JPH0830954B2 - Musical sound waveform generator - Google Patents

Description

TECHNICAL FIELD The present invention relates to a musical tone waveform generator.

[Conventional technology]

With the advancement of digital signal processing technology, as a first conventional example of an electronic musical instrument using the digital processing, not only a musical sound waveform having a simple characteristic but also a musical sound of a natural musical instrument, a sound of a human or a natural world (hereinafter referred to as a sound). , Collectively called natural sounds)
A PCM-type electronic musical instrument that can directly sample and memorize the sound and reproduce it at an arbitrary pitch has been realized.

On the other hand, as a second conventional example of an electronic musical instrument capable of digitally generating musical tone waveforms of various kinds of complicated characteristics,
There is an electronic musical instrument based on the FM system described in JP-A-54-33525 or JP-A-50-126406. This method is basically is intended to be _{e = A · sin {ω c} t + I (t) sinω m t} ··· (A) composed musical sound waveform of the waveform output e obtained by the calculation formula, the carrier frequency ω _c and the modulation wave frequency ω _m for modulating it are selected with an appropriate ratio, a modulation index I (t) that can change with time is set, and an amplitude coefficient that can also change with time is set. By setting A, it is possible to obtain a very unique synthesized sound or the like which has a complicated overtone characteristic and whose overtone characteristic can change with time.

Also, as a third conventional example that improves the FM system,
There is an electronic musical instrument described in JP-A-61-12279. In this method, a triangular wave operation is used instead of the sin operation of the expression (A), and a waveform output e obtained by an operation expression of e = A · T {α + I (t) T (θ)} (B) Is a musical tone waveform. Here, T (θ) is a triangular wave function generated by the modulated wave phase angle θ. Then, the carrier wave phase angle α and the modulated wave phase angle θ are advanced at an appropriate traveling speed ratio, and the modulation index I (t) and the amplitude coefficient A are set in the same manner as in the first conventional example, whereby the musical tone waveform is obtained. Can be synthesized.

[Problems to be Solved by the Invention]

Against the background of the above-mentioned conventional techniques, in recent years, there has been a demand for an electronic musical instrument to have a performance capable of dynamically producing a very unique musical tone peculiar to the electronic musical instrument to a natural sound.

However, the first conventional example of a PCM electronic musical instrument is:
Although he is very good at producing natural sounds themselves, he is not good at processing when processing natural sounds to produce individualized timbres.

That is, for example, when it is desired to continuously change from the original sound to a sine wave or the like, a sine wave is obtained by removing a harmonic component of the original sound with a digital filter or an analog filter or the like. Becomes relatively large, and if it is attempted to change the characteristic by a time function such as an envelope, it is necessary to store a filter coefficient corresponding to the characteristic of the filter in addition to natural sound data. On the other hand, the analog filter has a problem that it is difficult to obtain a desired characteristic and it is impossible to perform a time-division operation for producing a plurality of musical tones in parallel.

Further, in contrast to the above, when it is desired to continuously change from the original sound to a musical tone having a more complex overtone structure, a method of removing the overtone structure of the original sound with the above-described filter may not generate a new harmonic component. There is a problem that it is impossible.

Further, in the PCM electronic musical instrument, in order to save a storage capacity or to reproduce a continuous sound, a loop reproduction for repeatedly reproducing a specific waveform section of a natural sound from a memory is generally performed. However, if you simply perform loop playback,
A musical tone having the same characteristics is always reproduced in the waveform section, and even if the envelope of the amplitude is changed, the timbre itself does not change, resulting in a monotonous reproduced sound.
The PCM electronic musical instrument has a problem that it is difficult to add a tone color change during loop reproduction because it is difficult to process natural sound as described above.

On the other hand, for example, a musical tone of an actual musical instrument such as a piano includes harmonic components of a plurality of frequencies which are integral multiples of the fundamental wave component in addition to the fundamental frequency component based on the pitch frequency, and considerably higher harmonic components exist. Further, a non-integer multiple harmonic component may be included. Further, the proportion of each higher harmonic overtone, etc., differs depending on the type of musical instrument, and there are various overtone characteristics depending on the musical instrument. In this way, the existence of the overtone component peculiar to each musical instrument produces musical tones with rich sound quality. However, the electronic musical instrument based on the FM method, which is the second or third conventional method, is very good at manipulating the overtone composition of the musical tone to be produced, but as an output, it is unique to each instrument as described above. When it is desired to obtain a desired musical tone, it is difficult to optimally set the parameters.

That is, since the second conventional example is based on the modulation by the sine wave, the musical tone generated by the equation (A) has its frequency components concentrated in the low-order (low-frequency) overtone component, Even if the modulation index I (t) is set to a large value and the modulation is deeply performed, higher-order (higher frequency) overtone components do not appear well. Therefore, the second conventional example has a problem in that it is not possible to generate a musical tone having rich sound quality such as an actual musical tone, and the tone quality of the musical tone that can be generated is limited.

On the other hand, in the third conventional example based on the equation (B), since the modulation is basically based on the triangular wave containing many overtones, a tone having a higher order harmonic component as a frequency component is clearly present. Although it can be easily generated, when it is desired to obtain a desired musical sound as an output, the traveling speed ratio between the carrier wave phase angle α and the modulated wave phase angle θ in the equation (B) and the modulation index I ( It is difficult to optimally determine t) and the amplitude coefficient A. In addition to this, in the third conventional example, since the triangular wave is driven by a triangular wave, for example, in the process of gradually attenuating the tone after the tone starts, its amplitude decreases in order from the higher harmonic components. Eventually, there is a problem that a process in which only a single sine wave component corresponding to the pitch frequency is finally impossible cannot be realized.

An object of the present invention is to make it possible to faithfully produce natural sounds with a small circuit scale, to easily and continuously control the overtone component thereof, and to easily synthesize a musical sound such as a single sine wave, and The object is to enable loop reproduction and freely control the tone characteristic during loop reproduction.

[Means for solving the problem]

The present invention firstly has carrier signal generating means for generating a carrier signal. The means is means for inputting a carrier phase angle signal that repeats an operation in which the phase angle sequentially and linearly increases as time elapses during one cycle, converts the signal according to a constant function, and outputs the carrier signal. , ROM which receives the carrier phase angle signal as an address input.
The characteristics of the output carrier signal will be described later.

Next, there is provided a modulation signal generating means for generating a modulation signal. The means is, for example, a means for receiving the carrier phase angle signal as input, converting it according to a certain function, and outputting it as a modulation signal, and is constituted by a ROM or the like having the carrier wave phase angle signal as an address input. The characteristics of the output modulated signal will be described later.

Further, when the modulated signal is mixed with the carrier signal generated from the carrier signal generating means, the mixing ratio of the modulated signal to the carrier signal is controlled from 0 to an arbitrary mixing ratio, It has a mixing control means for outputting a mixed signal mixed with the modulation signal at the mixing ratio.
The means is, for example, a multiplier that multiplies the modulation signal output from the modulation signal generating means by a modulation index whose value can change between 0 and 1, an output signal of the multiplier and the carrier. It is an adder that adds the carrier signals generated from the signal generating means and outputs as a mixed signal. The mixing ratio may change with time as described below.

