JP2794730B2 - Electronic musical instrument - Google Patents

Description

Description: BACKGROUND OF THE INVENTION (a) Field of the Invention The present invention relates to a method for controlling a musical tone of an electronic musical instrument that generates a musical tone electronically, and more particularly to a method for controlling a pitch.

(B) Conventional technology Various electronic musical instruments that generate musical tones electronically have been put to practical use at present. In addition to conventional keyboards (keyboards), wind instruments and string instruments (guitars) Has also been put to practical use. These electronic musical instruments have various operation units (key switches and the like), and these operation units are operated (played).
Thus, the musical tone generated by obtaining the performance information is controlled. For example, in the case of a keyboard-type electronic musical instrument, 5
A key-on sensor that has a keyboard of about 7.5 octaves (each key corresponds to the pitch of each semitone (C, C #, D, E ♭ ……)), and detects on / off of each key , An initial touch sensor for detecting keying strength (initial touch), an after touch sensor for detecting key pressing strength (after touch) during key-on, and the like.
Furthermore, in addition to this keyboard, it also has a pedal and a wheel-type operation unit. On the other hand, in the case of a wind instrument-type electronic musical instrument, a key system and a blowing port similar to a woodwind instrument are provided as operation units, and a sensor for detecting the operation state of each key and a breath sensor and a lead for detecting the strength of breath to be blown are provided. It has a lip sensor for detecting the applied pressure.
Various tone control parameters such as pitch, pitch displacement, level (volume), waveform, vibrato and the like are determined based on the performance information obtained from such an operation unit, and the tone generator is caused to sound by using the tone control parameters. A musical tone with a rich expression can be generated. In an electronic musical instrument having such a configuration, it may be required to further enhance the playing effect and give a subtle expression to the musical tone. In such a case,
Conventionally, one tone control parameter is controlled by a plurality of pieces of performance information.

(C) Problems to be Solved by the Invention However, in a conventional electronic musical instrument, when one musical tone control parameter is determined based on a plurality of pieces of performance information, the control amount is individually determined based on each piece of performance information. Then, the tone control parameters are determined by adding or multiplying them. For example, pitch displacement (displacement of a minute frequency (pitch) at the same pitch: so-called choking), which is one of the performance techniques, is an technique often used during performance because it has the effect of increasing the tension of a song. However, the pitch displacement parameter (pitch parameter) is determined based on performance information such as initial touch, after touch, breath, and pitch bend wheel. Conventionally, the pitch parameter is determined by adding (multiplying) a control amount obtained from the initial touch performance information, a control amount obtained from the after touch performance information, and the like. However, if the pitch is to be controlled in a complicated manner using a large amount of performance information, the above-described parameter determination method requires a control amount to be individually obtained for each piece of performance information. In addition to the drawbacks, even when the plurality of obtained control amounts are biased to either the pitch rise or the pitch fall, this cannot be suppressed, and there is a danger that the pitch will rise and fall significantly and become undesirable as a musical tone. there were.

Further, when trying to express a nuance close to that of a natural musical instrument with an electronic musical instrument, it is necessary to control various musical information in a complex manner to control musical tones. When trying to perform the program processing by a general algorithm, there is a disadvantage that it takes a long time to perform the calculation and is not practical. Further, if the speed of the arithmetic unit is increased to compensate for this, there is a disadvantage that the size and the price are increased.

SUMMARY OF THE INVENTION The present invention has been made in view of such conventional disadvantages, and has as its object to provide a tone control method for an electronic musical instrument that can simplify and speed up more natural tone control by using fuzzy inference.

(D) Means for Solving the Problems The present invention comprises a plurality of keys, and a performance means for generating a plurality of types of performance information including initial touch information, after touch information and key-on time information by operating each key; A plurality of fuzzy inferences including the initial touch information generated by the playing means, the first fuzzy inference based on the key-on information information, and the second fuzzy inference based on the after touch information are executed in real time, and based on the inference result, Fuzzy inference means for calculating pitch variation data; and pitch control means for controlling the pitch of a tone signal in accordance with the pitch variation data determined by the fuzzy inference means.

(E) Function In the tone control method for an electronic musical instrument according to the present invention, the pitch parameter is controlled by fuzzy inference by referring to performance information (key-on signal, initial touch intensity, after touch intensity, etc.) in a complex manner.

