JP2678970B2 - Tone generator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the generation of musical tones for electronic musical instruments and the like in which a plurality of musical tone synthesizing methods can be efficiently mixed.
Related to the device .

[0002]

2. Description of the Related Art A musical tone of an ordinary musical instrument has a complicated overtone structure such that the frequency and amplitude of the overtone always fluctuate, and some musical instruments include non-integer harmonics. It has impact noise when the piano is attacking (rising). Such overtones and noise components greatly characterize the timbre of the musical instrument.

Such a musical sound is generated by a musical sound generator such as an electronic musical instrument.
In order to reproduce the sound realistically in a room , and to create a new sense of sound that has never existed before, various music synthesis methods are easy.
Used in sound generators .

Among these methods, there are many methods such as a PCM method, a frequency modulation method, a phase modulation method, and a harmonic overtone addition method, each of which has advantages and disadvantages. If these are used in combination as appropriate, the above-mentioned object can be achieved to some extent. For example, during an attack, the harmonic structure is usually complicatedly changed in addition to the noise described above. Therefore, the PCM method is suitable, and the sustain portion (steady portion) after that is the PC.
When the M method is used, a memory having a large storage capacity is required, so it may be possible to switch to another method.

As one example thereof, Japanese Patent Laid-Open No. 58-10229.
No. 6 synthesizes musical tones by using the PCM system as the musical tone waveform at the time of attack and the frequency modulation system after that. Hereinafter, this method will be described.

When a key is pressed, numerical data corresponding to the pitch of the pressed key is repeatedly accumulated to obtain an accumulated value whose value changes at a speed corresponding to the pitch. After that, the waveform data of the attack part is read from the waveform memory of the PCM system with this accumulated value, and the envelope of the attack part shown in FIG.

On the other hand, the numerical data corresponding to the pitch of the above-mentioned pressed key, the cumulative value that repeatedly accumulates the numerical data and changes at the speed corresponding to the pitch, and another constant are set to the frequency. Modulation method Input to a tone generator to obtain a frequency modulation waveform.

After that, the envelope shown in FIG. 12 (b) is given to the waveform, and it is added and synthesized by the adder with the tone waveform of the above-mentioned PCM system, and as shown in FIG. 12 (c), the attack part of the tone and the sustain. A full-waveform signal of a musical tone having a (steady) portion and a release (attenuation) portion is created.

As is clear from the above, in this tone synthesis method, the attack portion of the tone uses a synthesized waveform according to the PCM method, and thereafter, a synthesized waveform according to the frequency modulation method is used, and the tone synthesis is performed with the lapse of time from the attack. The method is switched.

In addition, it may be considered that the PCM system and the frequency modulation system or the phase modulation system are assigned to different sounding channels.

[0011]

In this case, the PCM system is a system for reading PCM waveform data from a waveform memory, and the frequency modulation system or the phase modulation system described above uses sine waveform data or the like stored in the waveform memory. This is a method of reading while modulating with an address signal.

In the PCM system, the musical tone waveform of the original natural musical instrument or the like is sampled at a frequency at least twice the effective frequency band of the musical tone waveform according to the sampling theorem, and stored in the waveform memory as PCM waveform data. ing. Then, by thinning out each sample point, it is possible to output musical tone waveforms of various pitches. In this case, by storing and utilizing the PCM waveform data for some typical tone ranges, it is possible to prevent the fidelity of the output waveform to the original tone from being deteriorated due to the above-described thinning processing.

On the other hand, in the modulation method, the modulation waveform data such as a sine wave stored in the waveform memory has a frequency twice that of a general frequency band which can be sounded as a musical tone waveform, for the following reason. Sample values need to be prepared at sample intervals corresponding to considerably higher frequencies. The reason is that in order to obtain waveform data that does not contain musically harmful noise components in the entire soundable range, the sample value of the lowest pitch, that is, in the state where sampling points are not thinned, is used. This is because a sufficiently large number of modulation waveform data are required. In particular, since the step width of the read address of the waveform memory can change greatly with time as the modulation becomes deeper, it is possible to suppress the generation of harmful noise components in response to such a change in the step width and to faithfully perform modulation. In order to obtain the generated waveform data, it is necessary to allow a sufficient margin in the number of sample values of the modulation waveform data.

