JPS59185391A - Musical sound generator - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field The present invention relates to a detailed musical tone generator used in a child musical instrument, etc., and particularly to a device for repeatedly reading out a plurality of periodic waveforms stored in a memory.

Prior Art To obtain a high-quality musical tone (a musical tone similar to the sound of a natural instrument), the musical waveform of the entire period from the start of the sound to the end of the sound is stored in advance in a waveform memory, and this is stored in a waveform memory. There is something to be read out. This method is particularly suitable for generating percussive musical tones. However, one drawback is that it requires a very large amount of memory and is expensive. In order to improve this point, a part of the multi-cycle waveform of the entire sound generation period is stored in the waveform memory.
By repeatedly reading this, a tone waveform signal over the entire sound generation period is obtained, while an envelope waveform signal is generated by an envelope generator, and an envelope is given to the tone waveform signal by this envelope waveform signal. Takes out a multi-period waveform, converts it into a flat envelope waveform (makes each waveform amplitude value of the multi-period waveform the same), stores it in a waveform memory, and then repeatedly reads out the stored waveform. This applies an attenuation envelope to the read waveform.

However, although this method can reduce the capacity of the waveform memory, it requires a complicated and expensive envelope generator, so there is room for further improvement.

OBJECT OF THE INVENTION An object of the present invention is to produce high-quality musical tones (especially percussive musical tones) with a simpler structure than conventional ones.
, j. The object of the present invention is to provide a musical tone generating device that can generate sounds.

SUMMARY OF THE INVENTION According to the present invention, one or more cycle waveforms stored in a waveform memory are repeatedly read out, and the levels of the one waveform read out are sequentially read out at a predetermined ratio for each repeated readout cycle. By lowering the waveform memory, it is possible to obtain a musical tone signal with a completely non-cussive envelope (easily used with a special envelope generator).Preferably, the waveform memory contains , storing at least a plurality of periods and a waveform to which an attenuation envelope from a first level A'a to a second level b is given in advance;
This tenth cycle waveform part is repeatedly read out, this repeated readout is memorized, and while this multiple cycle waveform is repeatedly read out, each repeated readout cycle is shown in Fig. 1(b).
As shown in , the level control coefficient is changed sequentially (from 1 to 3k, geometrically at a ratio of -1, with the initial value being ``1'').
k...), and the level of the read waveform for each repeated cycle is controlled by the level control coefficients that change in this way. As a result, a musical tone signal to which a percussive envelope is applied as a whole is obtained as shown in 1:9(c). In other words, by setting the common ratio l (upward), the final level of the multi-period waveform read out in one cycle will match the first level of the multi-period waveform read out in the next cycle, and the repeated The connections between the read waveforms flow smoothly.

By the way, generally speaking, the musical waveform changes in a complicated manner during the rising portion of the sound, and thereafter the waveform changes little and remains stable. Therefore, in the waveform memory, a first multi-period waveform is stored corresponding to the rising portion of the sound, a second multi-period waveform is stored corresponding to the subsequent portion, and the first multi-period waveform is stored in the waveform memory. After reading once, the second multi-period waveform portion is repeatedly read out (preferably by rubbing; this causes
It is possible to further improve the quality of the obtained musical tone signal. To illustrate this point in a diagram, as shown in FIG. 2(a), the first multiple synchronous waveform portion A (!: second multiple period waveform 115 minutes B) is stored in the waveform memory in advance, Part A
After reading out once, part B is read out repeatedly. As shown in FIG. 2(b), the level control coefficient is set to "1" when reading part A, and is equalized at a ratio of k-1 when reading part B repeatedly, as in FIG. 1(b). change relatively. As a result, a musical tone signal as shown in FIG. 2(C) is obtained.

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will now be described in detail with reference to the accompanying drawings.

The electronic musical instrument shown in FIG. 3 implements the present invention according to the method shown in FIG.
) A first multi-synchronized waveform part A corresponding to the rising part of a sound, and a second multi-period waveform part B corresponding to the rising part
is stored in advance.