Further, it has a waveform output means for outputting a modulated tone waveform with the mixed signal output from the mixing control means as an input, the input and output having a predetermined functional relationship. The same means
For example, it is a decoder that converts the mixed signal according to the predetermined functional relationship and outputs it as a musical tone waveform. Alternatively, it is a ROM or the like that uses the mixed signal as an address input.

Along with the above configuration, the predetermined functional relationship and the carrier signal are such that when the mixing control unit controls the mixing ratio of the modulation signal to be 0, the musical tone waveform generated from the waveform output unit is It has a relationship that becomes a sine wave or a cosine wave.

Further, the carrier signal generating means and the modulation signal generating means, when the mixing ratio of the modulation signal is controlled by the mixing controlling means to be a predetermined mixing ratio, the desired waveform of the waveform of the waveform output means is generated. To generate a carrier signal and a modulated signal, and repeat the arbitrary signal section of the carrier signal and the modulated signal so that the arbitrary waveform section of the tone waveform is repeatedly output from the waveform output means. appear.

Then, the mixing control means temporally changes the mixing ratio in synchronization with the repeated generation operation.

In addition to the above respective configurations, there is provided an amplitude envelope control means for temporally changing the amplitude envelope characteristic of the musical tone waveform output from the waveform output means in synchronization with the repetitive generation operation. The means multiplies, for example, the tone waveform output from the waveform output means by an amplitude coefficient whose value can change with time, for example, between 0 and 1 according to a predetermined amplitude envelope function after the start of the repetitive generation operation. It is a multiplier. The means may be a means for normalizing the amplitude of the output of the waveform output means to match the desired tone waveform.

[Action]

 The operation of the present invention is as follows.

The tone waveform output from the waveform output means basically has a characteristic obtained by converting the carrier signal output from the carrier signal generating means in accordance with a predetermined functional relationship, and further, the mixing control means modulates the carrier signal into the carrier signal. By mixing the signals, the characteristic that the tone waveform is modulated by the modulation signal is added.

In this case, the relationship between the predetermined functional relationship in the waveform output unit and the carrier signal from the carrier signal generation unit, when the mixing control unit controls the mixing ratio of the modulation signal to be 0, The relationship is set such that the tone waveform generated from the waveform output means becomes a sine wave or a cosine wave. Accordingly, if the mixing ratio of the modulation signal is set to 0 in advance by the mixing control means, it is possible to generate a musical tone waveform consisting of only a sine wave or a cosine wave.

Further, the carrier signal generation means and the modulation signal generation means output a desired tone waveform from the waveform output means when the mixing control means controls the mixing ratio of the modulation signals to be 1, for example. Generate a carrier signal and a modulated signal as described. Accordingly, if the mixing ratio of the modulation signal is set to 1 in advance by the mixing control means, it is possible to obtain a desired musical tone waveform such as a musical tone of a natural musical instrument.

Further, the carrier signal generation means and the modulation signal generation means repeatedly generate an arbitrary signal section of the carrier signal and the modulation signal so that the arbitrary waveform section of the musical tone waveform is repeatedly output from the waveform output means. . This allows so-called loop reproduction.

Then, in this case, the mixing control means sets the mixing ratio to, for example, 1 before the start of the loop reproduction operation, and changes the mixing ratio with time after the start of the loop reproduction operation to approach 0, for example. Even during the loop reproduction, the frequency characteristic of the musical tone waveform can be gradually controlled so that the desired musical tone waveform is changed to a state including only a single sine wave component or a single cosine wave component. Alternatively, by changing the mixing ratio continuously to, for example, 1 or more, it is possible to control so that a unique musical tone having a more complex overtone configuration is generated from a desired musical sound waveform state. This makes it possible to generate a variety of musical tone waveforms only with the carrier signal and the modulation signal corresponding to an arbitrary waveform section, save the storage capacity of, for example, the carrier signal and the modulation signal, and perform loop playback during loop reproduction. The tone can be controlled so as not to become monotonous.

In addition to the above operation, the amplitude envelope control means controls the amplitude envelope characteristic of the musical tone waveform output from the waveform output means so as to be attenuated, for example, in time in synchronization with the loop reproduction operation, so that an actual musical instrument is produced. It is possible to realize a process in which the musical tone waveform is gradually attenuated after the start of sound generation like the musical tone of FIG.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

Description of Basic Principle of One Embodiment of Musical Sound Waveform Generator First, before explaining the loop reproducing operation directly related to the present invention, the basic principle of the present invention will be explained.

FIG. 1 is a basic principle configuration diagram of an embodiment of a musical tone waveform generating apparatus according to the present invention.

First, from the waveform data ROM 1, according to the address data Add (t) which increases with time, the cycle data F
(T), the phase difference data M (t), and the normalization coefficient K (section) having a constant value for each waveform section (described later) are read in synchronization with each other. First, the phase difference data M (t) [ra
d] is multiplied by the modulation index I in the multiplier 4 and then added with the period data F (t) [rad] in the adder 3 to obtain the phase angle data F (for modulating the sin wave (sine wave). t) + IM
(T) is obtained. The same data is input to the same ROM as an address signal for modulating and reading the sine wave from sin ROM 2 which is a ROM memory storing the sine wave data. The modulation output D (t) read from sin ROM 2 is
The multiplier 5 multiplies the normalized coefficient K (section) read from the waveform data ROM 1 for each waveform section, and then outputs as a waveform output OUT (t). Here, the maximum absolute value of the amplitude of the sin wave stored in sin ROM 2 is normalized to 1.

The operation of the musical tone waveform generator of FIG. 1 based on the above basic principle configuration will be described below.

First, the original waveform ORG (t) of the musical sound waveform of a natural musical instrument, etc.
As shown in FIG. 2 (a), the waveform is divided into waveform sections A to D, for example, at each cycle section (pitch cycle) of the fundamental wave.

And within each waveform section, 0 [rad] or more and 2π [rad]
As shown in FIG. 2 (b), the phase angle data that linearly increases with time t during
The cycle data F (t) [rad] read from the ROM 1 is used. Now, assuming that the modulation index I in FIG. 1 is 0, the period data F (t) itself is stored in sin ROM 2.
Read out the in wave with its phase angle specified linearly and output si
When n (F (t)) is obtained, the sin wave of one cycle corresponding to the phase angle of 0 to 2π [rad] is not modulated in each waveform section A to D as shown in FIG. 2 (c). Is read by. The waveform sections A to D do not have to correspond to the pitch cycle accurately, and particularly for a musical sound having a weak periodicity such as a percussion instrument sound, for example, from an appropriate zero-cross point (time when the amplitude is 0) to a zero-cross point. May be one waveform section.

Next, the phase difference data M (t) read from the waveform data ROM 1 of FIG. 1 is output from the adder 3 with the value of the modulation index I multiplied by the multiplier 4 being 1 as shown in FIG. )
Using the phase angle data F (t) + M (t) shown in
When the sin wave for one cycle stored in sin ROM 2 is modulated and read out, the waveform output as shown in Fig. 3 (a)
The data is such that the original waveform ORG (t) for one waveform section in which the maximum absolute value of the amplitude is normalized to 1 is read as OUT (t), and is shown in FIG. 3 (c).