Here, when it is attempted to realize a tone generation device that combines the PCM system and the modulation system, it is conceivable that the waveform memory is shared by both systems in order to prevent the circuit configuration from becoming complicated. And both formulas in real time
When it is considered to access the waveform memory and generate a musical tone waveform while switching with, the musical tone waveform of the same pitch is output in the same manner between the two expressions in response to the addressing based on the performance operation. Need to be configured.
That is, from the user's point of view, if there is a difference in the sounding operation or the like due to the difference in the tone synthesis method, the tone generator will have a very poor operability. Therefore, in the conventional example in which the PCM system and the modulation system are combined as described above, the sampling intervals of the PCM waveform data and the modulation waveform data stored in the waveform memory need to be set to have the same rate.

[0015] However, for example, the sampling interval of the modulation waveform data widely set equal to the sampling interval of the PCM waveform data to reduce the number of sample values, musical harmful noise component generated as described above It has a problem that it is easy to do. On the contrary, if the sampling interval of the PCM waveform data is set to be narrower than the sampling interval of the modulation waveform data and the number of sampling values is increased, an enormous memory capacity is required and the musical tone generating apparatus There is a problem that the cost will increase significantly.

An object of the present invention is to provide, in a musical tone generator in which a PCM system and a modulation system are mixed, a sample suitable for both the PCM waveform data and the modulating waveform data without degrading the sound quality of the musical tone to be produced. It is possible to store in the waveform memory at intervals, and to enable the pronunciation of musical tones by simple addressing while switching both types in real time.

[0017]

SUMMARY OF THE INVENTION According to the present invention, a tone synthesis operation by the PCM system for accessing a waveform data storage means storing PCM waveform data to output a tone waveform and a waveform data storage storing modulation waveform data. It is premised on a musical tone generating apparatus capable of generating a musical tone signal corresponding to each musical tone synthesizing operation by appropriately switching the musical tone waveform synthesizing operation by a modulation method for accessing the means and outputting a musical tone waveform . In this case, for example, the tone generation channel is associated with each of a plurality of time division processing timings, and the tone synthesis operation is performed by the time division processing while switching the PCM system and the modulation system for each sound generation channel. Alternatively, the PCM system and the modulation system are switched to generate sound in the sound generation section from the start of sound generation of each tone to the mute.

Further, such a musical tone generating apparatus has the following control means. That is, the control means, when accessing the waveform data storage means that stores the PCM waveform data, converts the frequency specified by the supplied frequency information into a frequency based on the sampling frequency of the PCM waveform data, An address signal corresponding to the converted frequency is generated and accessed. Further, when the control means accesses the waveform data storage means that stores the modulation waveform data, it controls the frequency specified by the frequency information.
Transformed into the frequency relative to the sampling frequency of the modulation waveform data to generate a luer address signal to correspond to the converted frequency, performs access by appropriately modulating the address signal.

In the above configuration, the waveform data storage means for storing the PCM waveform data and the waveform data storage means for storing the modulation waveform data can be a common waveform memory, and the PCM waveform data and the modulation waveform data can be used. Can be configured to be stored in different address areas of the waveform memory.

[0020]

The PCM waveform data stored in the waveform data storage means is generally data obtained by sampling a musical tone waveform of a natural musical instrument or the like at a frequency that is at least twice the effective frequency band of the musical tone waveform according to the sampling theorem. . Also,
The modulating waveform data such as a sine wave stored in the waveform data storage means has a general frequency that can be generated as a musical tone waveform in order to prevent the generation of a musically harmful noise component in the entire soundable range. It is prepared with a sampling interval corresponding to a frequency considerably higher than twice the frequency of the band. That is, the sampling frequency of the stored waveform data differs between the PCM system and the modulation system.

On the other hand, a frequency designated by frequency information (pitch data, etc.) generated in response to a performance operation such as a key depression operation on a keyboard generally corresponds to a note name of a depressed key or the like. There is. Therefore, when the tone synthesis operation is performed while switching between the PCM system and the modulation system, the control means supplies the
The frequency information is converted into frequencies corresponding to the sampling frequencies of the PCM waveform data and the modulation waveform data, an address signal is generated based on the converted frequencies, and each waveform data is stored. The waveform data storage means is accessed.

As a result, it becomes possible to store the PCM waveform data and the modulating waveform data in the waveform storage means at a sampling interval suitable for both formulas without deteriorating the sound quality of the generated musical tone, and both formulas can be stored in real time. It is possible to generate a musical tone by simple addressing while switching to, and it is also possible to share the waveform storage means between the PCM system and the modulation system.

[0023]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, one embodiment in which the present invention is applied to an electronic keyboard instrument will be described with reference to the drawings.

FIG. 1 is an overall configuration diagram of one embodiment. First, the performance mode selection input section 3 has a performance mode selection switch, which is not particularly shown. The switch is a switch for selecting a performance mode desired by a performer from several performance modes in which single and plural tone synthesis systems are assigned to sound generation channels.

