a fourth multiplier for receiving the waveform data in the second loop unit as an input, and the fourth And a fifth adder for controlling the input to the second adder of the second loop unit, and a fourth adder for adding the output of the multiplier of No. 2 to the first loop unit. A plurality of string models, and a plurality of predetermined multiplication coefficients corresponding to key depression, to the first to fifth multipliers of each string model, and Electronic musical instrument having the configuration of a coefficient generator for varying the plurality of multiplication factor time output from the first corresponds to the fifth multiplier to key-release. 8. A first adder to which waveform data is input, a first multiplier to which the output of the first adder is input, and a first variable that delays the output of the first multiplier. A delay device, a second multiplier that receives the output of the first adder, a second adder that adds the output of the first variable delay device and the output of the second multiplier, and It comprises a second variable delay device that delays the output of the second adder, and a third multiplier that receives the output of the second variable delay device as an input and outputs a multiplication result to the first adder. An electronic musical instrument that has multiple string models. 9. The electronic musical instrument according to claim 8, further comprising a coefficient generator that outputs a plurality of multiplication coefficients that change in accordance with the amount of depression of the damper pedal to the first to third multipliers of each string model. 10. A plurality of predetermined multiplication coefficients corresponding to key depression are output to first to third multipliers of each string model, and first to third multiplications corresponding to key release. 9. The electronic musical instrument according to claim 8, further comprising a coefficient generation unit that temporally changes a plurality of multiplication coefficients output to the instrument. 11. The sum of the time delay amounts of the first and second variable delay devices is substantially equal to the time delay amount corresponding to the fundamental frequency of the corresponding pitch of the string model.
And electronic musical instruments described in 5, 6, 7, 8, 9 and 10. 12. The sum of the time delay amounts of the first and second variable delay units is smaller than the time delay amount corresponding to the fundamental frequency of the corresponding pitch of the string model.
And electronic musical instruments described in 5, 6, 7, 8, 9 and 10. 13. The ratio of the time delay amounts of the first and second variable delay devices is equal to the ratio of two parts of a string obtained by dividing a piano string by a damper into two parts. And electronic musical instruments described in 5, 6, 7, 8, 9 and 10. 14. The scope of claim 5, 6, 7 and 8 having a coefficient generator for outputting a multiplication coefficient according to a numerical value corresponding to a contact state between a damper and a string, which is calculated from a damper pedal and a key press / release of a keyboard. Electronic musical instrument described. 15. The stereo musical tone data is calculated by inputting the waveform data in which the output of the string model is distributed to the stereo to the two soundboard filters corresponding to each stereo LR. The electronic musical instrument according to 14 above. 16. The electronic musical instrument according to claim 16, wherein the outputs of the two soundboard filters are input to each string model.