Further, the normalization coefficient K (section) for each waveform section read from the waveform data ROM 1 of FIG. 1 is the absolute value of the amplitude of the sin wave stored in the sin ROM 2 of FIG. 1 as described above. Since the maximum value of is normalized to 1, it is a coefficient for returning to the amplitude of the final original waveform ORG (t) for each waveform section.

From the above relationship, OUT (t) = K (section) · sin (F (t) + IM)
(T))-(1) ORG (t) = K (section) -sin (F (t) + M (t))-
・ It turns out that there is a relationship of (2). As can be seen from these relationships, when it is desired to obtain the desired original waveform ORG (t) as the waveform output OUT (t) when the value of the modulation index I is 1, for each waveform section, the waveform section It is necessary to obtain the period data F (t), the phase difference data M (t), and the normalization coefficient K (section) corresponding to the original waveform ORG (t). How to obtain it will be described by taking the case of the waveform section A of the original waveform ORG (t) in FIG. 2 as an example.

First, regarding the method of deriving the periodic data F (t),
This has already been described with reference to FIG.

Next, in the waveform section A of FIG. 2 (a), the original waveform ORG
The maximum absolute value of the amplitude of (t) is set as the normalization coefficient K (section A). Then, by dividing each amplitude value of the original waveform ORG (t) in the same section by the normalization coefficient K (section), the amplitude is normalized within ± 1 as shown in FIG. 3 (a). To do. In FIG. 3 (a), the maximum values of the positive and negative absolute values are both 1, but the maximum value may be 1 in either one and may be 1 or less in the other.

Next, the normalized original waveform OR obtained in this way is
Using G (t), the phase difference data M (t) is obtained by the following processes (1) to (see FIG. 4).

First, for any time t _x within the waveform section A of the normalized original waveform ORG (t), the amplitude A _x of the normalized original waveform ORG (t) corresponding to the time t _x is obtained ( Fig. 4 (a)).

The maximum value of the absolute value of the amplitude on the sin wave is 1, obtains the phase angle p _x equal position with Motoma' amplitude A _x in the above (FIG. 4 (b)) In this case, the original waveform ORG (t) The position of the time t _{x in and} the position of the phase angle p _x in the sin wave are obtained so as to have substantially the same relationship. That is, for example, if the time t _x is within the position of about 1/4 from the beginning of the waveform section A, the phase angle is also determined in the vicinity of 0 to π / 2 within about 1/4 from the beginning.

And phase angle p _x which is obtained by the above, by using the period data F (t) of waveform segment A that is obtained in advance, p _x -F (t _x) corresponding to the time t _x as the phase difference data M (t _x) Is calculated (FIG. 4 (c)).

By changing the time t _x in the entire waveform section A,
The above process is repeated to obtain the phase difference data M (t) corresponding to each time t of the waveform section A.

The above processing is repeated for each waveform section A to D of the original waveform ORG (t) in FIG. 2A, and the cycle data F (t), the phase difference data M (t) and The normalization coefficient K (section) is stored in the waveform data ROM 1 of FIG.

By setting the value of the modulation index I multiplied by the multiplier 4 of FIG. 1 together with each of the above data to 1 and performing the modulation operation, the waveform output OUT (t) of FIG. 1 is obtained as the waveform output OUT (t) of FIG.
The original waveform ORG (t) can be obtained.

Next, in FIG. 1, by changing the value of the modulation index I multiplied by the multiplier 4, variously modulated waveform output OUT (t) can be obtained.

First, if the modulation index I = 0 is set, an unmodulated sin wave can be obtained in each waveform section as already shown in FIG. 2 (c).

Also, by changing the value of the modulation index I to 1.0, 1.5, 2.0, the modulation output D from the sin ROM 2 in FIG.
As (t), it is possible to obtain a deeply modulated waveform as shown in FIGS. 5 (a), (b), and (c).

As described above, by changing the value of the modulation index I, various modulation waveforms can be obtained from a sine wave to a deeply modulated waveform with the original waveform as the center.

Further, by continuously changing the modulation index I from the start of sound generation to the mute, for example, a waveform output OUT (t) that changes from a deep modulation state to a sin wave with attenuation of a musical sound is obtained. It is also possible.

In addition to the basic principle as described above, as an operation particularly related to the present invention, by repeating the address data Add (t) of FIG. 1 in a predetermined section, a predetermined waveform section is repeatedly output for the waveform output OUT (t). The so-called loop playback is performed. Then, in this case, by changing the value of the modulation index I in the loop reproduction section with time,
This makes it possible to perform loop playback rich in changes.
The specific configuration will be described below.

Description of Specific Configuration of One Embodiment of Musical Sound Waveform Generator FIG. 6A is a specific configuration diagram of one embodiment of the musical sound wave generator based on the basic principle of FIG. In the figure, the parts with the same numbers as in the principle configuration of FIG. 1 have the same functions as in FIG. The waveform data ROM 1 in FIG. 1 is composed of two ROMs, FM ROM1 ₁ and K ROM1 ₂ in FIG. 6A.

In FIG. 6A, address data ADD for reading the period data F (t) and the phase difference data M (t) from the FM ROM ₁₁
# 1 is output from the generation circuit of FIG. 6B described later.

The phase difference data M (t) output from the FM ROM1 ₁ based on the address data Add (t) is multiplied by the modulation index I latched in the latch 16 in the multiplier 4, and the FM ROM1 ₁ is added in the adder 3. Period data F output from
(T) is added. The phase angle data F (t) + IM · (t) thus obtained is input to the same ROM as the read address of sin ROM 2. The modulation index I is output from the generation circuit shown in FIG.
It is sequentially latched in the latch 16 in synchronization with the rising edge of.

As a result, the modulation output D from sin ROM 2 to multiplier 5
(T) is output.

On the other hand, the section identification data IB output from the FM ROM ₁₁ is
The clock CLK # 1 is inverted by the inverter 11 and input to the D flip-flop (F / F, the same applies hereinafter) 10 that operates in synchronization with the rising edge of the clock, and the exclusive OR circuit (EOR, apply the same below) 12 To the first input of. The positive logic output Q of the F / F10 is input to the second input of the EOR12. With the above circuit configuration, when there is a change in the value of the section identification data IB sequentially output from the FM ROM ₁₁ ,
The output of R12 becomes logic "1".

The address input to the K ROM1 ₂ address data ADD #
2 is an accumulator consisting of an adder 13, a selector 14 and a latch 15 and uses the head address b as an initial value for the above EOR12.
Are sequentially accumulated by one address each time the output of the memory becomes "1". The head address b is output from a control unit (not shown), and is selected by the selector 14 while the start pulse STRT is at the high level and latched as the initial value of the accumulated value.
Given to 15. If the start pulse STRT is low level, the output of the adder 13 is selected and the accumulation operation is executed.

Thus, K ROM 1 ₂ to given address data ADD # 2, K ROM 1 to ₂ from the multiplier 5 the normalization factor K (section)
Is output.

In the multiplier 5, the modulation output D output from the sin ROM 2
(T) is multiplied by the normalization coefficient K (section), and the multiplication result is further multiplied by the envelope value generated from the envelope generator 17 in the multiplier 18.