Next, a CPU (central control unit) in the control unit 2 controls the musical instrument system according to a program written in a ROM (Read Only Memory) not shown.

Further, the control unit 2 scans at a fixed cycle and fetches key depression / key release information of the keyboard unit 1. When a key is pressed, the pressed key is assigned to any of a plurality of sound generation channels and the operation information of the key is sent to the key code register unit 5. The key code register unit 5 temporarily stores the key code based on the key operation information sent from the control unit 2, and also generates pitch data corresponding to the key code, that is, step width data for reading the memory unit 9. The number of channels is output to the address generator 6.

The PCM system / modulation system / selection unit 4 indicates either the PCM system or the modulation system on the basis of the selection information of the musical tone synthesis system sent from the control unit 2, and the constant multiplier described later. Outputs a flag for activating.

Next, the address generator 6 outputs the address signal X corresponding to the accumulated value of the step width corresponding to the pitch data sent from the key code register section 5 (for this reason, the step for reading out the memory section). The speed changes according to the pitch of the depressed key). The address signal X is used to read the address of the waveform memory of the memory unit 9, and is also input to the modulator 7 as original phase angle data before being modulated.

Next, the memory section 9 stores in advance waveform data such as tone waveform data used in the PCM system and sine wave data (sine table) used in the modulation system, and each is read from another address.

The modulation section 7 is a circuit for modulating an address signal for reading the sine wave memory of the memory section 9 when performing the tone synthesis of the modulation method. The selection circuit 8 is a gate circuit, and when the flag of the PCM / modulation system / selection unit 4 is “1” (in the case of the PCM system), the address signal output X of the address generation unit 6 is selected as it is, and the flag is also set. When it is "0" (in the case of the modulation method), the address signal X'modulated by the modulator 7 is selected. In the selection circuit 8, the upper address corresponding to the integer part of the address is input to the memory unit 9. The lower address corresponding to the fractional part is input to the interpolating part 10, where the waveform data of the address corresponding to the integer part + the fractional part is obtained.

Next, the interpolation section 10 will be described. When the waveform data of the memory unit 9 is read, as shown in FIG. 2A, depending on the pitch of the pressed key, as shown in FIG. 2A, the middle part n + a of two adjacent addresses n and n + 1 of the memory unit 9 (where 0 <a Amplitude value X _m (X ₁
, Which is on the line connecting X ₁ and X _{2 in} the interpolated value of X ₂ and X ₂ . This a is the fractional part of the above-mentioned address and corresponds to the lower address.

This X_mIs calculated as follows. FIG.
(b) is a principle block diagram of the interpolation part 10. Smell
The amplitude values corresponding to the addresses n and n + 1, respectively.
X₁, X_Two10 for temporarily storing_a, 10 _bOut of
Force difference (X_Two-X₁) Is obtained by the subtractor 10c, and then the multiplier
10_dMultiplies the lower address value a by_Two-X₁)
Get. Then, adder 10_eAt address n
Amplitude value X₁And add X₁+ A (X_Two-X₁) = X_m Get.

In this way, the interpolation unit 10 adds
Less n₁, N_TwoAmplitude value X₁, X_TwoFrom address (n +
A) amplitude value X_mIs obtained. The lower address a
When 0, interpolation is not performed and the waveform data from the memory unit 9 is
Is the register 10 in Fig. 2 (b). _b, Adder 10_eThrough
It is directly output to the multiplication unit 12.

Next, the waveform data for one channel obtained as described above is multiplied by the envelope signal from the envelope generator 11 in the multiplier 12. Thereafter, the accumulator 13 accumulates the waveform data for all the channels for each sample, and the D / A converter 14 converts the waveform data into analog musical tone signals and sends them to the audio system.

Next, the relationship between the tone synthesis method assigned to the tone generation channel and the key code will be described with reference to FIG. FIG. 3 shows a case of eight-tone polyphonic as an example.

First, in the performance mode 1, the musical tone synthesis method is set to P.
When only the CM method is used, the flag of the PCM method / modulation method / selection unit 4 in FIG. 2 is always “1”, and as shown in FIG.
Musical tone waveform data according to the PCM method is assigned to all eight channels of the pressed eight keys (key codes K1, K2, K3, ..., K8).

The next performance mode 2 is a case where a modulation system such as a phase modulation system or a frequency modulation system is used independently, and the flag of the PCM system / modulation system / selection unit 4 is always "0", as shown in FIG. As described above, the musical tone waveform data by the modulation method is assigned to all eight channels of the eight keys (key codes K1, K2, K3, ..., K8) that are depressed as in the mode 1.

