JPH0318197B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an electronic musical instrument that achieves an effect of gradually changing the pitch such as a slur or portamento, and particularly relates to an electronic musical instrument that realizes an effect of gradually changing the pitch such as slur or portamento, and in particular, it relates to an electronic musical instrument that realizes an effect of gradually changing the pitch such as slur or portamento. This invention relates to making the waveform for the rising portion different from the waveform for the normal rising portion (to which no pitch change is applied).

[Conventional technology]

The effect of smoothly changing the pitch of a musical tone from a pitch corresponding to a first key to a pitch corresponding to a second key is conventionally known as portamento or slur (hereinafter referred to as slur in this specification). Japanese Patent Laid-Open No. 139095/1983 shows that a more natural slur effect can be achieved by smoothly changing not only the pitch but also the timbre during the period in which the slur is applied. has been done. In Japanese Patent Application Laid-Open No. 59-139095, by adding and synthesizing two series of musical tone signals with somewhat different tones at a level ratio that changes over time, a temporal change in tone color when adding a slur is realized as a result. It is becoming more and more like this. Specifically, the tones in one series are key-scaled according to the height of the key that was pressed just before, and the tones in the other series are key-scaled according to the pitch of the key that was pressed this time. It has a unique tone. Therefore, even though a timbre change is applied during a slur, it does not mean that a timbre change specific to the slur is applied, but rather the timbre changes from the timbre keyscaled by the previously pressed key to the timbre keyscaled by the currently pressed key. The timbre of the generated sound simply changed smoothly.

[Problem that the invention seeks to solve]

By the way, in natural musical instruments, there is a remarkable difference between the timbre of the rising part of a musical tone and the timbre of the sustaining part, and even the rising part of a new tone in a slur exhibits an independent timbre change. Furthermore, the timbre characteristics of the rising part of the new sound in this slur are closer to the timbre of the sustaining part than the normal timbre of the rising part. It has been difficult to realize such a timbre change peculiar to slurs only by controlling the key-scaled timbre to smoothly transition as in the prior art described above. In particular, in order to imitate the above-mentioned difference in timbre between the rising and sustaining parts of a musical tone, an electronic device employs a musical sound generation method that generates musical sound waveforms with different timbre characteristics in the rising and sustaining parts. In musical instruments, a musical sound waveform is generated that has the timbre characteristics of a normal rising part of a musical tone, regardless of whether it is a slur or not. Even if a smooth change was applied, the normal timbre characteristics of the rising part appeared strongly, and it was not possible to obtain the timbre change peculiar to slurs.

This invention was made in view of the above points,
By making the timbre of the rising part of a musical sound different from the timbre of the normal rising part when adding a slur,
The purpose of the present invention is to provide an electronic musical instrument that can produce a unique timbre change during slurs.

[Means for solving problems]

In order to solve the above-mentioned problems, the electronic musical instrument of the present invention has a first mode in which the pitch of the musical tone to be generated is gradually changed from a predetermined first pitch to a predetermined second pitch at the rise of the musical tone. , or
a second mode in which the second pitch is set from the beginning;
pitch control means for controlling the pitch of the musical tone according to one of the modes, and mode selection means for selecting the first mode or the second mode;
a musical tone generating means for generating a musical tone signal having a pitch controlled by the pitch control means, and a tone color of a rising portion of the musical tone signal generated by the musical tone generating means being different depending on the first or second mode. It is characterized in that it is equipped with a timbre control means.

According to the first invention, the musical sound generating means can selectively generate a musical sound waveform for a rising part and a musical sound waveform for a sustaining part, and the tone control means can generate a musical sound waveform in the first mode. A tone waveform for the sustaining portion is generated at the rising edge of the signal, and in the second mode, a tone waveform for the rising edge is generated at the rising edge of the musical tone signal.

According to the second invention, the musical sound generation means can selectively generate a musical sound waveform for a rising part and a musical sound waveform for a slur, and the timbre control means can generate a musical sound signal in the first mode. A tone waveform for a slur is generated at the rising edge of the tone signal, and a tone waveform for the rising edge is generated at the rising edge of the tone signal in the second mode.

In one embodiment, the selection of the first or second mode is made in response to a key operation method on a keyboard for specifying the pitch of a musical tone to be generated. for example,
When the legato style is used as the key operation method when switching the pressed key, that is, when the key operation method is used in which a new key is pressed before the previously pressed key is completely released, the first mode is selected. On the other hand, when a key operation method in which a new key is pressed after the previously pressed key is completely released, that is, a staccato method is used, the second mode is selected.

[Effect]

When the first mode is selected, the pitch of the musical tone to be generated gradually changes from a predetermined first pitch to a second pitch at the rise of the musical tone,
A pitch change similar to a slur is added. When the second mode is selected, the pitch of the musical tone to be generated is set to the second pitch (the pitch corresponding to the pitch of the musical tone) from the beginning, so pitch changes such as slurs are not applied. Under the control of the timbre control means, the timbre of the rising edge of the musical tone signal generated by the musical tone generating means is made different depending on the first or second mode. For example, in the first mode,
At the rising edge of a musical tone signal, a musical sound waveform for a sustained part or a musical sound waveform for a slur, which can achieve a unique tone at the time of a slur, is generated, whereas in the second mode, a musical sound waveform for a normal rising part is generated. generates a musical sound waveform. In this way, it is possible to make the timbre of the rising part of a musical tone when adding a slur different from the timbre of a normal rising part, and it becomes possible to realize a timbre change peculiar to the time of a slur.