The multiplication result is latched in synchronization with the rising edge of the clock obtained by inverting the clock CLK # 1 by the inverter 20.
It is latched by 19 and is output as the waveform output OUT (t).

Next, FIG. 6B shows a circuit configuration for generating the address data ADD # 1 shown in FIG. 6A.

First, basically, the address data ADD # 1 is stored in the accumulator including the adder 7, the selector 8 and the latch 9,
With the initial address a as the initial value, the pitch data d latched in the latch 6 are sequentially accumulated at the address intervals in synchronization with the rising edge of the clock CLK # 1 input to the latch 9. In this case, the head address a is particularly output from a control unit (not shown), and is selected by the selectors 25 and 8 (via the OR circuit 27 in the selector 8) while the start pulse STRT is at the high level, and the initial accumulated value is selected. It is given to the latch 9 as a value. If start pulse STRT is low level, adder 7
Select the output of and execute the accumulation operation. The pitch data d is output from a controller (not shown) and latched in the latch 6 in synchronization with the rising edge of the start pulse STRT.

On the other hand, in the latch 23, in synchronization with the rising edge of the start pulse STRT, the start address LS of the loop basic section (described later in FIG. 11 and the like) output from a control unit (not shown) is latched. In addition, the latch 21 latches the address LE, which is output from the control unit at the same timing and is +1 to the final address of the loop basic section.

Then, the subtractor 22 subtracts the value of the address LE latched in the latch 21 from the accumulated value output from the adder 7, and the result of the subtraction is decoded by the decoder 26.

The output of the decoder 26 becomes high level when the subtraction result becomes 0 or a negative value, that is, when the accumulated value becomes equal to or exceeds the address LE. As a result, the selector 8 selects the output of the selector 25 via the OR circuit 27, and the selector 25 selects the output of the adder 24 if the start pulse STRT is at the low level.

At this time, in the adder 24, the subtraction result of the subtractor 22, that is, the amount of the accumulated value from the adder 7 exceeding the address LE is added to the start address LS of the loop basic section latched by the latch 23, It is output to the selector 25. Therefore,
The latch 9 stores pitch data d from the end of the loop basic section.
The address folded back to the beginning with the width of is latched.

After that, the addresses of the loop basic section are accumulated, and when the address exceeds the address LE, the loop basic section is returned to the beginning again. In this way, the address data AD of the basic loop section is
D # 1 is repeatedly output.

Subsequently, FIG. 6C shows a circuit configuration for generating the modulation index I of FIG. 6A.

First, before starting the performance, the performer gives a control signal SEL to the selector 33 in advance from a control unit (not shown) so that I
Select ROM 31 or LFO (low frequency oscillator, same hereafter) 32.

Then, at the timing when the start pulse STRT rises to a high level before the performance starts, the RS-flip-flop (F / F
(F, the same applies hereinafter) 28 is reset, and its positive logic output Q becomes low level. As a result, the selector 29 selects the value I and outputs it as the modulation index I.

Subsequently, the F / F 28 is set at the timing when the output of the decoder 26 in FIG. 6B becomes high level, that is, the timing when the loop reproduction operation is started, and the positive logic output Q thereof becomes high level. After that, the selector 29 selects the output of the selector 33. At the same time, the address counter
30 or LFO32 is activated.

Now, in the mode in which the selector 33 selects the output of the I ROM 31, the start address x in which the modulation index data to be read from the I ROM 31 is stored in advance in the address counter 30 is output from a control unit (not shown). It is set. Then, based on the start instruction from the positive logic output Q from the F / F 28, the clock CLK # 1 starts from the above-mentioned start address x.
The address is incremented in sequence in synchronization with the rising edge of
Supply to 31. As a result, the modulation index data corresponding to the above address is read from the I ROM 31, and the selector 33
And is output as the modulation index I via the selector 29.
Therefore, the modulation index I has a value of 1 before the start of the loop reproduction operation.
Then, after the start of the loop reproduction operation, the modulation index data is output from the I ROM 31.

On the other hand, in the mode in which the selector 33 selects the output of the LFO 32, the LFO 32 starts operating based on the start instruction by the positive logic output Q from the F / F 28, which causes the LFO 32 to synchronize with the rising edge of the clock CLK # 1. Then, the LFO data is sequentially output, and is output as the modulation index I via the selector 33 and the selector 29. Therefore, the modulation index I has a value of 1 before the loop reproduction operation starts, and outputs the value of the LFO data output from the LFO 32 after the loop reproduction operation starts.

In the configurations shown in FIGS. 6A, 6B, and 6C, the clocks CLK # 1 and CLK # 2 and the start pulse STRT are output from a controller (not shown).

Next, FIG. 7 shows the structure of the data stored in the FM ROM ₁₁ of FIG. 6A.

In the figure, one set of waveform data from the start of sound generation to mute is stored in order from the head address a, and one address has one cycle data F (t), one phase difference data M (t) and One-bit section identification data IB is stored as a set. In this case, as the address progresses, the data of each sampling point of each waveform section A, B, C, D, ... Of FIG. As the waveform section changes to A, B, C, D,..., The value of each address of the section identification data IB becomes “0” in section A, “1” in section B, and “1” in section C. 0 ”,“ 1 ”in section D,
.., Etc., alternately in units of sections. The section identification data IB is data for identifying the boundary of the section and updating the address specified in the KROM ₁₂ as described later.

Next, FIG. 8 shows the structure of data stored in the K ROM ₁₂ of FIG. 6A.

In the figure, one normalization coefficient K (section) is stored in order from the head address b corresponding to each waveform section A, B, C, D, ...

 The operation of the embodiment having the above configuration will be described below.

First, the normal reproduction operation before entering the loop reproduction operation will be described with reference to the operation timing chart of FIG.

At the start of sound generation, a start pulse STRT output from a control unit (not simply shown) (hereinafter simply referred to as a control unit).
However, at the timing of t1 in FIG. 9, it rises to the logic "1" (hereinafter simply referred to as "1". The same applies to the logic "0"), and immediately after that, the clock CLK # 1 becomes "1" t2. The sounding operation is started from the timing. In this case, the cycle of the clock CLK # 1 corresponds to the sampling cycle of the tone generation, and the clock CLK # 2 generated from the control unit has the same cycle as the clock CLK # 1 and 1/4 cycle from the same clock. It is a clock delayed by a minute. The following operation is controlled according to the above two clocks CLK # 1 and CLK # 2. In addition, the start pulse STRT maintains “1” for one cycle from the time when the clock CLK # 1 falls to “0” at the start of sound generation to the next time it falls to “0”. Maintain "0" until pronunciation starts.

First, the description will be centered on FIGS. 6A and 6B. At the timing t1 when the start pulse STRT from the control unit rises to logic "1" as shown in FIG. 9 (a), the pitch data d from the control unit is latched in the latch 6 as shown in FIG. 9 (c). It

Then, the timing t1 when the start pulse STRT is "1"
During t4, the selectors 25 and 8 select the head address a from the control unit, and this head address a is the clock CLK # 1.
Is latched by the latch 9 at the timing t2 when the signal rises to "1", and the initial value of the address data ADD # 1 is determined as shown in FIG. 9 (d).