The next performance mode 3 is a feature of the present invention, in which P is generated by the control signal output from the controller 2.
The flag output by the CM system / modulation system / selection unit 4 is changed for each sounding channel, and the PCM system and the modulation system are switched for each of the four sounding channels. Then, for each of the key codes (K1, K2, K3, K4) of the sound generation channels 1 to 4 and the sound generation channels 5 to 8, sound is generated by two tone synthesis methods, and a four-tone polyphonic sound is generated in each. Done. In this case, each tone corresponding to each of the four key codes (K1, K2, K3, K4) of the two methods has different envelopes E1, E2, E3, E4 and E1 ', E2'.
, E3 ', E4' are given for each sounding channel timing. In the case of the performance mode 3, the musical tones synthesized by the two methods of the PCM method and the modulation method are produced at the same time in the sense of hearing, and a non-monotonic "thick" musical sound is obtained.

Next, the operation of the address generator 6 commonly used for both the PCM system and the modulation system will be described. In the case of the present embodiment, when the address is designated and the waveform data of the memory unit 9 (FIG. 1) is sequentially read out, the step width of the address needs to be changed according to the key code of the depressed key, that is, the pitch. There is. The step width is changed by the address generator 6, and its circuit configuration is shown in FIG.

In FIG. 4, an adder 20 and a current address register 22 for storing a time-varying address are stored.
Operates as an accumulator. The pitch register 19 temporarily stores the step width data sent from the key code register 5. The step width data enters the accumulator described above via a constant multiplier 23, which will be described later, and is accumulated with a step width corresponding to the pitch of the pressed key. As a result, for example, if the step width of a certain sound is 1, 1, 2, 3, ...
For octave-high sounds, a series of accumulated values such as 2, 4, 6, ... Is sequentially output from the current address register 22. This is the address signal X for reading the memory section 9. The address signal X is address data for reading the waveform data of the memory section 9 in the case of the PCM system, but corresponds to the original phase angle data before being modulated in the case of the modulation system.

Next, the constant multiplier 23 will be described. In the case of the PCM method, the memory unit 9 stores musical tone waveform data of a natural musical instrument or the like having a certain pitch frequency at a sampling cycle. On the other hand, the modulator 7
In the case of the phase modulation method described later as the first and second embodiments of the present invention and in the case of the frequency modulation method described later as the third embodiment, the memory unit 9 stores the same in order to increase the accuracy of the synthesized tone waveform. A certain pitch frequency stored (typically
The sine wave data of the lowest frequency of the musical sound is stored in a period (one-hundredth to one-tenths of the PCM method) smaller than the sampling period. And in this embodiment,
In order to generate sound in parallel by mixing the PCM system and the modulation system, the constant multiplier 23 is operated to multiply the step width in the phase modulation system or the frequency modulation system by a predetermined amount, and a musical tone waveform is produced. The pitch of is equal to that in the PCM system.

Then, in order to operate the constant multiplier 23, a constant multiplication flag is output from the PCM system / modulation system / selection input section 4. This constant multiplication flag is the same as the flag for instructing the tone synthesis method, and is "1" for the PCM method, "0" for the modulation method, and "1" (for the PCM method). The constant multiplication factor is 1, and when it is "0" (in the case of the modulation method), the above-mentioned predetermined multiplication factor is obtained.

By the way, in order to synthesize a musical tone based on the waveform data of the memory section 9, the reading is started from the start address of the waveform data (including the sine waveform) and the end data is read.
After the address has been read once, the memory is read repeatedly, returning to the start address again. This is an operation called loop reproduction. For that purpose, the address width (this is called the loop width) data from the start address to the end address is stored in the loop width register 1
Store in 5 beforehand.

Next, the operation of this loop reproduction will be described. The comparison circuit 18 constantly checks whether the current address C of the output of the adder 20 becomes equal to or exceeds the end address R set in the end address register 17. If it becomes equal to or exceeds the end address R, the subtracter 21 subtracts the loop width from the current current address C, and the current address C returns to the start address.

At this time, the reading of the memory section 9 (FIG. 1) may not be completed accurately at the end address depending on the step width. For example, from the memory unit 9 storing the tone waveform shown in FIG. 5A (in this embodiment, one cycle of the waveform is loop-reproduced), the waveform is stored at the address shown in FIG. Suppose you want to read with. When the step width output from the pitch register 19 of FIG. 4 is 1 or 2, the current address C is set as shown in FIG. 5 (b) or (c).
8 of 8 corresponds to 8 of the end address R. as a result,
The loop width 8 is the output 8 of the adder 20 in the subtractor 21.
And the output of the current address register 22 becomes zero. This returns to start address 0,
As before, again, 1, 2, 3, ... or
The address advances to 2, 4, 6, ....

