JPH0315758B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to an electronic musical instrument, and more particularly to an electronic musical instrument that extracts parameters characterizing each tone from a plurality of types of musical tones of existing natural musical instruments, and uses these parameters to form a musical tone corresponding to a pressed key. It is something. Conventionally, musical sound waveforms of natural instrument sounds corresponding to the pitches of each key or parameters characterizing the natural instrument sounds are stored in advance for each key on a keyboard section, and the stored contents are read out as the keys are pressed. There are electronic musical instruments that generate musical tones. However, since this electronic musical instrument stores musical sound waveforms or parameters corresponding to the pitch of each key for each key, the storage capacity of the memory becomes enormous, making the device configuration large-scale and expensive. It had the disadvantage of becoming a thing. This invention was made in view of the above-mentioned drawbacks of conventional electronic musical instruments, and its purpose is to create an electronic musical instrument that is small-scale and can utilize existing natural musical instrument sounds to form musical tones corresponding to pressed keys. It is about providing. To this end, the electronic musical instrument according to the present invention inputs a plurality of types of musical sounds of existing natural musical instruments as sample sounds from a microphone, extracts parameters characterizing each sample sound from these sample sounds, and stores them in a memory. The stored parameters are interpolated according to the key touch, such as the pitch of the pressed key or the strength with which the key is pressed, and this interpolated value is used to respond to the key touch, such as the pitch of the pressed key or the strength of the key pressed. It is designed to form musical tones. Hereinafter, the present invention will be described in detail based on illustrated embodiments. FIG. 1 is a block diagram showing one embodiment of the present invention, which can be roughly divided into parameter extraction circuit 1.
0, memory circuit 20, interpolation calculation circuit 30, keyboard section 4
0 and a musical tone forming device 50. The parameter extraction means 10 is a microphone
Multiple tones of existing natural musical instruments are input as sample sounds from the MIC, and the amplitude coefficients of each harmonic that constitutes each sample sound are extracted from these sample sounds by fast Fourier transform processing. The coefficients are output as parameters for each sample sound. In this example, as shown in FIG. 2, a total of nine sample sounds A to I of high, medium, and low sounds, each having volume levels of strong, medium, and low, are input from the microphone MIC. Harmonic amplitude coefficient Cn for sounds A to I
[A] to Cn[I] are extracted. Here, n is 1
~ Represents the harmonic order of W. The storage circuit 20 stores the harmonic amplitude coefficients Cn[A] to Cn[I] of the nine sampled sounds extracted by the parameter extraction circuit 10 for each sampled sound. It has The write control circuit 200 writes each harmonic amplitude coefficient of nine sample sounds.
When Cn[A] to Cn[I] are input, these amplitude coefficients Cn[A] to Cn[I] are stored separately in the memory 201.
Address signal WA [A] to be stored in
While supplying WA [I] to the memory 201, the amplitude coefficients Cn [A] to Cn input from the extraction circuit 10
[I] is supplied as write data. As a result, the parameter memory 201 stores each of the amplitude coefficients Cn[A] to Cn[I] regarding the nine sample sounds separately. The interpolation calculation circuit 30 calculates the harmonic amplitude coefficient Cn[A]~ of each sample sound stored in the storage circuit 20.
Cn [I] is interpolated according to the pitch (X) and the strength (Y) of the pressed key in the keyboard section 40,
It forms a harmonic amplitude coefficient Cn [X・Y] corresponding to the pitch (X) and key depression strength (Y) of the pressed key, and each time a key depression operation is performed on the keyboard section 40, the pressed key is The harmonic amplitude coefficient Cn corresponding to
Form [X・Y]. That is, the keyboard section 40 is provided with a plurality of keys, and a key switch that operates when a key is pressed corresponding to each key, and a key touch detection sensor (for example, When a certain key is pressed, the key switch and key touch detection sensor corresponding to the key operate, and the key code KC representing the pitch (X) of the key and the key code KC representing the pitch (X) of the key are activated. Key press strength (Y)
The configuration is such that touch data TD representing touch is output. The key code KC and touch data TD output from the keyboard section 40 are supplied to the interpolation calculation circuit 30. When a key is pressed on the keyboard section 40 and the key code KC representing the pitch (X) of the key and the touch data representing the key pressing force (Y) are supplied to the interpolation calculation circuit 30, the interpolation calculation circuit 30 performs interpolation. The arithmetic circuit 30 calculates amplitude coefficients for the nine sample sounds A to I stored in the storage circuit 20.