This allows the FM ROM1 ₁ after a slight delay from t2
The period data F (t), the phase difference data M (t), and the section identification data IB (see FIG. 7) at the leading address a are read as shown in FIGS. 9 (e), (k), and (l). .

After that, the address data ADD # 1 output from the latch 9 is fed back to the adder 7, and the pitch data d set in the latch 6 is sequentially accumulated. Here, the above accumulated value is sequentially latched in the latch 9 via the selector 8 at each timing t5, t8, t11, t14, t17, etc. of FIG. 9 (d) when the clock CLK # 1 rises to "1". And specified as new address data ADD # 1, FM ROM1 ₁
The period data F (t), the phase difference data M (t) and the section identification data IB (see FIG. 7) at the corresponding addresses above are read out as shown in FIGS. 9 (e), (k) and (l). Be done.
Since the start pulse STRT falls to "0" at t4, the selector 8 selects the output of the adder 7 after t4.

In this case, the musical tone generation start instruction is given by the player depressing one of the keys on the keyboard, which is not particularly shown, and if the key pressed at that time is a high-pitched key. A large value of the pitch data d is latched in the latch 6 from the control section. As a result, the address width skipped on the FM ROM ₁₁ becomes large, and the waveform output with a high pitch cycle
OUT (t) is obtained. Conversely, for example, when the lowest tone key is pressed, the value 1 is latched as the pitch data d, whereby each data is read one by _one on the FM ROM1 ₁ and the waveform output OUT of the lowest pitch period is output. (T) is obtained.

Of the data read from the FM ROM ₁₁ as described above, the phase difference data M (t) is input to the multiplier 4.
Now, after the start pulse STRT rises to "1", the clock CLK # 1 rises to "1" at the timings t2, t5, t8, t11, t14, t17, etc. of FIG.
The modulation index I having a constant value of 1 is sequentially set in the latch 16 from the selector 29 in FIG. This is because the F / F 28 in FIG. 6C is reset by the rising of the start pulse STRT and the positive logic output Q becomes low level. Therefore, in the multiplier 4, the phase difference data M (t) is multiplied by the modulation index I. This output is input to the adder 3, where the cycle data F (t) output from the FM ROM1 ₁
Is added to obtain phase angle data F (t) + IM · (t).

By the above phase angle data F (t) + IMM (t), sin
The ROM 2 is accessed to output the cycle data F (t) and the phase difference data M (t) (each timing t2, t5, t8, t11, t14, t17, etc. in FIGS. 9 (k) and (l)). Immediately after), the modulation output D (t) is output from sin ROM 2 as shown in FIG. 9 (n).

On the other hand, in FIG. 6A, the start pulse STRT is “1”.
Between the timings t1 and t4, the selector 14 selects the start address b from the control unit, and this start address b is latched at the timing t3 when the clock CLK # 2 rises to "1".
Latched to 5, address data A as shown in Fig. 9 (i)
The initial value of DD # 2 is set.

This allows K ROM1 ₂ after a short delay from t3.
The normalization coefficient K (section A) of the head address b of is read as shown in FIG. 9 (j). After that, the address data ADD # 2 output from the latch 15 is fed back to the adder 13, and the logical output value "0" or "1" of the EOR 12 is sequentially accumulated. In this case, the accumulated value is the timing t6, t in FIG. 9 (i) when the clock CLK # 2 rises to "1".
At 9, t10, t12, t15, t18, etc., new address data AD is sequentially latched by the latch 15 via the selector 14.
D # is specified as 2, K ROM 1 ₂ on the corresponding normalized coefficient address K (interval) is read as FIG. 9 (j). Here, since the start pulse STRT falls to "0" at t4, the selector 14 makes the adder after t4.
Select 13 outputs.

In parallel with the above operation, each timing t2, t5, t8, t1 in FIG. 9 (b) at which the clock CLK # 1 rises to "1".
After a slight delay from 1, t14, t17, etc., Fig. 9 (e)
Section identification data IB from FM ROM 1 ₁ are sequentially outputted as. This data corresponds to each timing t4, t7, t10, t13, t1 in FIG. 9 (b) when the clock CLK # 1 falls to "0".
At 6, t19, etc., it is set to F / F10 and its positive logic output Q is sequentially determined as shown in FIG. 9 (f). Then, the exclusive logical sum of the section identification data IB from the FM ROM 1 ₁ and the positive logic output Q is calculated by EOR12.

Here, the section identification data IB output from the FM ROM ₁₁
Since "0" or "1" is alternately stored in each waveform section unit as shown in FIG. 7, every time the waveform section changes from A to B, B to C, C to D, etc. To "1", "1" to "0", and "0" to "1". Therefore, the output of the EOR12 becomes "1" only when the above waveform section changes, and becomes "0" at other timings. In the example of FIG. 9, the output of the EOR12 becomes "1" only while the clock CLK # 1 at t16 is "1" from immediately after t14 when the waveform section A is changed to B. From this, only within the above timing, 1 is accumulated in the value b of the address data ADD # 2 from the latch 15 by the adder 13, and this accumulated value b +
1 is latched in the latch 15 as new address data ADD # 2 at the timing t15 in FIG. 9 (h) when the clock CLK # 2 rises to "1". Therefore, after the timing t15 when the clock CLK # 2 rises to "1", the normalization coefficient K (section B) on the KROM1 ₂ based on the value b + 1 of the new address data ADD # 2 is shown in FIG. 9 (j). Is read out.

On the other hand, at other timings when the output of EOR12 is “0”, the adder 13 does not perform the accumulation operation, so the latch 15
, The same address data ADD # 2 as before one timing is latched. Therefore, the timing t6 in FIG. 9 (h),
In t9, t12, t18, etc., 1 timing as before normalization factor K (interval) is read as the K ROM 1 ₂ of FIG. 9 (j).

Thus, the modulation output D based on the period data F (t) and the phase difference data M (t) of the waveform section A from the FM ROM _{11 is obtained.}
At the timing when (t) is read from sin ROM 2, the normalization coefficient K corresponding to the waveform section A from K ROM1 ₂ is read.
(Section A) is read out, and in the same way, in waveform section B, K ROM1
_The normalization coefficient K (section B) corresponding to the waveform section B is read from _2, and the modulation output D of each waveform section is read.
Corresponding to (t) is output, the normalization factor K of the waveform section (section) is read from the K ROM 1 _2.

As described above, the clock CLK # 1 shown in FIG.
Rises to "1" at timings t2, t5, t8, t11, t
After a slight delay from 14, t17, etc., the sin ROM 2 outputs the modulation output D (t) as shown in FIG. 9 (n), and in parallel with this, the clock CLK # of FIG. 9 (h). After a slight delay from each timing t3, t6, t9, t12, t15, t18, etc. when 2 rises to "1", K ROM1 ₂ outputs the normalization coefficient K (section) as shown in Fig. 9 (j). To be done. Then, the modulation output D (t) at each of these timings and the normalization coefficient K
(Section) is multiplied by the multiplier 5 and then further multiplied
At each timing t4, t7, t10, t13, t1 at which the clock is multiplied by the envelope data from the envelope generator 17 at 18 and the clock CLK # 1 of FIG. 9 (b) falls to "0".
At 6 and t19, the waveform output OUT (t) is latched by the latch 19 as shown in FIG. 9 (o) and is determined at each timing. Here, the envelope generator 17 and the multiplier 18
Although not shown in FIG. 1, it can be operated when it is desired to add an envelope other than the original waveform to the waveform output OUT (t). For example, amplitude control during loop reproduction is performed as described later.