The keyboard circuit 11 detects a key pressed on the keyboard, outputs a key code KC indicating the pressed key, and outputs a C key-on pulse KONP corresponding to the start of pressing the key. The note clock generating circuit 12 generates a note clock pulse having a frequency corresponding to the pitch of the pressed key in response to the 3%1 key code KCi given from the keyboard circuit 11.

In order to control reading of the waveform memory 10, an address counter 16, a repeating 7 address data supply circuit 14,
An end address detection circuit 15 is provided. Each sample point of the multi-period waveform portions A and B stored in the waveform memory 10 is assigned a consecutive address from 0 to a predetermined final value21. The first multi-cycle waveform part A is stored in the storage area from address O to just before a predetermined repeat address, and the second multi-cycle waveform part A is stored in the storage area from the predetermined repeat address to the end address (final value). Waveform portion B is stored and stored.

The counter 16 is reset at the beginning of pressing a key by the key-on pulse KON P Ic given from the keyboard circuit 11, and then counted up according to the note clock pulse corresponding to the pressed key given from the note clock generation circuit 12. . The count output of this counter 1.5 is used as an address signal from the waveform memory 10 to Me.
The tt waveforms are read out sequentially.

Therefore, while the count value of the counter 16 increases sequentially from 0 until reaching the predetermined repeat address, the first multi-cycle waveform portion A corresponding to the rising edge of 1 is read out, and the count value continues to increase. The second multi-cycle waveform portion B is read out from the predetermined repeating address value to the end address value while increasing sequentially.In the end address detection N path 15, the count value of the counter 16 is the end address. is detected, and based on this detection, a preset control signal is given to the counter 16.The preset data input PD of the counter 16 is given a repeating address value from the repeating address supply circuit 14, and responds to the end address detection. This repetitive address value is preset in the counter 16 by the preset control signal.After the first and second multi-period waveform portions A and B have been read out one time after being stored in the waveform memory 10, the counter 16 is 11: The count value of RO returns to the address by 1, and the second multi-cycle waveform portion B is read out again.Every time the count value reaches the end address (
・Since the repeat address is preset, the increase from the repeat address to the end address will not be repeated.
The second multi-period waveform portion B is repeatedly read out by 1 to 3t.

The musical waveform signal read out from the waveform memory 10 is applied to a multiplier 16, where the level is controlled according to the level control coefficient K.The level-controlled musical waveform signal is converted into an analog signal by a digital/analog converter 17. and reaches the sound system 18.

The level 1 lil J
In order to change this, a circuit 19, a multiplier 20,
A selector 21, a register 22, and an OR circuit group 26 are provided. The entire one-chip supply circuit 19 has a common ratio 1< corresponding to the attenuated second multi-period waveform part stored in the waveform memory
(=') is output and given to the multiplier 21]. The selector 21 normally selects “0” from the OR circuit group 26.
" n signals given to the input are selected, and the end address is detected by the end address detection circuit 15.
That is, when reading of part B for one cycle is completed, the output signal of the multiplier 20 given to the "1" input is selected. The register 22 takes in and stores the output signal of the selector 21 according to the system clock pulse φ.

The output signal of the register 22 is given to the multiplier 16 as a level control coefficient, and is also passed through the OR circuit group 26 to the "0" input of the multiplier 20 and the selector 21.

The OR circuit group 26 responds to the key-on pulse KONP by
Is the data with all bits “1” or the output data of register 22? This is what is output. When the key-on pulse KONP becomes ``1'' at the beginning of pressing the key, data of all bits ``1'' is input to the OR circuit group 26, and from the OR circuit 26 via the rOJ input of the selector 21. The data is loaded into the register 22.

The key-on pulse KONP is '0''' (7:: After that, it is taken into the register 22 and all 1 pulses are output.)
” data circulates through the OR circuit group 26 and the “O” input of the selector 21 and is held in the register 22.

In this way, the first multi-period waveform part A and the second multi-period waveform part B are read out one by one from the waveform memory 10: the tree is at level 5 and the control coefficient K maintains a total pitch of 1'''. The total pitch of this coefficient "1" corresponds to 11 in decimal notation, and the musical waveform signals of parts A and B read out from the waveform memory 10 are output to the multiplier 16 at the same level (or at the maximum level). means that it is output from.