FIG. 1 schematically shows one embodiment using a waveform diagram for understanding the present invention, in which the first musical tone that precedes the sounding order rises in the second mode (that is, in staccato form). key pressed, no slur pitch change is applied at the rising edge),
An example is shown in which the second musical tone rises in the first mode (that is, the key is pressed in a legato manner, and a slur pitch change is applied at the rising part). Since the first musical tone rises in the second mode, it is generated from the beginning at the first pitch corresponding to its original pitch, and the musical sound waveform of the rising part is a normal musical sound waveform for the rising part. . As is known, when the tone waveform for the rising part is completed, the tone waveform for the sustaining part is used. On the other hand, since the second musical tone rises in the first mode, its pitch gradually changes from the first pitch to the second pitch corresponding to its original pitch, and the musical sound waveform of the rising part changes gradually. The musical sound waveform for the sustained part is used from the beginning. The amplitude envelope of the sustain part musical waveform used at the rising edge of the second musical tone rises with a predetermined attack characteristic, while the amplitude envelope of the first musical tone falls with a predetermined decay characteristic.

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

〔Example〕

FIG. 2 shows an embodiment of the present invention in a single-tone electronic musical instrument, in which the tone generator 10 is
When adding a slur, the previous musical tone and the current musical tone are generated independently, and the amplitude of the previous musical tone is controlled by the decay envelope, and the amplitude of the current musical tone is controlled by the attack envelope, so that the two tones are crossfade. It is equipped with musical tone generation channels for each channel.

The keyboard 11 includes a plurality of keys for specifying the pitch of musical tones to be generated. Key press detection circuit 1
2 detects key presses and key releases on the keyboard 11;
If there is a key that has changed from a released key to a pressed key, a key-on event signal is sent along with the key code KC representing that key.
KONEV is output, and if there is a key that has changed from being pressed to being released, the key-off event signal KOFEV is output together with the key code KC. The output of the key press detection circuit 12 is given to the microcomputer section 13. In the microcomputer section 13, the keyboard 11
When a plurality of keys are pressed, a single note selection process is performed in which one of the keys is selected for sounding, and a process is performed in which two channels for musical tone generation are assigned so that they are alternately used. The output of the microcomputer section 13 is given to the interface 14,
Based on this, the interface 14 outputs the key code KC of the musical tone to be generated, the key-on signal KON, the slur mode signal SLM for controlling envelope signal generation during the slur application period, and the forced dump signal FD. Further, in order to control the generation period of the slur mode signal SLM, a slur off pulse SLOFP is applied from the envelope generator 15 to the interface 14.

The pitch information generation and slur adding circuit 16 inputs the key code KC of the musical tone to be generated given from the interface 14, generates pitch information SKC representing the pitch of the key indicated by the key code KC, and generates a slur effect. When the value of the input key code KC changes from a certain key (first key) to another key (second key), the pitch of the first key gradually changes from the pitch of the first key to the pitch of the second key. Pitch information SKC representing the pitch that changes to is generated. The pitch control means capable of controlling pitch changes for such slurs, that is, the pitch information generation and slur adding circuit 16, can be constructed using the known technology disclosed in, for example, Japanese Patent Laid-Open No. 54-107722 and others. Since it is possible to do so, its internal details will not be particularly explained.

The slur control operator 17 includes a slur switch 17a and a slur speed selector 17b for selecting on/off of the slur effect, and the output of the switch 17a is transmitted to the microcomputer section 13.
is input to the slur adding circuit 16 to control whether or not to add a slur, and the output of the selector 17b is input to the slur adding circuit 16 to control the slur speed (pitch change rate or pitch change time). In this embodiment, turning on the switch 17a does not necessarily add a slur, but whether or not to add a slur is automatically determined depending on the key pressing method. That is, when the switch 17a is on, if the key press method is legato, a slur effect is applied (this corresponds to the first mode), and if the key press method is staccato, no slur effect is applied (this corresponds to the second mode). mode). This key press operation method is determined by the microcomputer section 13, and when a slur is to be added, that is, in the first mode, the slur mode signal SLM output from the interface 14 becomes "1". Although this slur mode signal SLM is not given to the slur adding circuit 16, the slur adding circuit 16 recognizes that the output of the slur switch 17a is on and that the value of the input key code KC is different from one key to another. Based on the continuous change in the key, it is determined that a slur should be added, and pitch information SKC that instructs pitch change is generated as described above.
Note that the slur effect on/off selection and slur speed selection are not limited to selection operations using the dedicated switches 17a and 17b, and may be performed automatically in conjunction with the tone color selection in the tone color selection circuit 18.

