JPH0414356B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an electronic musical instrument of the type that synthesizes musical tone signals using frequency modulation calculations, and particularly relates to controlling the frequency ratio of a modulation signal and a carrier signal according to a musical range.

The instantaneous amplitude value e(t) of a musical tone signal synthesized by the frequency modulation method is basically expressed by the following equation.

e(t)=A(t)sin {ω _c +I(t)sinω _n t} ...(1) Here, A(t) is the amplitude coefficient expressed as a function of time, and ω _c is the angle of the carrier wave. The frequency, I(t), is the modulation index expressed as a function of time, ω _n is the angular frequency of the modulated wave, and t is the time.

By the way, the musical tones of natural musical instruments have different harmonic compositions (spectral distributions) depending on their range, and their timbres vary slightly depending on their range. However, in conventional frequency modulation type electronic musical instruments, the frequency ratio ω _c /ω _n (hereinafter this ratio will be expressed as N) between the carrier signal and the modulating signal was set corresponding only to the timbre type. The composition is the same in all ranges, and therefore it was not possible to obtain musical tones with different overtone compositions depending on the range, as is the case with natural instruments.

In view of this point, we decided to apply for the earlier application
In the electronic musical instrument shown in No. 1, modulation ratio control information for variably controlling the frequency ratio described above according to the tonal range is stored in advance in the memory, and this information is transferred to the memory according to the tonal range of the pressed key. The carrier frequency or modulation frequency is controlled using this information. However, since it is necessary to store modulation ratio control information corresponding to all sound ranges or pitches, there is a problem in that the memory capacity becomes large.

An object of the present invention is to increase the frequency ratio N between a carrier signal and a modulation signal in a frequency modulation type electronic musical instrument.
The purpose of the present invention is to make it possible to generate musical tones having different overtone compositions depending on the tonal range, like the sounds of natural musical instruments, by controlling the musical tones according to the pitch or range of the tones to be generated. A further object of the present invention is to realize control of the frequency ratio N according to pitch or range with a simple configuration. Therefore, in the present invention, data indicating the frequency ratio N of the carrier signal and the modulation signal is calculated by a predetermined calculation using information indicating the pitch or range of the musical sound to be generated and a predetermined calculation parameter, The carrier frequency or modulation frequency is controlled according to the frequency ratio N thus calculated.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an example in which the present invention is implemented in an electronic musical instrument that synthesizes musical tones according to the basic frequency modulation calculation formula shown in equation (1) above. The key switch circuit 10 includes a key switch corresponding to each key on the keyboard, and corresponds to a key press with a key code KC indicating the pressed key and a key on indicating whether the key is kept pressed (or released). Outputs the signal KON. The frequency number table 11 stores in advance a predetermined numerical value (referred to as a frequency number) corresponding to the musical tone frequency of each key, and corresponds to the pressed key according to the key code KC given from the key switch circuit 10. Read out the frequency number ω.

The frequency ratio calculation device 12 includes a key switch circuit 1
A predetermined calculation is performed based on the key code KC given from 0 and the timbre selection information given from the timbre selection device 13, and the frequency ratio N (=ω _c /ω _n ) between the carrier frequency ω _c and the modulation frequency ω _n is indicated. Calculate the data. The accumulator 14 outputs data indicating the instantaneous phase angle ω _n t of the modulation frequency ω _n by repeatedly adding (or subtracting) the frequency number ω read from the table 11 with a predetermined modulus. On the other hand, the frequency number ω read from the table 11 is also given to the multiplier 15, where it is multiplied by the frequency ratio N given from the arithmetic unit 12. The output "N·ω" of this multiplier 15 is given to the accumulator 16 as data indicating the carrier frequency ω _c . Similar to 14, the accumulator 16 repeatedly adds (or subtracts) the data ω _c with a predetermined modulus, and calculates the instantaneous phase angle ω _c t of the carrier frequency ω _c .
Outputs data indicating. Since ω=ω _n ,
ω _c =N·ω _n , and the carrier frequency ω _c is determined according to the frequency ratio N.