On the other hand, in the case of the step width 3 in FIG. 5C, the current address increases to 0, 3, 6, and 9 and exceeds the end address of 8. As a result, the loop width of 8 is subtracted from the output 9 of the adder 20, and the current address register 2
The output of 2 becomes 1. Therefore, the loop reproduction is started from the address 1 of the previous start address, not the address 0, and thereafter, the addresses are sequentially advanced with the step width 3, 4, 7 ,.

In this way, a waveform having no discontinuity, which is repeatedly reproduced with the loop width defined by the loop width register 15, is obtained. Next, how the address signal X is changed in the modulator 7 will be described in order using the first and second embodiments of the phase modulation method and the third embodiment of the frequency modulation method. << First Example of Modulating Unit 7 >> First, a first example of the modulating unit 7 will be described. In this embodiment, by changing the read address signal X of the memory section 9 (FIG. 1) storing the sine wave data, the read phase angle of the sine wave is changed and various waveforms are obtained. In the case of this phase modulation method, the address signal X will be referred to as original phase angle data X.

First, before describing the circuit configuration of this embodiment, the operating principle of this embodiment will be described. FIG.
(g) shows the relationship between the original phase angle data X and the modified phase angle data X'obtained in the circuit of FIG. 7 described later. As can be seen from the figure, in this embodiment, X and X'are related by the slope of the polygonal line function. From FIG. 6 (g), the slopes of the polygonal lines in the figure are 0 ≦ X <M and NM ≦ X.
If α for <N and β for M ≦ X <NM, α = (N / 4) / M (1) β = (N / 4) / (N / 2-M ) (2) From these equations, 1 / α + 1 / β = 2 (3) is obtained. Here, N represents 2π radians (one cycle) of the phase angle, and M is a switching point of the slope of the polygonal line. Then, when X = M, X ′ = N / 4 = π / 2, and the maximum value of the sine wave data in the memory unit 9 (FIG. 1) is accessed.

From FIG. 6 (g), in the circuit of FIG.
For the original phase angle data X, 0 ≦ X <M and NM ≦
The modified phase angle data X ′ is obtained by equivalently performing an operation of multiplying α when X <N and β when M ≦ X <N−M.

Further, in the circuit of FIG. 7 described later, β = α / 2 ^k ・ ・ ・ (4) (However, k = 0, 1, 2, ・ ・ ・, 7), α and β are determined, and equation (3) and (4)
From the formula, α = (1 + 2 ^k ) / 2 (5) β = (1 + 2 ^−k ) / 2 (6) is obtained. From the equations (5) and (6), in the circuit of FIG. 7, which will be described later, after k is determined in advance, 0 ≦ X <M and NM
When ≦ X <N, X ′ = αX = {(1 + 2 ^k ) / 2} X = 2 ⁻¹ X + 2 ^k−1 X (7) When M ≦ X <N−M, X ′ = βX = {(1 + 2 ^−k ) / 2} X = 2 ⁻¹ X + 2 ^{−k −1} X (8) is equivalently calculated. The expressions (7) and (8) can be executed by a binary bit shift operation, and the circuit of FIG. 7 described later is also configured to perform the bit shift operation.

In order to equivalently perform the operations of the above equations (7) and (8), the operation shown below is sequentially executed in the circuit of FIG. 7 which will be described later. The lower bits excluding the most significant bit X _MSB of the original phase angle data X are extracted, and the data that repeats at N / 2 as a boundary with respect to changes in the original phase angle data X as shown in FIG. 6 (a). X ₀ ′ is obtained.

X ₀ ′ is appropriately inverted, and data X ₁ ′ and X ₁ ″ having the characteristics as shown in FIGS. 6B and 6C are obtained with respect to changes in the original phase angle data X. ₁ ′ is multiplied by β and X ₁ ″ is multiplied by α, and the data X ₂ ′ and X ₂ ″ having the characteristics as shown in FIGS. can get.

When X ₂ ′ ≧ N / 4, X ₂ ″ is selected, and when X ₂ ′ <N / 4, X ₂ ′ is selected. Data X ₃ ′ having the characteristics as shown in FIG. 6 (f) is obtained.

When 0≤X <M, X ₃ 'is selected and M
When ≦ X <N / 2, the operation (N / 2−X ₃ ′) is executed, and when N / 2 ≦ X <N−M, the operation (X ₃ ′ + N / 2) is executed and N− When M ≦ X ≦ N (N / 2−
The calculation X ₃ ′) + N / 2 is executed.

By the above-mentioned operations of to, FIG.
In the circuit of FIG. 6, operations equivalent to the expressions (7) and (8) described above are executed, and as shown in FIG.
The modulated phase angle data X'having the polygonal line characteristic as shown in (g) is obtained.

Next, as the configuration of the modulation section 7 of FIG. 1, an example of the circuit of FIG. First, the lower bits except the most significant bit X _MSB of the original phase angle data X output from the address generator 6 of FIG. 1 are taken out to obtain X ₀ ′. This corresponds to the arithmetic operation described above.

Next, this X ₀ ′ is the inverting circuits 31 and 3
Enter in 4. At the R terminal of the inverting circuit 31, the most significant bit X _MSB of the original phase angle data X is transferred to the inverter 30.
The signal 1 inverted by the
X _MSB is connected to the R terminal of

[0058]

[Outside 1]

Input as it is. Thus, the inverter circuit 31, the X _MSB is 'inverts the X _1' input X ₀ when "0" is output as the output as 'it is X ₁ an' input X ₀ when "1" To do. On the other hand, the inverting circuit 34
Outputs the input X ₀ ′ as it is as X ₁ ″ when the above X _MSB is “0”, and inverts the input X ₀ ′ and outputs it as X ₁ ″ when it is “1”. 6 (b), (c)
The phase angle data X ₁ ′ and X ₁ ″ are obtained as described above. The above corresponds to the above-mentioned arithmetic operation.

After this, the phase data X₁'To the right
The right shift operation of k bits is performed by the shift circuit 32.
2^-kX₁′ Is calculated. Then, the adder 33
Output 2 of the right shift circuit 32^-kX₁′ And phase data
X₁Each of ’is shifted by 1 bit to the right and then added,
As a result, X_Two′ = ΒX₁′ = {(1 + 2^-k) / 2} X₁′  = 2^-1X₁′ +2^-k-1X₁′ ・ ・ ・ (9) is obtained. On the other hand, the phase data X₁″ For left shift times
The path 35 performs a left shift operation of k bits,^k
X₁″ Is calculated, and the adder 36 operates in the above left system.
Output 2 of shift circuit 35^kX₁″ And phase data X₁"of
After shifting each by 1 bit to the right, they are added and the result is
X_Two″ = ΑX₁″ = {(1 + 2^k) / 2} X₁″  = 2^-1X₁″ +2^k-1X₁″ (10) 1 is obtained, where the right shift circuit 32 and the left shift circuit 3
The coefficient k that determines the shift amount in 5 is the control unit of FIG.
3-bit control data S input from 2 to the modulator 7 ₀~ S
_TwoGiven by

Next, the selection circuit 37 outputs the carry-out CO from the adder 33, that is, X ₂ ′ ≧
Select X ₂ ″ only when N / 4, X otherwise
'It operates to select, with respect to the original phase angle data X, the phase data X ₃ having such properties in FIG 6 (f)' ₂ outputs a. The above corresponds to the above-described arithmetic operation.

The next inverting circuit 38 is the selecting circuit 37.
Selectively inverts the output X ₃ 'of The circuit is a signal obtained by inverting X _MSB by the inverter 30 and a carry of the adder 33.

[0063]

[Outside 2]

X ₃ 'is inverted when the output of the exclusive OR circuit 39, which receives the output CO, becomes "1". That is, one of Outer 3 and Carry Out CO

[0065]

[Outside 3]

It is M that one is "1" and one is "0".
Since the range is ≦ X <N / 2 and NM ≦ X ≦ N, the output of the exclusive OR circuit 39 is “1” in this range.
And X ₃ 'is inverted. 0 ≦ X <M and N / 2 ≦
In the range of X <NM, the exclusive OR circuit 39
The output of "0", the X ₃ 'is output as it is.

Then, the output of the exclusive OR circuit 39 is added as an upper bit to the output of the inverting circuit 38, and the above-mentioned most significant bit X _MSB is further added as the most significant bit to obtain a final modulation phase angle. Data X'is obtained. In this case, the output of the exclusive OR circuit 39 and the value of X _MSB are (0, X
0), M ≦ X <N / 2, (1, 0), N / 2 ≦ X
Since (0, 1) in NM and (1, 1) in NM ≦ X ≦ N, the value of the modulation phase angle data X ′ is
When 0 ≦ X <M, it becomes equal to the output value of the inverting circuit 38, and when M ≦ X <N / 2, it becomes a value obtained by adding N / 4 to the output value of the inverting circuit 38, and N / 2 ≦ X <N. At −M, it becomes a value obtained by adding N / 2 to the output value of the inverting circuit 38, and at N−M ≦ X ≦ N, it becomes a value obtained by adding 3N / 4 to the output value of the inverting circuit 38. Thus, the modulated phase angle data X'having the characteristic shown in FIG. 6 (g) with respect to the original phase angle data X is obtained.

FIG. 8 shows various waveforms obtained by accessing the sine wave data of the memory section 9 using the modulation phase angle data X'obtained by using the circuit of FIG. 7 as the modulation section 7 of FIG. Show the data. It can be seen that the sine wave gradually approaches the sawtooth wave as the value of the coefficient k increases.

The above is an example in which the circuit of FIG. 7 generates the modulated phase angle data X'having the polygonal line characteristic as shown in FIG. 6 (g) with respect to the original phase angle data X. Also, for example, with respect to the original phase angle data X, FIGS.
By providing a circuit for generating the modulated phase angle data X ′ having the polygonal line characteristic as shown in FIG. 5, variously modulated waveform data can be obtained as shown in FIG. << Second Embodiment of Modulating Unit 7 >> Next, a second embodiment of the modulating unit 7 will be described.

FIG. 10 is a block diagram showing the configuration of a second embodiment of the modulator 7 based on a phase modulation method different from that of the first embodiment described above. In the figure, the address signal X output from the address generation unit 6 (FIG. 1) is a sine wave,
Alternatively, a waveform R that stores another waveform different from the sine wave
Read OM40. The read waveform is multiplied by the envelope output from the envelope generator 41 in the multiplier 42, and then the waveform is input to the selection circuit 43. In addition, the address signal X is directly input to the selection circuit 43,
In particular, a control signal output from a CPU (not shown) selects these two inputs. Thereafter, the output of the selection circuit 43 is multiplied by a constant multiplier by the constant multiplier 44, and the memory unit 9 of FIG.
The step width for reading is expanded by a constant multiple of several times to several tens of times the step width in the case of PCM. The output of the constant multiplier 44 thus obtained is the output X'of the modulator 7 in FIG.
become. Since the constant multiplier 44 is used, the constant multiplier 23 (see FIG. 4) of the address generator 6 in FIG. 1 is omitted.

When the address signal is directly selected in the circuit of FIG. 10, the sine wave is read from the sine table of the memory section 9. Further, when the output of the multiplier 42 is selected, a more complicated waveform can be obtained. In this case, in particular, by storing various waveforms in the waveform ROM 40, the output X'can be a more complicated signal than in the case of the first embodiment described above, and the waveform read from the memory unit 9 in FIG. The data can also be modulated to more complex characteristics. << Third Embodiment of Modulation Unit 7 >> Next, a third embodiment of the modulation unit 7 will be described.

FIG. 11 shows the modulation based on the frequency modulation method.
It is a block diagram which shows the structure of the 3rd Example of the part 7. Addition
Input signal ω to the calculation unit 54 and the multiplication unit 50_ct is the address
Corresponding to the address signal X of the scanning unit 6 (FIG. 1)
It is an accumulated value whose value changes with the step width corresponding to the pitch of.
This accumulated value repeats the phase change of 0 → 2π radian.
This ω_ct is a constant value ω in the multiplication unit 50_{m /}ω_cIs multiplied by
Got ω_mSign table with t as address signal
Access 51. Furthermore, its output signal sinω_m
The modulation index is generated based on the control signal in the multiplier 52 to t
The modulation index I (t) generated from the unit 53 is multiplied to obtain I
(t) sinω_mt is obtained. Next, in the adding section 44,
ω _ct and I (t) sin ω_mt and are added, ω_ct +
I (t) sin ω_m(t) is obtained. After that, the addition unit 54
The constant multiplier 55 for the output of
Is multiplied, and the step size of reading of the memory unit 9 in FIG. 1 is expanded.
Will be great. The output of the constant multiplier 55 thus obtained
Becomes the output X'of the modulator 7 in FIG. The constant multiplier 5
5, the constant multiplier 2 of the address generator 6 of FIG. 1 is used.
3 (see FIG. 4) is omitted.