Here, the fuzzy inference is performed on "how much the pitch should be raised or lowered based on various performance information". Therefore, a plurality of fuzzy rules are determined. Fuzzy rules are generally expressed in the form if (x = A, y = B,...) Then (u = R). In the present invention, for example, if the initial touch (x) is large (A) and the after touch (y) is large (B), the pitch (u) is greatly increased (R). "If the initial touch (x) is small (A) and the after touch (y) is small (B), the pitch (u) is slightly lowered (R). "Initial touch (x) is large (A) and after touch (y) is small (B) and key-on time (z)
Is longer (C), the pitch (u) is not changed (R). ], Etc. are configured.

Here, a general fuzzy inference method will be described with reference to FIG. This method is called the mini, max rule. In this example, two fuzzy rules “if (x =
A ₁ ) then (u = R ₁ ), if (x = A ₂ , y = B ₂ ) then (u =
R ₂ )]. Each proposition (x =
A ₁ , x = A ₂ , y = B ₂ , u = R ₁ , u = R ₂ ) are represented by a membership function. The membership function of the conditional part (the proposition below “if”) indicates how much the input variable value (x ₀ , y ₀ ) belongs to the given fuzzy set (A ₁ , A ₂ , B ₂ ) This is a function for calculating a function value (membership values: α _x , β _x , β _y ), and the output of the condition part is the minimum one of the obtained membership values (α _x , β _x ) . This is a function to output the conclusion of this rule by the membership function of the conclusion part (the propositions below “then”). The spread in the control amount direction (u-axis direction) limited (cut off) by the output value of the condition part It is output as a value that has The final control amount (u ₀ ) ORs the conclusions of multiple phage rules,
The value of the center of gravity is used.

By using such inference for pitch control, it is possible to easily perform inference processing in which a plurality of pieces of performance information are considered in a complex manner, so that fine pitch control can be performed with a simple circuit configuration. Also in this case, the processing speed does not decrease. Further, by variously changing the fuzzy rules and membership functions, the characteristics of the musical instrument can be easily changed, and the pitch displacement can be easily characterized for each musical instrument.

(F) Embodiment FIG. 1 is a block diagram of an electronic musical instrument to which a pitch control method according to an embodiment of the present invention is applied. This electronic musical instrument has a keyboard-type musical instrument. The keyboard 1 has keys corresponding to a plurality of pitches. Each key has a key-on sensor for detecting key on / off, an initial touch sensor for detecting initial touch intensity (speed), and an after touch intensity. Aftertouch sensors for detection are provided, and the states of these sensors are detected by the key-on detection circuit 2, the initial touch detection circuit 3, and the aftertouch detection circuit 4. Here, the initial touch sensor includes, for example, two photo sensors that are continuously turned on in accordance with a key-on operation, and a photo sensor that is turned on later also serves as a key-on sensor. The key-on detection circuit 2 is always the keyboard 1
(Photo sensor) is scanned to monitor the on / off of each key. When there is a key-on, the key code (code indicating the pitch: KC), key-on signal (KON) and key sound duration (KONT) Is output. When a key is turned on, the initial touch detection circuit 3 detects and outputs the strength of the keystroke. Further, the after touch detection circuit 4 detects and outputs the pressing strength of the key that is turned on. The key code (KC) is input to the synthesizer 7, the key-on signal (KON) is input to the tone generator 8 and the envelope generator 9, and the key-on time signal (KONT) is input to the pitch parameter inference circuit 5. Pitch parameter inference circuit 5, sound source circuit 8, envelope generator 9
Is also input with an initial touch intensity signal and an after touch intensity signal. The pitch parameter inference circuit 5 is connected to a synthesizer 7 via a signal generation circuit 6. The pitch parameter inference circuit 5 infers the amount of pitch displacement based on the input signal and outputs it to the signal generation circuit 6. The signal generation circuit 6 generates a displacement frequency signal (CS) from the pitch displacement amount and outputs it to the synthesizer 7. The synthesizer 7 is a circuit that converts the input key code (KC) into a frequency signal (F number: digital value representing frequency) and adds the displacement frequency signal (CS) to this frequency signal. In this way, the frequency of the musical sound is determined and input to the tone generator 8. The tone generator 8 generates a digital signal (quantized signal) representing a predetermined waveform (tone color) at this frequency, and inputs this signal to the multiplier 10. An envelope generator 9 is connected to the multiplication circuit 10. The digital signal is amplitude-modulated by an envelope signal generated by the envelope generator 9, and is input to a D / A conversion circuit 11. The digital tone signal is sampled and held in the D / A conversion circuit 11, converted into an analog tone signal, and input to the amplifier 12.