The output of the accumulator 14 is the sine wave table 1
The sine wave amplitude value sinω _n t corresponding to the phase angle ω _n t is read from the table 17 . This sine wave signal, that is, the modulation signal, is given to the multiplier 15, and the envelope generator 1
The modulation index I(t) given from 9 is multiplied. The envelope generators 19 and 20 each generate a predetermined envelope waveform signal in response to the key-on signal KON applied from the key switch circuit 10,
This envelope waveform signal is output as a modulation index I(t) and an amplitude coefficient A(t), respectively. The output signal “I(t)sinω _n t” of the multiplier 18 is provided to the adder 21 and added to the phase angle data ω _c t of the carrier signal provided from the accumulator 16 . The sine wave table 22 is read using the output signal "ω _c t + I(t) sin ω _n t" of the adder 21 as an address signal, and the multiplier 23 reads out the output signal of the table 22 and the amplitude coefficient A given from the envelope generator 20.
(t), a frequency modulated signal e(t), that is, a musical tone signal, as expressed by the above equation (1) is obtained. The musical tone signal output from the multiplier 23 is given to a sound system 24.

FIG. 2 shows an example of the frequency ratio calculation device 12, which calculates formant center frequency information ogFo, key scale slope information α, and frequency ratio offset information.
Nof is given from the timbre selection device 13 (FIG. 1) in accordance with the timbre selected by the device 13 (FIG. 1). In this example, a predetermined fixed formant is realized according to the selected timbre, Fo indicates the fixed formant center frequency, and information ogFo represents this in logarithmic form. Fixed formant musical tones (timbres) are realized by varying the overtone composition of musical tones depending on pitch or range (this is called a key scale) so as to obtain a desired fixed formant. In other words, the fixed formant adjusts the frequency ratio N of the carrier signal and the modulating signal so that the carrier frequency ω _e is always located at or near the fixed formant center frequency Fo, regardless of the pitch or range of the musical note. This is achieved by changing the pitch according to the pitch or range. The key scale slope information α is used to set the slope of the key scale (that is, the degree of difference in overtone structure depending on pitch and range). When performing key scaling to realize a completely fixed formant, the information α is set to a predetermined value (in this example, α=
1), and when it is desired to move the formant center frequency somewhat depending on the sound range, the information α is set to an appropriate value other than the predetermined value (α≠1). The frequency ratio offset information Nof is used to entirely shift (translate) the slope of the key scale determined by the information α.

The frequency information conversion circuit 25 receives the key code KC of the pressed key given from the key switch circuit 10 (FIG. 1), and converts this key code KC into a musical tone frequency Fn (that is, pitch) of the pressed key in negative logarithmic form. Convert to information expressed as "-ogFn". This pressed key frequency information -ogFn and the above-mentioned formant center frequency information ogFo are added by an adder 26, and a logarithmic expression ogFo/Fn of Fo/Fn is output from the adder 26. The output of adder 26 is applied to multiplier 28 via gate 27. The adder 26 has a comparison function in addition to the addition function, and determines whether the condition "Fo≧Fn" is satisfied based on both input signals, and when the condition is satisfied, opens the gate 27.
A control signal is given to the gate 27 to close the gate 27 when the condition is not satisfied. This is because it is impossible to form a fixed formant if the formant center frequency Fo is smaller than the frequency Fn of the pressed key.
This is to force ogFo/Fn to be 0. In other words, when "Fo<Fn", the output signal of the adder 26
A gate 27 is provided to change ogFo/Fn to the same value as when "Fo=Fn", that is, 0.

The multiplier 28 multiplies the signal ogFo/Fn given from the gate 27 by the above-mentioned key scale slope information α. As a result, the value α·ogFo/Fn=og(Fo/Fn) is output from the multiplier 28. Furthermore, information α
is a numerical value including decimal digits, and the multiplier 28 outputs a signal obtained by rounding off fractions below a predetermined significant digit in the multiplication result. The output signal og(Fo/Fn) 〓 of the multiplier 28 is given to the logarithmic/linear conversion table 29,
Converted to a numerical value (Fo/Fn) in linear representation format. This value (Fo/Fn) corresponds to the frequency ratio N between the carrier frequency ω _c and the modulation frequency ω _n . That is, for a completely fixed formant, Fo = ω _c , and Fn
=ω _n , so if α=1, (Fo/Fn) = ω _c /ω _n =N holds true. Therefore, if you want to achieve a completely fixed formant, set α to 1, and if you want to move the formant slightly depending on the range of the musical note, set α to 1.
α is set to an appropriate value other than the above (for example, a number including a decimal number near 1, such as 0.5).

The output (Fo/Fn) of the logarithmic/linear conversion table 29 is given to an adder 30, and the above-mentioned frequency ratio offset information Nof is added thereto. The output signal of this adder 30 is the first numerical data indicating the frequency ratio N.
The signal is applied to multiplier 15 in the figure. The adder 30 is designed to round off the decimal part of the addition result and output the result, and the frequency ratio N is given by an integer.

FIG. 3 shows an example of the key scale of the frequency ratio N calculated by the arithmetic unit 12 of FIG. 2. The horizontal axis shows the musical tone frequency Fn of the pressed key, and the vertical axis shows the frequency ratio N. Fo=1kHz, Nof=0, and the broken line shows the key scale curve when α=1. The actual frequency ratio N corresponding to the dashed curve has a step shape as shown by the solid line. This is because the frequency ratio N is rounded to the nearest integer to the dashed curve. Referring to the curve of α = 1, for example, when Fn, that is, ω _n = 125 Hz, N = 8 can be found, and by multiplication in the multiplier 15 (Fig. 1), ω _c = 125 × 8 = 1 kHz. I understand. Furthermore, it can be seen that when Fn=250Hz, N=4, and ω _c =250×4=1kHz. In this way, a perfect fixed formant is realized when α=1. In Figure 3, the dashed line is α
The curve when = 0.5 is shown. In this case, Fn=1k
Hz, N=1 and ω _c =1kHz, but Fn=250Hz
When N = 2, ω _c = 500Hz, and when Fn = 125Hz, N
=3, ω _c =375Hz, and the formant moves.

Note that the information ogFo, α, and Nof provided corresponding to the timbre selected by the timbre selection device 13 are provided from a ROM (abbreviation for read-only memory), not shown. This ROM may be included on the timbre selection device 13 side, or may be included on the arithmetic device 12 side.

The calculation format in the frequency ratio calculation device 12 is the second
It is not limited to what is shown in the figure, and other formats may be used, for example, interpolation calculations as described below. To explain the frequency ratio calculation device 12 using interpolation calculation, as an example, the frequency ratio N corresponding to a key with a predetermined note name (for example, note name C) for each octave of the keyboard is explained.
The frequency ratio N corresponding to each key or range is determined by preparing values in advance and performing interpolation calculations between these previously prepared numerical values. Assume that the keyboard has 6 octaves (73 keys) from keys C1 to C7, and key C is prepared in advance.
1, C2, C3, ... Frequency ratio N corresponding to C7
The values of are shown as N ₁ , N ₂ , N ₃ ...N ₇ , respectively. The number indicating the number of the pressed key from key C in the same octave is indicated by x, and the frequency ratio values N ₁ to N ₇ of the keys C1 to C7 corresponding to the octave to which this pressed key belongs is denoted by Ny, and values N ₁ to N ₇ one octave above N y are denoted by N _y+1 , then the frequency ratio N corresponding to the pressed key is determined by performing interpolation calculation according to the following formula.

_{N = Ny +}
x=4, and the above equation (2) is executed based on these parameters. Of course, the values N ₁ , N ₂ , N ₃ ...
It goes without saying that N ₇ is a parameter corresponding to the timbre selected by the timbre selection device 13.

In the above equation (2), the numerical value 12 of "X/12" is the number of whole tones (number of keys) within one octave. the result,
A different frequency ratio N can be calculated for each individual key. However, the key scaling need not be performed so precisely, and the frequency ratio N may be calculated for each predetermined tone range (key range). When one octave is divided into four parts and the frequency ratio N is calculated for each range (key range) consisting of three keys, the above equation (2) is changed as follows.

N=Ny+X/4(Ny ₊₁ −Ny)...(3) Here, unlike x in equation (2), This is a number indicating the number of key C in the key range (key range consisting of three keys) to which it belongs. for example,
When the key E3 is pressed, the calculation parameters in the above equation (3) are Ny=N ₃ , N _y+1 =N ₄ , and X=1.

FIG. 4 shows an example of a frequency ratio calculation device 12 that performs the interpolation calculation of equation (3) above. The basic value memories 31 and 32 store in advance values N ₁ , N ₂ , N _{3 .} . . N ₇ that are the basis for interpolation calculations, respectively. A plurality of sets of these basic values N ₁ to N ₇ are stored corresponding to each timbre, and the timbre selection information given from the timbre selection device 13 (FIG. 1)
A set of basic values N ₁ to _{N 7} corresponding to the selected timbre is selected according to the TS. Of the key code KC given from the key switch circuit 10 (FIG. 1), the code portion indicating the octave is stored in each memory 31.
and 32 address inputs. However, the memory 31 is addressed to the same octave as the octave code of the key code KC, and the memory 32 is addressed to the octave one octave above. As a result, the basic value data Ny of the octave to which the pressed key belongs is stored from the memory 31.
is read out, and basic value data N _y+1 one octave above it is read out from the memory 32.

The data Ny read from the memory 31 is added to the adder 33 and converted into a negative value "-Ny" by the sign conversion circuit 34, and then given to the adder 35. Data output from memory 32 N _y+1
is given to the adder 35. The output of the adder 35 is given to the multiplier 36 and multiplied by "1/4". The output of multiplier 36 is given to multiplier 37.

On the other hand, the key code KC is the key range data generation circuit 3
8, and the aforementioned numerical value X indicating the number of the key range to which the pressed key belongs is generated in accordance with the key code KC. In this circuit 38, the key code
Based on the note chord part of KC, the following value of X is generated depending on the note name. That is, when the pitch names are C, C#, and D, X=0, D#,
When E, F, X=1, when F#, G, G#, X=
2, A, A#, B, then X=3. This numerical value X is given to a multiplier 37 and multiplied by the output signal of the multiplier 36. The output of multiplier 37 is output to adder 3
3 and is added to Ny. In this way, the interpolation calculation of equation (3) is executed, and the adder 33 outputs data indicating the frequency ratio N, which is the solution.

Note that since the memories 31 and 32 store the same contents, one memory can be shared in time division.

FIG. 5 shows an example in which the present invention is implemented in a device that synthesizes musical tones according to a nested frequency modulation calculation as shown in equation (4) below.

e(t)=A(t)sin[ω _c t+I ₂ (t)sin {ω _n2 t +I ₁ (t)sinω _n1 t}] …(4) Here, ω _n1 and ω _n2 are the modulation frequencies, I ₁ (t) and I ₂ (t) are modulation indices, and the others are the same as in equation (1).

In FIG. 5, a key code KC and a frequency number ω are generated corresponding to the pressed keys in the same way as in FIG. Also, modulation index I ₁ (t), I ₂ (t) and amplitude coefficient A
(t) is also generated in response to a key press from an envelope generator (not shown) as in FIG. The first and second frequency ratio calculation devices 39 and 40 are configured as shown in FIG. 2 or 4, similar to the calculation device 12 in FIG. 1, and have key codes indicating pressed keys.
Frequency ratios N(1) and N(2) are calculated according to the timbre selection information given from KC and the timbre selection device 41, respectively.