The tone generator 10 independently generates a musical tone signal having a pitch according to the pitch information SKC given from the circuit 16 through two musical tone generation channels, controls the tone, volume, etc. of this musical tone signal in each channel, and controls the final tone signal. The musical tone signals of both channels are then added together and given to the sound system 19. This tone generator 10 includes a timbre control means, and controls the timbre of the rising edge of the musical tone signal generated by the tone generator 10 between a first mode (that is, when adding a slur) and a second mode. (when not adding a slur) and when not adding a slur. In order to control tone switching according to such a mode, the output of the microcomputer section 13 is provided to the tone generator 10 via the interface 14 as necessary.

The envelope generator 15 generates envelope signals VL for two channels (that is, first and second envelope signals VL1, VL2) in a time-division manner and supplies them to the tone generator 10. Each channel of the tone generator 10 controls the volume, tone, etc. in accordance with the corresponding envelope signal VL. Note that the two musical tone generation channels in the tone generator 10 may be time-sharing. The signals KON, SLM, and FD given from the interface 14 to the envelope generator 15 are given in two channels in a time-division manner according to the allocation in the microcomputer section 13, and based on these, the envelope signal of each channel is given.
VL (VL1, VL2) is formed in a time-division manner. Further, a key code KC is given to the envelope generator 15 from the interface 14 for key scaling of the envelope signal VL, and a timbre selection information TC is given from the timbre selection circuit 18 to control the envelope signal according to the timbre. is given.
Furthermore, an envelope signal according to the slur rate
Slur selector 17 to control VL changes
Slur speed data SSD is given from b.
Further, the envelope signal VL is also given to the microcomputer section 13.

The microcomputer section 13 executes a key-on event program as shown in FIG. 3 when the key-on event signal KONEV is given, and executes a key-off event program as shown in FIG. 4 when the key-off event signal KOFEV is given. Execute.

In the key-on event program, first, the key code KC given from the key-press detection circuit 12 is taken into the pressed key code register KCR together with the key-on event signal KONEV (block 20).
This register KCR has multiple storage locations,
All key codes KC of keys currently being pressed can be stored. In block 21, the register
One key code is selected from among the pressed key codes stored in the KCR according to a predetermined priority selection criterion, and is stored in an internal register as a priority selection key code NKC. As the priority selection criterion, an appropriate criterion such as giving priority to the highest note or lowest note, or giving priority to the last note (keys pressed later are given priority) may be adopted. On the other hand, the key code PKC of the key currently being sounded is stored in another internal register, and block 22 checks whether the priority selection key code NKC detected in block 21 is the same as the key code PKC currently being sounded. Find out. If they are the same, exit this program without performing any remaining processing (return). In the case of the last-arrival priority criterion, block 22 does not normally become YES, but in the case of the highest or lowest tone priority criterion, a key-on event may occur even for keys that are not prioritized, so block 22 becomes YES. There is.

When the priority selection key code NKC is not currently being sounded, that is, when the musical tone to be generated should be switched from the one corresponding to the currently sounding key code PKC to the one corresponding to the priority selection key code NKC,
The NO in block 22 leads to block 23 and the program continues. Block 23 checks whether the slur switch 17a is turned on.
If it is on, check whether the current key-on event was any key-on. (Block 2
4). Any key on means that there are no keys pressed other than the key pressed this time (in other words, a key is pressed for the first time when no keys are pressed), and this is the pressed key code register KCR. This can be determined by checking whether there is only one key code or two or more stored in the key code. If any key on is YES,
This means that the key has been pressed in a staccato style key operation, and the content of the slur register SL is set to "0" so that no slur effect is applied (block 25). On the other hand, if any key-on is NO, it means that the key was operated in legato style, and the content of the slur register SL is set to "1" to apply a slur effect (block 26). If the slur switch 17a is off, the slur register SL is set to "0" without determining whether any key is on (block 27). In this way, through the processing of blocks 23 to 27, it is determined whether a slur should be added, that is, it is determined whether the first mode or the second mode has been selected.

The channel flag CHFLG indicates which of the two channels the sound to be produced is assigned to, and the channel to be assigned is switched alternately each time the key to be produced changes. In block 28, the channel flag
Find out which channel CHFLG is pointing to. If the first channel is indicated, check the contents of the slur register SL in block 29,
If “1” (when adding a slur, that is, when the first mode is selected), block 30,
After executing the process in block 31, proceed to block 32.
If “0” (when no slur is added, that is, when the second mode is selected), block 30,
Skip block 31 and proceed to block 32. In block 30, the forced dump pulse of the second channel
Output FDP2. In block 31, it is checked whether the level of the envelope signal VL2 of the second channel has become zero based on the envelope signal VL given from the envelope generator 15 (FIG. 2), and the program waits in this step until it becomes zero. When it is determined in block 28 that it is the first channel, it means that the front sound is assigned to be pronounced in the first channel, and the current sound related to New Key On is newly assigned to be pronounced in the second channel. This means that it should be assigned to . