JPS5827520B2 - Denshigatsuki - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an improvement in an electronic musical instrument in which partial tones are determined by recursive calculations.

For this type of electronic musical instrument, it has been considered to generate sample values of partials at time intervals T to synthesize musical tones (here, partials are constituent elements of musical tones, and the fundamental tone and a musical tone element having a frequency that is an integral multiple of the frequency of the fundamental tone;
(This is a concept that includes musical tone elements whose frequencies are not integral multiples of the frequency of the fundamental tone.)

For example, when synthesizing a musical tone from M partials, as shown in FIG. 1, for each partial, a sample value, fm(0)1. fm(T), fm(2T),...
・・・・・・・・・frn(nT)・・・・・・・・・
・For each time T, 2T,......nT・
Sample values of musical tones are obtained for..., and digital signals representing these sample values are converted into analog signals to obtain musical tones.

FIG. 2 shows a specific example of the electronic musical instrument described above, which operates according to clock pulses CK1 and CR2 (FIG. 3) generated from a synchronization control section (not shown).

Clock pulse CK1 has period T and pulse width τ/2=T
/2M, and the clock pulse CK2 has a period τ−T/
M, pulse width τ/2.

Here, the period τ corresponds to the time for calculating one partial tone sample value.

That is, in this example, from time nT to time (n+1)T
Partial sample values f s (n
T ), f2(nT), f m(n
T ), ...... fM (nT) is calculated, and at time (n-+-1)T, the sample value F (nT
) is generated.

Now, when the keyboard is operated and a signal indicating the key number of the operated key is issued from the key number designation device 1, the memory address control device 2 receives the initial partial sample value storage device 6 from the initial partial tone sample value storage device 6 by the address signal. Time τ (=T/M
) every initial partial sample value (fl(O), f, (T))
, (f2(0), f2(T)......(fm
(0), fm(T)・・・・・・**・・・(fM(0
)1. fM(T) is sequentially outputted, and the partial tone sample value storage control device 5 stores these values, and also stores the partial tone sample values fl(0), f2 (0')-. in the accumulator 7.
−−−−−・f m (0)−・・・−・fM (0
) is sent.

The accumulator 7 performs the following calculation to obtain the sample value F(0) of the musical tone at time T.

In the time interval between time T and time 2T, the signal from the memory address control device 2 causes the parameter storage devices 3P and 3Q to change pl, p at every time interval τ(T/M).
2・・・・・・・・・pm・・・・・・・・・PM and ql, ql, -”””” (1m””””°(IM is sent to the arithmetic unit 4, partial tone The sample value storage control device 5 stores the partial tone sample values, ft (o), f2 from the terminal Tn-2.
(0),...fm(0),...
..., fM(0), partial tone sample values from terminal Tn-1, fl(T), f2('r),...
fm(T), . . . fM(T) is sent to the arithmetic unit 4.

The arithmetic unit 4 performs the following calculations from these values at intervals of time τ to obtain partial tone sample values f1 (2T) and f2.
(2T),...------・f m (2T)--
−・−・・Output f M (2T).

The partial tone sample value storage control device 5 receives and stores these values, and also stores the previously stored partial tone sample values f1(T),
f2(T),... fm(T)...
...Outputs fM(T).

Accumulator 71, +〉-Jy+l"II>Noko-A
, /'', +Dou (The sample value F(T) at time 2T of the musical tone is obtained by performing the following calculations.

Between time (n-1)T and time nTO, the parameter storage devices 3P and 3Q record pl, p2°""""pm""""" pM and q at every time interval τ as described above.
l, q2°"""" (1 m """"' qM is sent to the arithmetic unit 4, and the partial tone sample value storage control device 5 connects the terminal T
Partial sample value f't((n 2)T), f from n-2
2((n-2)T), -・=, fm((n-2)T)・
・・・・・・・・・fM((n-2)T) and terminal T
Partial sample value f t ((n 1 ) T
), f2((n 1)T), ---fm((n
2) T)...-...fM((n-2)T) is sent to the arithmetic unit 4.