The frequency number ω is repeatedly added (or subtracted) by an accumulator 42 and is provided to a sine wave table 43 as phase angle data ω _n1 t of the first modulation signal. Also, in the multiplier 44, the frequency number ω
and the frequency ratio N(1) determined by the first arithmetic unit 39
The output N(1)·ω is repeatedly added (or subtracted) by the accumulator 45. Also,
The multiplier 46 multiplies the frequency number ω by the frequency ratio N(2) obtained by the second arithmetic unit 40, and the output N(2)·ω is repeatedly added (or subtracted) by the accumulator 47. . The output of the accumulator 45 is given to the adder 48 as phase angle data ω _n2 t (where ω _n2 =N(1)·ω) of the second modulation signal. The output of the accumulator 47 is given to the adder 49 as carrier signal phase angle data ω _c t (where ω _c =N(2)·ω).

The sine wave table 43 is read out using the first modulation signal phase angle data ω _n1 t output from the accumulator 42, and the multiplier 50 adds the readout output to the sine wave table 43.
is multiplied by the modulation index I ₁ (t), and the output is sent to the adder 48
give to The sine wave table 51 is read out by the output of the adder 48, the read output is multiplied by the modulation index I ₂ (t) in the multiplier 52, and the output is given to the adder 49. The sine wave table 53 is read out based on the output of the adder 49, and the readout output is multiplied by the amplitude coefficient A(t) in the multiplier 54. thus,
A musical tone signal e(t) as shown in equation (4) above is output from the multiplier 54.

Here, the frequency ratio ω _n2 /ω _n1 between the first modulation frequency ω _n1 t and the second modulation frequency ω _n2 is N(1), and the first
This is set as a result of calculation by the frequency ratio calculation device 39. Further, the frequency ratio ω _c /ω _n1 between the first modulation frequency ω _n1 and the carrier frequency ω _c is N(2), and is set as a result of calculation by the second frequency ratio calculation device 40. In this way, the first and second modulation frequencies ω _n1 , ω _n2 in the multi-frequency modulation calculation
The frequency ratio N(1) and the frequency ratio N(2) between the first modulation frequency ω _n1 and the carrier frequency ω _c are also variable depending on the range of the pressed key, as in the case of basic frequency modulation calculations. can be controlled.

According to the basic frequency modulation calculation formula shown in equation (1) above, it is easy to form relatively simple fixed formants such as human voice sounds, but it is easy to form relatively simple fixed formants such as the sounds of natural instruments such as piano. It is not good at forming formants. In this regard, according to the multi-frequency modulation calculation formula (configuration shown in Figure 5) as shown in equation (4) above, it is possible to simulate the soundboard effect of a piano or stringed instrument (fixed formant components generated by the soundboard). Suitable for synthesizing complex fixed formants (synthesis) or other natural instrument sounds.

An example of the spectral distribution of natural musical instrument sounds is shown in FIGS. 6a and 6b. This is the spectral distribution of the piano sound, where a is the spectral distribution in the bass range and b is the spectral distribution in the treble range. As you can see from the figure, there are many high-order overtones in the low range, but the higher the range, the fewer high-order overtones become, resulting in a fixed formant (or something close to it).

When simulating the spectral distribution shown in FIGS. 6a and 6b using the apparatus shown in FIG. 5, the following is an example. First, when the key code KC belongs to a predetermined bass range, the first arithmetic unit 39 outputs a decimal number "1" as the frequency ratio N(1), and the second arithmetic unit 40 outputs the frequency ratio N(2). ) outputs the decimal number "8". This is an arithmetic unit 39, 40 having a configuration as shown in FIG. 2 or 4.
This is possible by setting the calculation parameters to predetermined values depending on the range and timbre. As a result, the frequency ratio of the modulation signal and carrier signal ``ω _n1 :
ω _n2 :ω _c ” becomes “1:1:8”, and a musical tone signal e(t) having a spectral distribution as shown in FIG. 7(a) can be synthesized. As is clear from FIG. 7(a), the spectral distribution of the bass range as shown in FIG. 6(a) is simulated.