Address signals RA[A] to RA for reading out single or multiple amplitude coefficients related to key code KC and touch data TD among Cn[A] to Cn[I]
[I] is given to the parameter memory 201. For example, as shown by symbols Xi and Yi in Fig. 2, the key code KC of pitch Xi and the touch data of key pressing force Yi
When TD is given to the interpolation calculation circuit 30, the calculation circuit 30 calculates the harmonic amplitude coefficients Cn[B], Cn[C], Cn[E],
Address signal RA[B] for reading Cn[F],
Output RA[C], RA[E], and RA[F] to the parameter memory 201. As a result, four sample sounds B, C, E,
Harmonic amplitude coefficients Cn [B], Cn [C], Cn regarding F
[E] and Cn[F] are read. Thereafter, as shown in FIG. 3, the interpolation calculation circuit 30 performs interpolation calculation between the coefficients Cn[E] and Cn[B] of the sample sounds E and B according to the pitch Xi, and obtains a calculation value _P1. After that, interpolation is performed between the coefficients Cn[F] and Cn[C] of the sample sounds F and C according to the pitch Xi to obtain the calculated value _P2 , and further these calculated values P An interpolation calculation is performed between ₁ and P ₂ according to the key pressing strength Yi, and this calculated value P ₃ is calculated as the harmonic amplitude coefficient Cn for the pressed key with the pitch Xi and the key pressing strength Yi.
Output as [Xi・Yi]. At the same time, the coefficients Cn [Xi, Yi] obtained in this way are applied to the tone forming device 5.
An address signal WAn [Xi·Yi] for writing into the amplitude coefficient memory 510 of 0 is output. In this way, the amplitude coefficient memory 510 stores a set of harmonic amplitude coefficients Cn [X, Y] that are interpolated according to the pitch (X) and the strength (Y) of the pressed key. It is memorized each time a key is operated. FIG. 4 shows a generalized interpolation process in the interpolation calculation circuit 30, and Table 1 below shows the contents of the calculation. In addition, in Figure 4
FLAG=1 to FLAG=5 mean flag signals for identifying which range the pitch (X) of the pressed key should be, and S1 to S5 are the pitches (X) of the pressed key.
and the processing steps of interpolation calculation according to the key depression strength (Y). Therefore, for example, when a key press operation indicated by Xi and Yi in FIG. 2 is performed, the flag signal FLAG=5 is set and the process of step S5 is executed. As a result, as is clear from Table 1, the above-described processing is performed.

[Table] As described above, the amplitude coefficient memory 510 of the musical tone forming device 50 stores a set of harmonics corresponding to the pitch (X) and the strength (Y) of the pressed key using natural musical instrument sounds. The amplitude coefficient Cn [X・Y] is stored, but
The musical tone forming device 50 forms musical tones by sequentially calculating the amplitude value X ₀ (qR) at each sample point of the musical sound waveform based on the following equation (1). X ₀ (qR)= _N/2 〓 ⁿ⁼¹ Cn[X・Y]sinπ/WnqR…(1) (q=1, 2,…) Here, X ₀ (qR)…each sample of the musical sound waveform Waveform amplitude value at a point. R...A numerical value proportional to the frequency (treble) of the generated musical tone (hereinafter referred to as frequency number). n...Represents the order of each harmonic component including the fundamental wave, where n=1 is the fundamental wave (fundamental tone), n=2 is the second harmonic (second overtone), and n=3 is the third harmonic (second overtone). 3rd harmonic) corresponds to... Cn[X・Y]...Amplitude coefficient (Fourier coefficient) for harmonic components of each order. N: Number of sequential sample points of one musical tone waveform at the highest frequency of the generated musical tone W: Total number of harmonics to be synthesized at each sample point. There is a relationship of W=N/2. Note that in the following description, harmonics include the fundamental wave, and the fundamental wave is assumed to be the first harmonic. That is, when a key press operation is performed on the keyboard section 40, the keyboard section 40 outputs a key code KC representing the pitch of the pressed key, and stores this key code KC as a frequency number R corresponding to each pitch. The frequency number memory 500 is supplied as an address signal to the frequency number memory 500. Then, the frequency number memory 500
A frequency number R corresponding to the pitch of the pressed key is read out. On the other hand, a clock oscillator 501 outputs a clock pulse tc of a constant period, and this clock pulse tc is frequency-divided by W in a counter 502 to become a calculation interval timing signal tz. In this case, "W"
is the total number of harmonics to be synthesized as described above, and for example, when synthesizing up to the 16th harmonic, "W=16". The calculation interval timing signal tx created in this way is supplied to the gate 503. This gate 503 is a calculation interval timing signal.
Frequency number memory 5 opens every time tx is supplied
The frequency number R output from 00 is supplied to the pitch section adder 504. Pitch interval adder 50
4 accumulates the frequency number R every time the frequency number R is supplied via the gate 503 (that is, every time the calculation interval timing signal tx is generated) and calculates 1R,
Outputs the cumulative value qR that increases as 2R, 3R, etc. Then, the interval adder 504 overflows when the accumulator qR exceeds the modulus N of the adder 504, and from then on, the calculation interval timing signal
A similar accumulation operation is performed again every time tx is generated. In this way, the accumulated value qR, which changes every time the calculation interval timing signal tx is generated, is determined by the clock pulse tc.
is fed to a harmonic section adder 506 via a gate 505 gated by . In this case, the clock pulse tc is the calculation interval timing signal
Since the frequency is W times that of tx, the gate 505 is opened W times during one cycle of the calculation interval timing signal tx. As a result, the harmonic interval adder 506 operates at the gate 5 every time the clock pulse tc occurs.
The cumulative value qR outputted from 05 is sequentially added and the cumulative value qR is output. When the harmonic interval adder 506 completes W accumulations, it is reset by the calculation interval timing signal tx and performs the same operation thereafter. Therefore, this harmonic section adder 506 calculates the cumulative value nqR (n=1, 2, 3, . . . W) that increases sequentially according to the clock pulse tc during one period of the calculation section timing signal tx. This means that it is occurring. This accumulated value nqR is decoded by the memory address decoder 507,
This decoded output is supplied as an address signal to a sine function memory 508 which stores sequential sample point amplitude values of one cycle of the sine wave waveform at each address.
The sine amplitude value sinπ/WnqR is read from the memory 508. As is clear from the above explanation, the cumulative value qR of the pitch interval adder 504 indicates the sequential sample points at which the musical waveform amplitude should be calculated, and the cumulative value nqR of the harmonic interval adder 506 is currently being calculated. sample point of
It represents the phase of the nth harmonic in qR. As a result, each harmonic (including the fundamental wave) at the sample point qR is output from the sine function memory 508.
The sine amplitude value sinπ/WnqR of is generated sequentially in the order of fundamental wave (first harmonic), second harmonic, . . . W-th harmonic. In this case, the sample points of the musical waveform to be calculated shift sequentially every time the calculation interval timing signal tx is generated, but which sample point to shift to next is determined by the frequency number R. This frequency number R is proportional to the pitch of the pressed key. Therefore, the sine amplitude value (sinπ/WnqR) of each harmonic corresponding to the pitch of the pressed key is sequentially generated from the sine function memory 508 in a time-sharing manner in accordance with the clock pulse tc. On the other hand, the counter 509 is composed of a modulus W counter, and the counter 502
The clock pulse tc is sequentially counted in synchronization with the clock pulse tc, and the count value is outputted to the amplitude coefficient memory 510 as an address signal n. The memory 510 stores the amplitude coefficient Cn of each harmonic calculated according to the pitch (X) and the strength (Y) of the pressed key as described above.
Since [X・Y] is stored, the counter 50
When an address signal n (indicating the harmonic order) that changes sequentially in synchronization with the clock pulse tc is supplied from 9 to 9, the harmonic amplitude coefficient Cn [which sets the amplitude value of each harmonic stored in each address] is supplied. X and Y] are read out sequentially. This harmonic amplitude coefficient Cn [X・
Y] is output to the multiplier 511. Multiplier 511
is the product value Fn obtained by multiplying the sine amplitude value sinπ/WnqR of each harmonic read out in a time-sharing manner from the sine function memory 508 by the harmonic amplitude coefficient Cn [X・Y] set for each harmonic. =Cn[X·Y]sinπ/WnqR is supplied to the accumulator 512. In this case, since the counter 509 is synchronized with the harmonic interval adder 506, each harmonic is read out sequentially, and the harmonic amplitude coefficient Cn [X・Y] is equal to the corresponding harmonic sine amplitude value sinπ/WnqR. By this, the amplitude value Fn for each harmonic is set. Accumulator 51
2 sequentially accumulates the amplitude value Fn for each harmonic output from the multiplier 511. Then, when the calculation interval timing signal tx is generated, the gate 513 opens and the accumulated value of the accumulator 512 (representing the amplitude value X ₀ (qR) at a certain sample point of the musical waveform)
is output to the DA converter 514, and the accumulator 512 is reset to perform the same accumulation operation as described above again in order to calculate the amplitude value at the next sample point. Therefore, the DA converter 514 has a cycle corresponding to the pitch of the pressed key, and each harmonic amplitude coefficient Cn
The amplitude value X ₀ (qR) (digital signal) of the musical sound waveform with the waveform shape set by [X・Y] is input every time the calculation interval timing signal tx occurs, and this digital amplitude value is converted into an analog signal. By converting it into a signal and supplying it to the sound system 515, the pitch corresponding to the pressed key is output to the memory 510.
A musical tone having a tone corresponding to the harmonic amplitude coefficient Cn [X.Y] stored in is generated. In the above embodiment, a harmonic synthesis method is adopted as the tone forming method in the tone forming device 50, and in connection with this, the parameter extraction circuit 10
Although the case has been described in which the extraction parameter is the harmonic amplitude coefficient Cn, the present invention is not limited to this, and any method can be adopted as the tone forming method in the tone forming device 50. The extracted parameters in the parameter extraction circuit 10 may also be adapted to extract appropriate parameters in accordance with the tone forming method of the tone forming device 50. As is clear from the above explanation, the electronic musical instrument according to the present invention inputs multiple types of musical sounds of existing natural musical instruments as sample sounds from a microphone, extracts parameters characterizing each sample sound from these sample sounds, and stores them in memory. The memorized parameters are interpolated according to the touch of the key, such as the pitch or strength of the pressed key, and the interpolated values are used to determine the pitch and strength of the pressed key. It is designed to form musical tones according to the musical tone. Therefore, it is possible to generate musical tones with a tone similar to that of a natural musical instrument with a small memory capacity, that is, a small-scale configuration.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing the relationship between pitch and volume of sample sounds used in the embodiment of FIG. 1, and FIG. 3 is an embodiment of the embodiment of FIG. 1. FIG. 4 is a flowchart showing the calculation process of the interpolation calculation circuit in the embodiment of FIG. 1. 10...Parameter extraction circuit, 20...Storage circuit,
30...Interpolation calculation circuit, 40...Keyboard section, 50...Tone forming device.

Claims (1)

[Scope of Claims] 1. Parameter extracting means for extracting sound parameters for each musical tone from a plurality of types of inputted musical tones, and a storage means for storing sound parameters related to each musical tone extracted by the parameter extracting means. and interpolating sound parameters related to at least two musical tones stored in the storage means in accordance with the pitch of the key pressed on the keyboard section or the touch of the key due to the key press to correspond to the pitch of the key or the touch of the key. An electronic musical instrument comprising: interpolation calculation means for forming a sound parameter formed by the correction calculation means; and musical tone formation means for forming a musical tone corresponding to a key pressed on the keyboard section using the sound parameter formed by the correction calculation means. 2. The plurality of musical tones are high, medium, and low.
The electronic musical instrument according to claim 1, wherein the musical tone has two frequency steps. 3. The electronic musical instrument according to claim 1, wherein the plurality of types of musical tones are high, medium, and low tones in three frequency levels, each having volume levels of strong, medium, and weak.