The overall operation of the embodiment shown in FIG. 1 described above is summarized as follows. First, when the performer selects the performance mode shown in FIG. 3 and performs a key-depressing operation, the control signal output from the control unit 2 for selecting a musical tone synthesis system causes the PCM system / modulation system / selection unit 4 to select the PCM system. In the case of, the flag "1" is output, and in the case of the modulation method, the flag "0" is output. By this flag, the memory unit 9 and the selection circuit 8
Each of the selection operations is performed. That is, when the flag is "1" and the most significant bit (MSB) of the address of the memory unit 9 is "1", the musical tone waveform data used in the PCM system has the most significant bit "1". When the flag is "0", the sine wave data used for the modulation method is read from the address area corresponding to the most significant bit "0". Then, based on the key information corresponding to the pitch of the pressed key, the address generator 6 outputs an address signal X having a step width corresponding to the pitch of the pressed key. In the case of the PCM method, the tone waveform data of the memory section 9 is directly read out via the selection circuit 8 by the address signal X. On the other hand, in the case of the modulation method, this address signal X is
The sine wave data of the memory section 9 is read out while being modulated by the output X'through the selection circuit 8.

The above-described selecting operation of the musical tone synthesizing method is performed for each sounding channel timing corresponding to the key operated for performance, based on the performance mode selection information. As for the output of the selection circuit 8, the upper bit corresponding to the integer of the address signal is sent to the memory unit 9 and the lower bit corresponding to the fractional part is sent to the interpolating unit 10, and the interpolating unit 10 performs the above-described interpolating operation, and the multiplying unit 12, Sound is emitted from the audio system from the accumulator 13 via the D / A converter 14.

In this embodiment, the electronic musical instrument to which the PCM method in which the amplitude value of the waveform itself is coded is applied has been described. However, the so-called DPCM and ADPCM methods in which the difference value of the waveform is stored are also PCM methods. Needless to say, it is included in the present invention as one and is applicable.

[0076]

According to the present invention, when the musical tone synthesizing operation is performed while switching between the PCM system and the modulation system, the supplied frequency information is the sampling frequency of each of the PCM waveform data and the modulation waveform data. Is converted to a frequency corresponding to, the address signal is generated based on the converted frequency, and the waveform data storage means in which each waveform data is stored is accessed. It is possible to store the PCM waveform data and the modulation waveform data in the waveform storage means at a sampling interval suitable for both equations without deterioration.

It is possible to generate a musical tone by simple addressing while switching both types in real time, and it is also possible to share the waveform storage means between the PCM system and the modulation system. This makes it possible to reduce the overlapping parts of the circuit and generate musical tones with excellent sound quality.
The device can be realized at low cost.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of an embodiment of the present invention.

2A and 2B are explanatory diagrams of an interpolation unit 10. FIG.

FIG. 3 is a diagram showing, for each performance mode, a relationship between a tone synthesis method assigned to a sound generation channel and a key code.

FIG. 4 is a block diagram of an address generator 6.

5A to 5D are explanatory diagrams of loop reproduction.

6A to 6G are diagrams showing a process of changing phase angle data in the modulator 7 according to the first embodiment.

FIG. 7 is a block diagram of a first embodiment of the modulator 7.

8A to 8C are various waveform charts generated by the first embodiment of the modulator 7.

9 (a) to 9 (g) are diagrams showing examples of modulated waveform data.

FIG. 10 is a block diagram of a second embodiment of the modulator 7.

FIG. 11 is a block diagram of a third embodiment of the modulator 7.

12 (a) to 12 (c) are envelope waveform diagrams of musical tones.

[Explanation of symbols]

 1 keyboard section 2 control section 3 performance mode selection input section 4 PCM method / modulation method / selection section 5 key code register section 6 address generation section 7 modulation section 8 selection circuit 9 memory section 10 interpolation section 11 envelope generation section 12 multiplication section 13 Accumulator 14 D / A converter

Claims (2)

(57) [Claims] 1. A musical sound synthesizing operation for accessing a waveform data storage means storing PCM waveform data to output a musical tone waveform, and a musical sound for accessing a waveform data storage means storing modulation waveform data to output a musical tone waveform. In a musical tone generating device capable of appropriately switching between waveform synthesizing operation and generating a musical tone signal corresponding to each musical tone synthesizing operation, when accessing the waveform data storage means storing the PCM waveform data, frequency information to be supplied. The frequency specified by
When the sampling frequency of the waveform data is converted to a reference frequency, an address signal corresponding to the converted frequency is generated and accessed, and the waveform data storage means storing the modulation waveform data is accessed. The frequency specified by the frequency information is converted into a frequency based on the sampling frequency of the modulation waveform data, an address signal corresponding to the converted frequency is generated, and the address signal is appropriately modulated. A musical tone generating apparatus having control means for accessing. 2. The waveform data storage means for storing the PCM waveform data and the waveform data storage means for storing the modulation waveform data are a common waveform memory, and the PCM waveform data and the modulation waveform data are the waveforms. The musical tone generating apparatus according to claim 1, wherein the musical tone generating apparatus is stored in different address areas of a memory .