Next, when the key code KC belongs to a predetermined treble range, the first calculation device 39 outputs "1" as the frequency ratio N(1), and the second calculation device 40 outputs "1" as the frequency ratio N(1).
Output "4" as (2). Similarly to the above, this is also possible by setting the calculation parameters in the calculation devices 39 and 40 to predetermined values depending on the musical range and timbre. As a result, “ω _n1 :
ω _n2 :ω _c ” is 1:1:4, and a musical tone signal e(t) having a spectral distribution as shown in FIG. 7b can be synthesized. As is clear from FIG. 7b, the spectral distribution in the treble range as shown in FIG. 6b is simulated.

As described above, by controlling the frequency ratio of the carrier signal and the modulation signal according to the sound range according to the present invention, it is possible to accurately simulate the spectral distribution of natural musical instrument sounds as shown in FIGS. 6a and 6b. can. If you do not change the frequency ratio according to the range,
If we try to simulate the spectral distribution shown in Figure 6b by controlling the modulation degree, we will get Figure 7c.
This is not desirable. That is, in Figure 7c, the frequency ratio ``ω <sub>n1</sub> :ω <sub>n2</sub> :ω <sub>c</sub> '' remains ``1:1:8'' as in Figure 7a, and the spectral level can be changed by controlling the modulation degree. ing. Therefore, the bandwidth of each peak in the spectral distribution is the same as in FIG. 7a, and overtones are dropped by the amount that the level is lowered. On the other hand, FIG. 7b simulates FIG. 6b by setting the bandwidth of each peak to half that of FIG. 7a, so there is no such inconvenience.

In FIG. 5, the first modulation frequency ω _n1 is made the same as the fundamental frequency ω, but the second modulation frequency ω _n2 or the carrier frequency ω _c may be made the same as ω. Further, the frequency to be made the same as the fundamental frequency ω may not be fixed to one of ω _n1 , ω _n2 , and ω _c , but may be changed depending on the tone color. In that case, a multiplier is also provided upstream of the accumulator 42 shown in FIG. 5, and an extra frequency ratio calculation device is provided to provide frequency ratio data to this multiplier. Furthermore, only one term in the multi-frequency calculation equation may be used to change the frequency ratio depending on the sound range. For example, if only the frequency ratio N(2) corresponding to the carrier frequency ω _c is to be variably controlled depending on the sound range, the second
It is sufficient to provide only the arithmetic unit 40.

Although a single multiple frequency modulation calculation is performed in FIG. 5, the present invention can be implemented in the same manner as described above even when multiple frequency modulation calculations are performed. It goes without saying that the present invention can also be implemented when polynomial frequency modulation calculations are performed. As is well known, in a multitone electronic musical instrument, a key assigner is provided in association with the key switch circuit 10, and the key code KC of the pressed key assigned to each channel and the key-on signal KON are output in a time-sharing manner. In that case, frequency ratio calculation devices 12, 3
9 and 40, it goes without saying that arithmetic operations are performed in a time-division manner in correspondence with the time-division time slots of the key code KC. Furthermore, it goes without saying that the electronic musical instrument according to the present invention can be implemented not only by discrete circuits but also by software processing using a microcomputer.