With the operation described above, the same operation as that of the musical tone waveform generator of FIG. 1 shown in FIGS. 2 to 5 can be realized.

Here, the overall operation timing including the above operation based on FIG. 9 is shown in FIG. At t1, after the start pulse STRT rises as shown in FIG. 7B, every time the clock CLK # 1 of FIG.
The address data ADD # 1 output from the generation circuit of FIG. 6B to the FM ROM ₁₁ of FIG. 6A gradually increases as shown in FIG. 10 (e) (corresponding to FIG. 9 (d)). This allows FM R
OM1 from _1, FIG. 10 (f), the cycle data F (t) and the phase difference data M (t) as shown in (g), the clock CLK #
In synchronization with the rising edge of 1, the contents of each address in FIG. 7 are sequentially output (strictly, as shown in FIGS. 9 (k) and 9 (l), they are slightly delayed from the rising edge of the clock CLK # 1). Here, each waveform section W ₁ , W ₂ , W ₃ , in FIG.
... is the waveform section A, B of FIG. 2 (a) or FIG.
It corresponds to C, D, .... And FIG. 10 (f)
2 (b), the phase data of the cycle data F (t) has a phase angle of 0 to 2π for each waveform section W ₁ , W ₂ , W ₃ , ...
The operation of increasing linearly is repeated until.

Each clock CL in each waveform section W ₁ , W ₂ , W ₃ , ...
At the rising edge of K # 1, the value of the modulation index I by which the phase difference data M (t) is multiplied by the multiplier 4 in FIG. 6A is 1, so the phase angle data F (t )
+ I · M (t) becomes F (t) + M (t), and as shown in Fig. 3 (a) or the above formula (2), sin of Fig. 6A
The waveform output OUT (t) output from the latch 19 from the ROM 2 via the multipliers 5 and 18 is the desired original waveform ORG (t).

Description of Loop Reproduction Operation In this case, after performing a certain waveform section, the player may want to perform loop reproduction in which the last waveform section is repeatedly reproduced.

When loop reproduction is simply performed, the waveform output OUT (t) is roughly as shown in FIG. 11 (a). That is, the waveform section W _n
By performing the loop reproduction with the loop basic section as the loop basic section, the reproduction in which the waveform of the waveform section W _n is stored is performed as shown in FIG. If this is reproduced as it is, a waveform output OUT (t) of a continuous sound system such as an organ can be obtained.

Here, there is a case where it is desired to attenuate the waveform output while the loop reproduction is being performed. For that purpose, the envelope generator 17 of FIG. 6A outputs the envelope data as shown in FIG. 11 (b) and the multiplier 18 multiplies it. By the 11th
The waveform output OUT (t) in FIG. 7A is converted as shown in FIG.

This makes it possible to generate a sound with a damped sound system,
However, in this case, since the musical tone characteristics of the loop playback section of are repeated the musical tone characteristics of the loop basic section W _n ,
The output is monotonous as a whole.

Therefore, in the present embodiment, the modulation index I multiplied by the phase difference data M (t) by the multiplier 4 of FIG. 6A in the loop reproduction section is set to the value 1 before the loop reproduction as shown in FIG. 11 (d). It is controlled so that it is gradually attenuated. The modulation index I is shown in Fig. 2 (c).
Alternatively, as described with reference to FIG. 5, etc., the original waveform ORG (t) changes to a sin wave as the value approaches 1 to 0. Therefore, by controlling the value of the modulation index I as shown in FIG. 11 (d), in the loop reproduction section, like natural sound,
Waveform output O that changes to sin wave as the amplitude attenuates
UT (t) can be obtained.

Also, for example, when the continuous sound waveform output OUT (t) is output by loop reproduction as shown in FIG. 11 (a),
In order to avoid a monotonous output, it is possible to add a change to the tone color by swinging the value of the modulation index I at a low frequency as shown in FIG. 11 (e) in the loop reproduction section.

The specific operation of the embodiment of FIGS. 6A, 6B and 6C for realizing the above operation will be described below.

First, when the performer wants to perform loop playback with the waveform section W _{n of} FIG. 10 as the basic loop section,
Start address LS of the basic loop section W _n for FM ROM1 ₁
Is latched in the latch 23 from a control unit (not shown) in synchronization with the rising edge of the start pulse STRT. At the same timing, the latch 21 latches the address LE obtained by adding 1 to the final address of the loop basic section from the control unit.

After the above operation, as already explained, address data AD
D # 1 is an adder 7, a selector 8 and a latch 9 shown in FIG. 6B.
In the accumulating section consisting of, the clock CLK input to the latch 9 after the timing t1 at the address interval of the pitch data d latched in the latch 6 with the initial address a as the initial value.
In synchronism with the rising edge of # 1, they are sequentially accumulated as shown in FIG.

In this case, the LE value latched in the latch 21 is subtracted from the accumulated value output from the adder 7 in the subtractor 22 shown in FIG. 6B. Then, in FIG. 10 (e), until the address data ADD # 1 reaches the address LE, the subtraction result becomes a positive value, and the output of the decoder 26 maintains the low level. As a result, the selector 8 selects the output from the adder 7 and performs the normal accumulation operation described above.

Then, at timing t15 of FIG. 10, the final address of the output loop basic intervals W _n latch 9 in Figure 6B (LE
−1) or approaching, and when the output accumulated by the adder 7 becomes equal to or exceeds the address LE, the subtraction result of the subtractor 22 becomes 0 or a negative value, and the decoder 26 The output becomes high level at the timing t16 immediately after the timing t15 as shown in FIG. 10 (c). As a result, the selector 8 selects the output of the selector 25, and
25 selects the output of the adder 24 because the start pulse STRT is at low level at the timing t16 in FIG.

At this time, in the adder 24, the subtraction result of the subtractor 22, that is, the amount of the accumulated value from the adder 7 exceeding the address LE is added to the start address LS of the loop basic section latched by the latch 23, It is output to the selector 25. Therefore,
At the timing t17 at which the clock CLK # 1 rises immediately after t16, the latch 9 latches the address folded back to the beginning with the width of the pitch data d from the end of the basic loop section. As a result, the address data ADD # 1 becomes the first
As shown in Fig. 0 (e), the loop is returned to the beginning of the basic interval.
The return destination address differs depending on the value of the pitch data d. Also, since the output of the adder 7 falls below the address LE due to the folding back at t17, the output of the subtractor 22 becomes a positive value again, and the output of the decoder 26 becomes as shown in FIG. 10 (c) immediately after t17. Return to low level.

After that, the addresses of the loop basic section are accumulated, and when the address exceeds the address LE, it is returned to the beginning of the loop basic section again at the timing t18. In this way, the address data ADD # 1 of the loop basic section is repeatedly output (t19, t20, etc. in FIG. 10 (e)).