FIG. 2 is a detailed block diagram of the pitch parameter inference circuit 5. This circuit is constituted by a fuzzy inference circuit. The fuzzy inference performed by this circuit is as follows: "if" large after touch "then" decrease pitch "" ... (1) "if" large initial touch "and" immediately after key-on "then" increase pitch "" ... (2) "if" large after touch "nor""large initial touch" and "Immediately after key-on""then" do not change pitch "" ... (3). The pitch fluctuation amount is determined by integrating these inference results. 3 (A) to 3 (D) show membership functions of each proposition constituting this rule. Where AT, KO,
IT is a membership function representing a fuzzy set (condition part) of “large after touch”, “immediately after key-on”, and “large initial touch”. Also, PN, PZ,
PP "lower pitch""do not change pitch"
It is a membership function corresponding to the conclusion of "raising the pitch". By executing the fuzzy rule using such a membership function, FIG. 4 (A),
The pitch control as shown in FIG. 10A and 10B, the upper row shows the intensity of the initial touch and the after touch, and the lower row shows the pitch displacement amount. FIG. 3A shows an example in which the initial touch is increased and the aftertouch is gradually increased. By performing in this way, it is possible to obtain a musical tone that gradually decreases from a high pitch. FIG. 6B shows an example in which the initial touch is slightly increased and the after touch is continuously weakened. By performing in this manner, it is possible to obtain a musical tone that starts sounding at a slightly higher pitch and continues at a substantially central pitch. By changing the fuzzy rule and the membership function in various ways, an arbitrary pitch displacement characteristic other than the above example can be obtained.

The configuration of the pitch parameter inference circuit 5 will be described. Membership function generator (MFC: Membership Function Ce)
rcuit) 101 to 103 are circuits for generating membership functions of the AT, IT, and KO of the condition part, and respectively receive an after touch intensity signal, an initial touch intensity signal, and a key-on time signal, and obtain a corresponding membership value. The membership function generator circuits 108 to 110
This is a circuit that generates N, PP, and PZ membership functions.
The minimum circuits 111 to 113 are (1) to (3), respectively.
This is a circuit for inferring the conclusion of the fuzzy rule.
The minimum circuit 111 receives the membership values and the membership functions of the membership function generation circuits 101 (condition part) and 103 (conclusion part) and receives the fuzzy rule (1).
Infer the conclusion. The membership values of the membership functions 102 and 103 are input to the minimum circuit 104, and a logical product (minimum value) is obtained, which is obtained by the fuzzy rule (2).
Is input to the minimum circuit 112 as the value of the condition part of. The minimum function 112 receives the membership function (PP) of the membership function generation circuit 109 and infers the conclusion of the fuzzy rule (2). The outputs of the membership function generating circuit 101 and the minimum circuit 104 are logically ORed (a maximum value is obtained) by a maximum circuit 106, and a subtractor 109
Is subtracted from "1" (a complementary set is obtained).
This value is input to the minimum circuit 113 as the value of the condition part of the fuzzy rule (3). The minimum circuit 113 has a membership function (P
Z) is input and the conclusion of the fuzzy rule (3) is inferred. The three conclusions inferred by the minimum circuits 111 to 113 are ORed by the maximum circuit 114 and the area is obtained. The obtained logical sum pattern and area are input to the center of gravity calculation circuit 115, and the center of gravity is calculated by the center of gravity calculation circuit 115. This center of gravity becomes a parameter of the amount of pitch fluctuation.

The minimum circuit, the maximum circuit, and the center-of-gravity calculation circuit can be constituted by a discrete circuit or a microcomputer. Fig. 5 (A) ~
(E) shows a flowchart in the case where the minimum circuit, the maximum circuit, and the center of gravity calculation circuit are constituted by a microcomputer.

FIGS. 7A and 7B are flowcharts for executing the operations of the maximum circuit 106 and the minimum circuit 104. FIG. In FIG. 3A, first, two scalar quantities (scl,
scl ₂₎ compares reads (n1), writes the scl ₁ if scl ₁ is larger in the memory scl ₀ (n2), when the scl ₂ is large writes scl ₂ in memory scl ₀ (n3). Fig. (B)
In, first, two scalar quantities (scl ₁ , scl ₂ ) are read and compared (n4). If scl ₁ is small, scl ₁ is stored in memory.
Write to scl ₀ (n5). If scl ₂ is small, write scl ₂ to memory scl ₀ (n6).