When the end address is first detected, the multiplier 20
So, when the common ratio is 1, all the bits are "1"', that is, the decimal number is "1".
” is output, that is, the multiplication result k.

Therefore, when the first end address is detected, the common ratio k is taken into the register 22, and k is maintained as the level control coefficient eK during one subsequent read cycle of the multi-period waveform portion H. When the next end address is detected, the result of the multiplication of ni = ni and k]<2 is taken into the register 22; When the next end address is detected, the result of the multiplication of k = k2 and k is taken into the register 22). In this way,
The level control coefficient f changes geometrically with the rate at which the repeated readout cycle of the second multi-period waveform portion B progresses as shown in FIG. 2(b). do.

In this way, the level is controlled by a coefficient K that varies by 3%.
As shown in FIG. 2(C), a percussive envelope signal is applied to the entire signal, and a one-chord waveform signal is outputted from the multiplier 16.

In addition, theoretically (well, by raising the power of k-1, the value of the gain coefficient is 1 (0 means that n becomes a dog (accompanied by this, it approaches "0" but never becomes rOJ, so by K) The level of the musical waveform signal that is pushed to the side is also "0" (which should never happen. However, in reality, the lower digits below the predetermined effective digits of the multiplication result in the multiplier 20 are discarded, so K becomes "0" (!), and the musical tone must be surely muted. To stop the operation (rubbing is a good way to prevent noise).

In Fig. 3, a common musical sound waveform is used for each weight (each pitch), and by changing the readout rate according to the note clock, a musical sound waveform signal corresponding to the pitch of each weight is generated. As shown in Figure 4, each weight (each pitch) is independent (
This multi-cycle waveform may be stored and read out at a constant readout rate. The waveform memo IJ10A shown in Fig. 4 stores in advance a multi-period waveform as shown in Fig. 2(a) for each key, and the storage area for the waveform corresponding to each weight is separate. Identified by corresponding start and end addresses. -For example,
The waveform storage capacity for one key is 20 kilowords, and the start address of each spindle is located every 20 kilowords (for example, the start address of the lowest key C2 is "0", and the start address of the next rlc+2 is located). is r20000
J, and the start address of the next D2 is r4
-0000 J).

The start address ROM (ROMft read-only memory) 24 stores a start address corresponding to each weight, and the repetition address ROM 25 stores a repetition address corresponding to each weight. The start address and repetition address corresponding to the pressed key are read out from both ROMs 24 and 25 according to the key code KC given by the keyboard circuit 11.The selector 26 reads the end address from the ROM 24 according to the output of the detection circuit 15A. This selects one of the outputs of .25 and repeats the address RO'IVI 2 when detecting the end address.
Select output 5 and start address R when other than that pillar.
Select the OMi 24 output. The output of the selector 26 is the preset data manual PDI of the address counter 13A.
This is given. Start of key press (Keyboard circuit 11
A preset command is given to the counter 13A via the OR circuit 27 by the key-on pulse KONP output from KONP, and the start address read from the ROM 24 and selected by the selector 26 is preset to the counter A. . The counter 13A is counted up by fixed clock pulses unrelated to the key (pitch).

In this way, counting starts from the start address corresponding to the pressed key, the count value increases sequentially at a constant rate, and the multi-period waveform corresponding to the pressed key (see Fig. 2) is stored in the waveform memory 10A. Parts A and B of a)
) are read out sequentially starting from the 7th bastard address. Counter 16A for end address detection time 1 bath 15A
It detects whether the count value of has reached the end address (f), and outputs a signal "]" when the end address is detected. Based on this end address detection signal, the selector 26 selects the output of the repeat address RoMzb, and ``1'' is given to the preset command PS of the counter 13A via the circuit 27. When the reading of the second multi-period waveform portion B is completed, the repeat address is preset to the counter 15A, and the Reading is repeated.

If the waveform storage capacity for one key is 20 kilowords (predetermined value) as in the previous example, the overflow signal every 20,000 (predetermined [direct]) of the count value of the counter 16A is used as the end address detection signal. There are old things that can be used,
It is not necessary to provide a special end address detection circuit 15A.