On the other hand, in the circuit for generating the modulation index I in FIG. 6C, the F / F is set until the address data ADD # 1 exceeds the address LE.
Since the positive logic output Q of 28 is maintained at the low level, the selector 9 outputs the value 1 as the modulation index I. And
The F / F 28 is set at the timing t16 when the output of the decoder 26 in FIG. 6B becomes high level, and the positive logic output Q thereof becomes high level. After that, the selector 29 selects the output of the selector 33. At the same time, the address counter
30 or LFO32 is activated.

Now, in the mode in which the selector 33 selects the output of the I ROM 31, the start address x in which the modulation index data to be read from the I ROM 31 is stored in advance in the address counter 30 is output from a control unit (not shown). It is set. Then, based on the start instruction from the positive logic output Q from the F / F 28, the clock CLK # 1 starts from the above-mentioned start address x.
The address is incremented in sequence in synchronization with the rising edge of
Supply to 31. As a result, the modulation index data corresponding to the above address is read from the I ROM 31, and the selector 33
And is output as the modulation index I via the selector 29.
Therefore, the modulation index I has a value of 1 before the start of the loop reproduction operation.
Then, after the start of the loop reproduction operation, the modulation index data is output from the I ROM 31. This modulation index data is the 6th A at each timing when the clock CLK # 1 rises.
Latched in latch 16 in the figure, which results in timing t17
After that, the value of the modulation index I has a characteristic of being attenuated as shown in FIG. As a result, the characteristic shown in FIG. 11 (d) is realized.

As described above, first, the address data ADD # 1 generated in FIG. 6B accesses the FM ROM1 ₁ in FIG. 6A, whereby the period data F (t) and the phase difference data in the loop basic section W _n are M (t) is read as shown in FIGS. 10 (f) and (g), and the modulation index I is latched in the latch 16 sequentially as shown in FIG. 10 (h). Therefore, the phase angle data F (t) + I · M (t) generated by this is sin
By accessing the ROM 2, a modulated output D (t) having a waveform characteristic that gradually changes from the waveform characteristic of the loop basic section W _{n to} the characteristic of the sin wave is generated.

On the other hand, in FIG. 6C, in the mode in which the selector 33 selects the output of the LFO 32, the LFO 32 starts operating based on the activation instruction by the positive logic output Q from the F / F 28, which causes the LFO 32 to operate with the clock CLK # LF in sync with the rising edge of 1
O data is sequentially output, and is output as the modulation index I via the selector 33 and the selector 29. Therefore, the modulation index I has a value of 1 before the loop reproduction operation starts, and outputs the value of the LFO data output from the LFO 32 after the loop reproduction operation starts. As a result, the characteristic of the modulation index I shown in FIG. 11 (e), which is not particularly shown in FIG. 10, is realized. This allows
From sin ROM 2 in FIG. 6A, a modulation output D (t) having a swing characteristic is obtained centering on the waveform characteristic of the loop basic section W _n .

During the above loop playback operation, the loop basic section W _n
As shown in FIG. 7, the section identification data IB output from the FM ROM _{11 of} FIG. 6A takes the same value of either "0" or "1". Therefore, as long as the address within the loop basic section is designated, the output of EOR12 remains low level as described above. Therefore, the address data ADD # 2 output from the latch 15 maintains the same value, whereby the K ROM1 ₂ continues to output the normalization coefficient K (section W _n ) corresponding to the loop basic section W _n. . Therefore, from the multiplier 5 in FIG. 6A, a waveform output whose amplitude is correctly normalized corresponding to the loop basic section W _n can be obtained.

In addition to the above operation, the envelope generator of FIG. 6A outputs the envelope data having the characteristics shown in FIG. 11 (b) and the multiplier 18 multiplies the amplitude data so that the amplitude of the natural sound becomes like that of natural sound. Si as it decays
It is possible to obtain a waveform output OUT (t) that changes to n waves.

Another Mode in Specific Configuration of One Embodiment of Musical Sound Waveform Generator In FIGS. 6A and 6B, one of the waveform sections defined in FIG. However, the circuit configuration is not limited to this, and loop reproduction over a plurality of waveform sections, loop reproduction in an arbitrary section unrelated to the waveform section, and the like may be adopted.

Further, before the start of the loop reproduction operation, the original waveform is reproduced with the value of the modulation index I set to 1. However, naturally, the value of the modulation index I may be changed to another value also in this section. Similarly, after the loop playback operation is started, the I ROM in Fig. 6C is
Although the characteristic of the modulation index I output from 31 is the characteristic of being attenuated as shown in FIG. 10 (h), it may be set to have another characteristic.

On the other hand, the one corresponding to the waveform data ROM 1 in FIG.
Although it is composed of two ROMs, M ROM1 ₁ and K ROM1 ₂ , it may be composed of one ROM if the addressing can be complicated a little.

Further, in FM ROM 1 _1, period data F (t) is the seventh
As shown in the figure, the value for each sampling point is stored, but within one waveform section, the periodic data F (t)
2 is a linear characteristic in which the phase angle changes from 0 to 2π [rad] in each waveform section width as shown in FIG. 2B, so that if the calculation time has a margin, the periodic data F is obtained at each sampling point.
A storage capacity can be saved by calculating and outputting (t).

Note that in FIG. 6A, the waveform output OUT for one musical tone
Although (t) has been realized, it is also possible to output a plurality of musical sound waveforms in parallel by operating each unit in a time division manner.

Another Mode of Principle Configuration of One Embodiment of Musical Sound Waveform Generator In the principle configuration of one embodiment of the music sound waveform generator of FIG. 1, when the modulation index I = 0, OUT ( t) = K (section) .sin (F (t)) .. (3), and the cycle data F (t) is data that increases linearly within each waveform section as shown in FIG. 2 (b). Since the normalization coefficient K (section) is constant in each waveform section,
A sine wave was obtained as the waveform output OUT (t). When the modulation index I = 1, ORG (t) = K (section) · sin (F (t) + M (t)) ·
The original waveform ORG (t) was obtained as (2). By changing the value of the modulation index I between 0 and 1, the waveform can be continuously changed from the sine wave to the original waveform. If the value of the modulation index I is set to 1 or more, the waveform can be continuously changed from the original waveform to the further modulated waveform.

As described above, in this embodiment, at least the period data F (t) that linearly changes in each waveform section and the phase difference data M (t) that is difference data from the data are used.
The feature is that it can output sin wave and original waveform. Therefore, if the sine wave can be output when I = 0 and the original waveform can be output when I = 1, it is not necessary to stick to the equations (1) and (2), and OUT (t) = K ( Section) ・ f (g (F (t)) + I ・ M
(T)) (4) Any embodiment may be used as long as it is an embodiment that realizes the calculation having the relation of.

For example, in the above formula (4), f is the input, and f (α) = (2 / π) α ··· (0 ≦ α ≦ π / 2) f (α) =-1+ (2 / π) ( 3π / 2-α) ··· (π / 2 ≦ α ≦ 3π / 2) f (α) =-1+ (2 / π) (α-3π / 2) ··· (3π / 2 ≦ α ≦ 2π) · .. (5) is defined as a triangular wave function, and g is set with β as an input, g (β) = (π / 2) sin β. (0 ≦ β ≦ π / 2) g (β) = π- (π / 2) sin β ·· (π / 2 ≦ β ≦ 3π / 2) g (β) = 2π + (π / 2) sin β ·· (3π / 2 ≦ β ≦ 2π) ・ ・ ・ ( If the value of the modulation index I is 0, that is, no modulation, by substituting the equations (5) and (6) into the above (4), f (g (β) ) = K (section) ・ f ((π / 2) sin β) = K (section) ・ (2 / π) (π / 2) sin β = K (section) ・ sin β ・ ・ ((0 ≦ β ≦ π / 2) f (g (β)) = (Section) * f (([pi]-([pi] / 2) * sin [beta])) = K (section) * (-1+ (2 / [pi]) (3 [pi] / 2- [pi] + ([pi] / 2) sin [beta])) = K (section) ・ sin β ・ ・ (π / 2 ≦ β ≦ 3π / 2) f (g (β)) = K (section) ・ f (2π + (π / 2) ・ sin β = K (section) ・(-1+ (2 / π) (2π ++ (π / 2) sin β-3π / 2)) = K (section) · sin β ··· (3π / 2 ≦ β ≦ 2π) (7) That is, a single sin wave is output when there is no modulation.

Further, in order to obtain the original waveform ORG (t) when the modulation index I = 1 is set by substituting the equations (5) and (6) into the equation (4), the above-mentioned FIG. Replace the sine wave of (b) with the triangular wave defined by the above equation (5), and
The period data F (t) in FIG. 7C is replaced with the function g (t) defined by the equation (6), and the phase difference data M (t) is determined as the difference data from the g (t). do it.

In this case, as the means for generating the triangular wave corresponding to sin ROM 2 in FIG. 1, it is possible to generate the triangular wave by a decoder circuit or the like in addition to the ROM.

In addition to the above aspect, the functions f and g satisfying the above equation (4)
Can define various combinations of functions.

〔The invention's effect〕

According to the present invention, if the mixing rate of the modulation signal is set to 0 in advance by the mixing control means, a musical tone waveform consisting of only a sine wave or a cosine wave can be generated, and the mixing rate of the modulation signal is set in advance by the mixing control means. If it is set to a predetermined mixing ratio (for example, 1),
It is possible to obtain a desired musical sound waveform such as a musical sound of a natural musical instrument.

In the present invention, so-called loop reproduction is possible, and even during the loop reproduction, the musical tone is gradually changed from the desired musical tone waveform state to the state containing only a single sine wave component or a single cosine wave component. The frequency characteristics of the waveform can be controlled. Alternatively, by changing the mixing ratio continuously to, for example, 1 or more, it is possible to control so that a unique musical tone having a more complex overtone configuration is generated from a desired musical sound waveform state. This makes it possible to generate a variety of musical tone waveforms only with the carrier signal and the modulation signal corresponding to an arbitrary waveform section, save the storage capacity of, for example, the carrier signal and the modulation signal, and perform loop playback during loop reproduction. The tone can be controlled so as not to become monotonous.

In addition to the above operation, the amplitude envelope control means controls the amplitude envelope characteristic of the musical tone waveform output from the waveform output means so as to be attenuated, for example, in time in synchronization with the loop reproduction operation, so that an actual musical instrument is produced. It is possible to realize a process in which the musical tone waveform is gradually attenuated after the start of sound generation like the musical tone of FIG.

As described above, according to the present invention, it is possible to easily generate both a state in which a natural tone is sounded and a state in which only a single sine wave component or a single cosine wave component is included, and various overtone characteristics are also obtained. In addition, it is possible to obtain the harmonic overtone characteristic at the time of loop reproduction. And as a structure for realizing it, it can be realized only by a combination of a normal ROM, a decoder, an adder, a multiplier, etc., so that a complicated tone waveform can be realized with a simple circuit structure, and as a result, Thus, it becomes possible to provide high-quality electronic musical instruments and the like at low cost.

[Brief description of drawings]

FIG. 1 is a basic principle configuration diagram of an embodiment of a musical tone waveform generator according to the present invention, FIG. 2 is a diagram showing a relationship between ORG (t) and F (t), and FIG. 3 is ORG ( t), F (t), and M (t), FIG. 4 shows how to determine M (t), and FIG. 5 shows D () when I is changed. t), F (t) and M (t), FIG. 6A is a concrete configuration diagram of one embodiment of the tone waveform generator, and FIG. 6B is a diagram showing generation of address data ADD # 1. Circuit configuration diagram, FIG. 6C is a configuration diagram of a modulation index I generation circuit, FIG. 7 is a data configuration diagram of FM ROM, FIG. 8 is a data configuration diagram of K ROM, and FIG. An operation timing chart of a concrete configuration of one embodiment of the waveform generator, FIG. 10 is an operation timing chart at the time of loop reproduction operation, and FIG. 11 is a principle operation explanatory view at the time of loop reproduction operation. 1 ₁ ... FM ROM, 2 ... sin ROM, 3, 7, 13, 24 ... Adder, 4, 5, 18 ... Multiplier, 6, 9, 15, 16, 19, 2
1,23 …… Latch, 8,14,25,29,33 …… Selector, 1
0 …… D-flip-flop (F / F), 11, 20 …… Inverter, 12 …… Exclusive OR circuit (EOR), 17 …… Envelope generator, 22 …… Subtractor, 26 …… Decoder, 27
... OR circuit, 28 ... RS-flip-flop (F / F), 3
0 …… Address counter, 31 …… I ROM, 32 …… Low frequency oscillator (LFO), ADD # 1, ADD # 2 …… Address data,
F (t) ... period data, M (t) ... phase difference data,
I ... Modulation index, F (t) + IM (t) ... Phase angle data, IB ... Section identification data, K (section) ... Normalization coefficient, D (t) ... Modulation output, OUT ( t) ... Waveform output. CLK # 1, CLK # 2 ... clock, STRT ... start pulse,

Claims (2)

[Claims] 1. A carrier signal generating means for generating a carrier signal, a modulation signal generating means for generating a modulation signal, and the modulation signal when the modulation signal is mixed with a carrier signal generated from the carrier signal generating means. Mixing control means for controlling the mixing ratio of the carrier signal from 0 to an arbitrary mixing ratio, and outputting a mixing signal in which the carrier signal and the modulation signal are mixed at the mixing ratio, and input and output. Has a predetermined functional relationship and outputs a modulated tone waveform with the mixed signal output from the mixing control means as an input, and the predetermined functional relationship and the carrier signal, When the mixing control means controls the mixing ratio of the modulated signal to be 0, the musical tone waveform generated from the waveform output means has a relationship of a sine wave or a cosine wave. Signal The means and the modulation signal generating means output a desired musical tone waveform from the waveform output means when the mixing control means controls the mixing rate of the modulation signal to be a predetermined mixing rate. Generating a carrier signal and a modulation signal, and repeatedly generating an arbitrary signal section of the carrier signal and the modulation signal so that the arbitrary waveform section of the tone waveform is repeatedly output from the waveform output means, and the mixing control The means changes the mixing ratio with time in synchronism with the repeated generation operation. 2. An amplitude envelope control means for temporally changing the amplitude envelope characteristic of the tone waveform output from the waveform output means in synchronization with the repetitive generation operation. Musical tone waveform generator.