FIG. 9C is a flowchart for executing the operation of the minimum circuits 111 to 113. First, i representing the value of the horizontal axis of the membership function is set to 0 in n7. When the value of i becomes equal to or larger than the size of the horizontal axis of the membership function (size), the operation ends with the determination of n8. At n9, the value (mem (i)) at i of the membership function is read, and it is determined whether or not this value is equal to or less than the membership value scl of the condition part (n10). If mem (i) is less than scl, mem
Write the value of (i) to buffer (buf) (n12), mem
If (i) exceeds scl, buffer the value of scl (buf)
(N11), the value of this buffer is written to the conclusion memory memo (i) corresponding to i (n13), then 1 is added to i (n14), and the process returns to n8.

FIG. 11D is a flowchart for executing the logical sum and area calculation operation of the maximum circuit 114. First
At n15, 0 is set to the horizontal axis value i and the area accumulation memory acc. The value of i is the size of the horizontal axis of the membership function (siz
When the value exceeds e), the operation ends with the judgment of n16. At n17, the conclusion function values (mem1 (i), mem2 (i), mem3 (i)) of the fuzzy rules (1), (2), and (3) at i are read, and the largest one is determined at n18. . mem1
If (i) is the maximum, mem1 (i) is written to the buffer (buf) (n19), and if mem2 (i) is the maximum, mem2
Write (i) to buffer (buf) (n20), mem3
If (i) is the maximum, mem3 (i) is written to the buffer (buf) (n21). At n22, the value of the buffer is written to mem0 (i) (n22), and the value of the buffer is added to the area accumulation memory acc (n23). Thereafter, 1 is added to i (n
24) Return to n16.

FIG. 11E is a flowchart for executing the same center of gravity calculation operation as the center of gravity calculation circuit 115. First, a half value of the area (acc) obtained in FIG. 4D is stored in the storage area (half) (n25). Next, j corresponding to the horizontal axis of the ORed conclusion function and 0 in the area integration area hac are set to 0.
Is set (n26). Mem0 (j) is read into the buffer (buf) (n27), and this value is integrated into the area integration area hac (n28). Compare the integrated value of hac with (half) (n2
9) If (hac) is equal to or more than (half), the operation is terminated assuming that the value of j at that time is the center of gravity, and (hac) is changed to (hal).
If the value does not satisfy f), 1 is added to j, and the process returns to (n30) n27.

In the above embodiment, an electronic musical instrument that sounds in real time based on an operation of a keyboard or the like has been described. However, the present invention can be similarly applied to an apparatus that stores performance information in advance and performs automatic performance.

(G) Effects of the Invention As described above, according to the electronic musical instrument of the present invention, the use of fuzzy inference for pitch parameter control makes it possible to easily infer a pitch control amount that considers a plurality of pieces of performance information in a complex manner. With a simple circuit configuration, fine pitch control can be performed. Also in this case, the processing speed does not decrease. Further, by variously changing the fuzzy rules and membership functions, the characteristics of the musical instrument can be easily changed, and the pitch displacement can be easily characterized for each musical instrument.

[Brief description of the drawings]

FIG. 1 is a block diagram of an electronic musical instrument using a pitch control method according to an embodiment of the present invention, FIG. 2 is a block diagram of a pitch parameter inference circuit of the electronic musical instrument, and FIGS.
FIG. 4D is a diagram showing a membership function used in the pitch parameter inference circuit. FIGS. 4A and 4B are diagrams showing examples of pitch displacement accompanying keyboard operation of the electronic musical instrument.
FIGS. 7A to 7E are flowcharts showing the operation when each operation unit of the pitch parameter inference circuit is constituted by a microcomputer. FIG. 6 is a diagram for explaining a general fuzzy inference technique. 1 ... keyboard, 2 ... key-on detection circuit, 3 ... initial touch detection circuit, 4 ... after touch detection circuit, 7 ... pitch parameter inference circuit.

Claims (1)

(57) [Claims] 1. A playing means having a plurality of keys and generating a plurality of types of performance information including initial touch information, after touch information and key-on time information by operating each key, and initial touch information generated by the playing means. Fuzzy inference means for executing, in real time, a plurality of fuzzy inferences including a first fuzzy inference based on key-on information and a second fuzzy inference based on after-touch information, and calculating pitch variation data based on the inference results An electronic musical instrument comprising: and a pitch control means for controlling a pitch of a tone signal in accordance with pitch variation data determined by the fuzzy inference means.