The common ratio ROM 28 stores the common ratio k of the level control function aK corresponding to each spindle individually.
The common ratio k of the values corresponding to the pressed keys is read out in accordance with the key code KC given from 1 to 1.

Second multi-period waveform portion B to be repeatedly read out
The first level a and the last level b are different for each key. Naturally, the ratio l (-± also differs for each weight. Therefore, the value of the common ratio corresponding to each weight is The selection operation of the selector 21 is controlled by the end address detection signal as in FIG. 3. In addition, the same reference numerals as in FIG. 3 are used in FIG. The other circuits that are shown in the diagram are circuits with the same function.

When implementing the method shown in FIG. 1, the waveform shown in FIG. 1(a) is stored in the waveform memo IJ10, 1OA shown in FIGS. The repetitive address ROM 25 is removed, the counter 16 is reset when the end address is detected, and the counter 13A is
All you have to do is preset the start address.

Although FIGS. 3 and 4 show single-tone electronic musical instruments, it goes without saying that the present invention can also be practiced with multi-tone electronic instruments. In that case, a key assigner (means for assigning a pressed key to one of a specific number of sound generation channels) is provided in conjunction with the keyboard circuit 11, and the address counter 16. Assigned to each channel from l″l.

The musical waveform signals of the keys are read out in a time-sharing manner, and the register 22 is used as a shift register with stages corresponding to the number of channels to supply the level control coefficient (calculation to the power of k) for each channel (this time-sharing). You can do it with .

It goes without saying that the musical tone generating device of the present invention can be used not only for generating scale tones as shown in the above embodiments, but also for generating rhythm tones (percussion instrument tones) in automatic rhythm performance devices and the like.

According to this invention, since multiple periodic waveforms are memorized and read out in their entirety, the resulting musical tone signal can be made into a commercial product, and by repeatedly reading out the multiple periodic waveforms, all pronunciations can be achieved. γ which is the configuration to obtain the period waveform
Therefore, since the storage capacity of the waveform memory can be relatively reduced, and the level of the readout waveform is sequentially lowered at a predetermined ratio for each repeated readout cycle, a special envelope generator is provided. without any
Attenuation envelopes can be applied extremely easily. Therefore, commercial quality musical tones can be produced with a simpler structure (thus, compared to the conventional method).

[Brief explanation of the drawing]

Figure 1 is a diagram showing the basic concept of the musical tone generation method according to the present invention, Figure 2 is a diagram showing the basic concept of the musical tone generation system according to the invention from a different perspective, and Figure 3 is a diagram showing an embodiment of the present invention. FIG. 4 is an electrical block diagram of an electronic musical instrument according to another embodiment of the present invention. 10.10A... Waveform memory, 11... Dance board circuit,
13, 13A...address counter, 14...repetitive address data supply circuit, 15, 15A...end address detection circuit, 16...multiplier for level control, 19...common ratio supply circuit, 20... Multiplier, 21
...Selector, 22...Register, 28...Common ratio R
OM. Patent applicant Yoshihito Iizuka (19) 664-

Claims (1)

[Claims] 1. A waveform memory for storing waveforms in advance;
A reading means for reading out the multi-period waveform stored in the waveform memory and repeatedly reading out a predetermined multi-period waveform portion; A musical tone generator comprising level control means for sequentially lowering the level at a predetermined ratio. 2. The waveform memory has a first level a and a second level a.
At least a plurality of periodic waveforms are stored in advance with attenuation envelopes up to 7 bells b, and the reading means stores a plurality of periodic waveforms to which attenuation envelopes from the first level a to the second level b are applied. 2. The musical tone generating device according to claim 1, wherein the periodic waveform portion is repeatedly read out, and the level control means (the level control means) is geometrically lowered by a negative ratio for each repeated readout cycle. 3. The waveform memory stores in advance a first multi-period waveform including a multi-period waveform from the onset of the sound and a second multi-period waveform following this, and the reading means 2. The musical tone generating device according to claim 1, wherein after reading out the first multi-period waveform once, the second multi-period waveform is read out repeatedly.