The calculation device 4 performs calculations at intervals of time τ from these values to obtain partial tone sample values f1(nT), f
2 (nT), −・−f m(nT),・−= f
M (nT), and the partial tone sample value storage control device 5 stores these values and also stores the previously stored partial phoneme *Kimoto values ft((n 1)T), f2((n
1)T), ...... f m ((n-1)T
), ・”−・fM((n 1)T) is output.

The accumulator 7 receives these values and performs calculations to obtain the sample value F(
(n-1)T) is determined.

From time nT to time (n+1)T, partial tone sample values f, (nT), f are output from the partial tone sample value storage control device 5.
2 (nT), -・f m (nT)...
. . . f M (nT ) is output, and the accumulator 7 receives these values and performs arithmetic operations to obtain the sample value F (nT ) of the musical tone at time (n+1)T.

The DA converter 8 receives sample values F(0), F(T), etc., which are sequentially sent from the accumulator 7 at time intervals T.
. . . A digital signal indicating F(nT) is converted into an analog signal indicating a musical tone, and the audio device 9 receives this analog signal and emits a sound.

However, in such electronic musical instruments, both the parameters T)m and (1m) are involved in the frequency and amplitude fluctuations of each partial, so in order to obtain a musical tone with a desired partial tone structure, the parameters pm, (1m) are The problem is that it is very difficult to set up.

In addition, all parameters pms qm must be stored for each partial, and the memory capacity must be extremely large in order to form musical tones with a large number of partials.1 This invention solves the above points. To provide an electronic musical instrument in which parameters can be easily set, musical tones with a wide variety of partial tone configurations can be easily created, and the capacity of the parameter memory can be reduced.

In order to achieve this objective, the present invention uses the above parameters pm and (1m) as a factor that determines the frequency of each partial (hereinafter referred to as frequency factor) and a factor that determines amplitude fluctuations (
Hereinafter, the frequency factor is calculated from the amplitude variation factor), and the frequency factor is calculated sequentially based on a plurality of past sample values.

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

FIG. 4 shows an embodiment in which the present invention can be applied to the electronic musical instrument shown in FIG.
and operates in synchronization with CR2.

That is, during the time interval T, the frequency factors λ1, λ2, . . . λ □ . . .
・λM and amplitude variation factors α1, α2, ...
・・α ・・・・・・・・・CtM is generated, and parameters (pt, qt), (p
2, (12)・・・・・・・・・(pml 9m)
is obtained for each time τ.

However, the first j frequency factors λ1 to λj are stored, and λj+1 to λM are calculated based on the previously obtained j frequency factors.

Here, partial sound f1(nT) is generated at every time τ between time nT and time (n+1)T.
・・・・・・fHl(nT), ・・・・・・・・・j'M(
Let us consider the case of finding nT).

In the first time interval, that is, the time corresponding to calculating the partial tone sample value f1 (nT), the address signal from the memory address control device 2 is assisted by the address signal from the memory address control device 2 based on the signal indicating the key number issued from the key number designation device 1. The factor storage device 11 stores auxiliary factors ε1, ε2...
... is sent to the frequency factor calculation device 16.

Further, the gates G1 to Gj are opened by the clock pulse CK1, and the frequency factor λ is stored from the initial frequency factor storage device 13.
1 to λj are sent to registers R1 to R•, respectively.

The frequency factor calculation device 16 includes j multipliers M1 to Mj and ■
The multiplier M1 consists of a storage device 11
The auxiliary factor ε1 sent from the register R1 is multiplied by the frequency factor λ1 which is the content of the register R1, and ε1·λ1 is sent to the adder Sa.

Similarly, the multipliers M2~M* have values ε2*λ2~εj*λj
is sent to the adder Sa.

Adder Sa adds the outputs from these multipliers M1 to M-, and uses the addition result as a frequency factor λj+1.

That is, the frequency factor λ・+l is calculated based on the previous (past) frequency factors λ1 to λ・,
This value λj+1 is sent to register Rj+1.