As explained above, according to the present invention, the frequency ratio between the carrier frequency and the modulation frequency can be varied depending on the pitch or range, so for example, a sound with many overtones in the low range and a sound with few overtones in the high range can be compared. It is possible to synthesize musical tones that more closely match the characteristics of natural musical instruments, such as rounded tones. Moreover,
Since the frequency ratio corresponding to the pitch range of each tone is determined by calculation, the circuit configuration can be simplified compared to the case where the frequency ratio is stored in a memory. The present invention also exhibits excellent effects in multiplex frequency modulation calculations.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the overall configuration of an electronic musical instrument showing one embodiment of the present invention, FIG. 2 is a block diagram showing an embodiment of the frequency ratio calculation device of FIG. 1, and FIG. 3 is a block diagram of the same frequency ratio calculation device. The frequency ratio N thus realized is
FIG. 4 is a block diagram showing another embodiment of the same frequency ratio calculation device; FIG. 5 is a block diagram showing the main parts of another embodiment of the electronic musical instrument according to the present invention. , FIG. 6a is a graph schematically showing an example of the spectral distribution of natural musical instrument sounds in the bass range, FIG. 6b is a graph schematically showing an example of the same spectral distribution in the treble range, and FIG.
Figure a is a spectral distribution diagram schematically showing an example of simulating the distribution in Figure 6a using the embodiment shown in Figure 5; ) is a spectral distribution diagram schematically showing an example of simulating the distribution of FIG. 12, 39, 40... Frequency ratio calculation device, 1
3, 41... Tone selection device, 15, 44, 46...
... Multiplier for controlling carrier frequency or modulation frequency according to frequency, 14, 16-23... Circuit for synthesizing musical tones by basic frequency modulation calculation, 42, 43, 45, 47- 54...Circuit for synthesizing musical tones by multiple frequency modulation calculation, 31, 32...Basic value memory, 33-3
8...Circuit for interpolation calculation.

Claims (1)

[Scope of Claims] 1. A musical tone generating means that generates a musical tone signal corresponding to a desired tone by frequency modulation calculation, and information indicating the pitch or range of the musical tone to be generated by this musical tone generating means as a calculation parameter. a calculation means that executes a predetermined calculation and calculates the frequency ratio of the carrier signal and the modulation signal according to the pitch or range of the musical tone to be generated; It is characterized by comprising a control means for changing the harmonic composition of the musical tone signal corresponding to the desired timbre according to the pitch or range by controlling the frequency of the carrier signal or modulation signal in the musical tone generating means. An electronic musical instrument. 2. The musical tone generating means generates a musical tone signal by a multiplex frequency modulation calculation, and the calculation means has at least a frequency ratio of the carrier signal and the modulation signal in the multiplex frequency modulation calculation. One frequency ratio is determined, and the control means controls the frequency of the carrier signal or modulation signal corresponding to the frequency ratio determined by the calculation means in accordance with this frequency ratio. The electronic musical instrument according to item 1. 3. The calculating means calculates a ratio of both frequencies based on frequency information indicating the pitch of the musical tone to be generated and fixed frequency information determined according to the timbre, and converts numerical information corresponding to this ratio into the frequency ratio. 3. The electronic musical instrument according to claim 1, further comprising an arithmetic circuit that provides information to the control means as the information. 4. The calculation circuit calculates a ratio between both frequencies based on frequency information indicating the pitch of the musical tone to be generated and fixed frequency information determined according to the timbre.
and a second circuit that calculates a constant determined according to the timbre to the numerical value indicating the ratio determined by the first circuit, and the output of the second circuit is used as information indicating the frequency ratio. The electronic musical instrument according to claim 3, wherein the electronic musical instrument is provided to a control means. 5. The calculation means stores numerical data indicating the frequency ratio corresponding to a plurality of predetermined pitches or ranges in advance as basic values, and responds to the information indicating the pitch or range of the musical tone to be generated. a basic value memory for reading out the two basic values in the vicinity of the pitch or range; and information indicating the pitch or range of the musical tone to be generated and the two basic values read from the basic value memory. The electronic musical instrument according to claim 1 or 2, further comprising an interpolation calculation circuit that executes a predetermined interpolation calculation as a parameter to obtain the frequency ratio corresponding to the pitch or range of the musical tone to be generated. .