The amplitude variation factor storage device 12 stores partial tones f1(nT)1, f
2(nT)・・・・・・fm(nT)・・・・・・
...Amplitude variation factors α1, α2 of f M (nT)
・・・・・・・・・α□・・・・・・・・・αM is memorized, and the first time interval τ, that is, partial tone ft(nT
) is calculated, the amplitude variation factor α1 is output by a signal from the memory address control device 2.

Multiplier 17 calculates the frequency factor λ1, which is the content of register R1.
is multiplied by the amplitude variation factor α1 to output the value λ1·α1, and the coefficient unit 18 doubles this value.

That is, the parameter p4 is determined by the following calculation.

On the other hand, the multiplier 19 squares the amplitude variation factor α1, and the inverter 20 inverts the sign of this value.

That is, the parameter q1 is determined by the following calculation.

In the next time interval τ, that is, the time interval in which the partial f2 (nT) is calculated, the contents of the register group Rj+1 to R1 are shifted one by one (from right to left), and the contents of the registers R-R.
・・・・・・・・・The contents of R2, Ro are that J)
J-1 It frequency factor λ・+1, λj・・・・・・λ3
, λ2.

In response, the multiplier M1 multiplies the auxiliary factor ε1 sent from the auxiliary factor storage 11 by the frequency factor λ2, which is the content of the register R1, and adds the value ε1λ2 to the adder Sa.
Send to.

Similarly, the multipliers M2~M· have values ε2·λ3·ε··λ·
+1 is sent to adder Sa.

As a result, the adder Sa performs the following calculation and sends a new frequency factor λj+2 to the register R·+1.

On the other hand, the amplitude variation factor storage device 12 stores the amplitude variation factor α2
Output.

Multiplier 17 calculates the frequency factor λ2, which is the content of register R1.
is multiplied by the amplitude variation factor α2 to output a value ct2·λ2, and the coefficient unit 18 doubles this value.

That is, the parameter p2 is Find it using the following calculation.

The multiplier 19 squares the amplitude variation factor α2, and inverts the inverter 2.
0 inverts this value.

That is, the parameter q2 is determined by the following calculation.

The mth time interval τ, that is, the partial sample value, fm(n
At the time when T) is calculated, the contents of registers Rj+1 to R1 are emptied one by one, and register R...
・・・・・・The contents of R2 and R1 are respectively the frequency J) Wave number factor λ・+m−1, ・・・・・・・・・λ1+□,
λ□.

Accordingly, multiplier M1 has auxiliary factor ε1 and register R
The product is multiplied by the frequency factor λ□ which is the content of 0, and the value ε1
- Send λ□ to adder Sa.

Similarly, multipliers M2~Mj have values ε2・λ1+□~εj・
λ·+m−t is sent to the adder Sa.

As a result, the adder Sa performs the following operation and stores the frequency factor λj+□ in the register Rj
- Send to h.

On the other hand, the amplitude variation factor storage device 12 stores the amplitude variation factor Ctm
The multiplier 17 multiplies the frequency factor λ□, which is the content of the register R1, by the amplitude variation factor ctm, and the coefficient unit 18 receives this value and calculates the parameter pm by the following calculation.

-By data 20 Also, the multiplier 19 and the inverter The following calculation is performed to obtain the parameter (1m).

The Mth time interval τ, that is, the partial sample value f M (
At the time when nT) is calculated, the content of the register R1 becomes a frequency factor, the width variation factor storage device 12 outputs the amplitude variation factor αM, and the multiplier 17 and the coefficient unit 18 calculate the parameter pM by the following calculation.

In addition, the multiplier 19 and the inverter 2o The parameter (IM) is determined by the calculation.

In this way, in the time interval T, the parameters (pl, ql), (p2.q2)...
・・・・・・(pm, qm)・・・・・・・・・(pM
, (LM) are sequentially output, and the same operation is performed in the next time interval T.

These parameters (pm, (1m)) are input to the arithmetic unit 4 shown in FIG. 2, and the arithmetic unit 4 calculates the equation (5) to obtain partial tone sample values f1(nT), f2(nT), . . .
・・・・・・fm(nT)・・・・・・・・・fM(n
Find T).

In addition, in the above embodiment, the frequency factor λ, 1-4-1 to λ
Although M is obtained by a calculation mechanism, the amplitude variation factors α1 to α
Similarly to □, it may be possible to continue from the storage device.

Furthermore, although the amplitude variation factors α1 to Ctm are successively retrieved from the storage device, they may be replaced by a calculation mechanism similar to that used to obtain the frequency factors λj−h to λM, for example.

Also, the pulse width of clock pulses CK1 and CR2 is τ/
It is not limited to 2(-T/2M).

Next, the parameter pm calculated as above,
The effect of qm on musical tones will be explained.

The sample value of the partial at time nT is Calculated by

Solve this difference equation (and 01 solution is obtained.

However, A and θi are constants determined by initial values.

By the way, this is because the parameters pm and qm are shown by equations (15) and (16).

Therefore, as is clear from equations (20) and (22), j
The amplitude of 'm(nT) is determined only by the factor α□,
It can be seen that the frequency is determined by the factor λ□.

As a result, the parameters pm and q for forming the desired musical tone are
The setting of m becomes easy.

Next, in the embodiment of FIG. 4 of the present invention, the initial frequency factor λ1. λ2......λj is set arbitrarily, and the remaining frequency factors λ・+1 to λM are calculated using equations (8) and (1
1), (This is determined by 1 match, but the specific effect this shows will be explained using an example.

j-2, the initial frequency factor is 1, the auxiliary factor is 1, and the remaining frequency factors are expressed by equations (8), (11),
(Due to 1m years.

This synthesizes musical tones in which the frequencies of the partials are integral multiples of the frequency of the fundamental tone.

Various other frequency configurations can be obtained by appropriately selecting j initial values and auxiliary factors.

As is clear from the above explanation, by replacing the parameter pm-qm for musical tone (partial tone) formation with a frequency factor and an amplitude variation factor, each factor can be generated and controlled independently. Therefore, it is easy to set the frequency and amplitude variation of partials, and it is possible to create musical tones with a wide variety of partial tones. You can easily choose from a variety of options.

Furthermore, since the frequency factors are calculated sequentially based on their past values, only a small storage capacity is required to generate the parameters Pm and (1m).

That is, when storing j initial frequency factors, the storage capacity can be reduced by (M-j).

[Brief explanation of drawings]

Fig. 1 is a waveform diagram showing musical tones and partial tones, Fig. 2 is a block diagram showing an example of an electronic musical instrument to which this invention is applied, and Fig. 3 is a time diagram showing clock pulses used in the embodiment of this invention. FIG. 4 is a block diagram showing an embodiment of the present invention. DESCRIPTION OF SYMBOLS 1...Key number designation device, 2...Memory address control device, 11...Auxiliary factor storage device, 12...Amplitude variation factor storage device, 13 ...
...Initial frequency factor storage device, 01~G...
・Gate, R1 to Rj + 1...Register,
16... Frequency factor calculation device, Ml ~ Mj・
... Multiplier, Sa ... Adder, 17...
... Multiplier, 18 ... Coefficient unit, 19 ...
...multiplier, 20...inverter.

Claims (1)

[Claims] 1. M partials f (M is a positive integer) constituting a musical tone
The sample value of m (m-1, 2, . . . . ,...
...M) and qm(m-1,2, ...
...M) and then add them together to form a musical tone by adding the obtained partial sample values, and the following (a) and (b) ←→ Equipped with an electronic musical instrument. (b) M factors αm(m-1, 2,...
...A device that generates M). (b) M factors λrrl(m-1, 2,...) that determine the frequency of each partial corresponding to each partial f□ above
...A device that generates M). ←→ A device that calculates the parameters pm and qm by calculating the following equation from the amplitude variation factor α□ generated from the device (a) and the frequency factor λ□ generated from the device (b). 2. The electronic musical instrument according to claim 1, wherein the frequency factor λ□ is determined by recursive calculation.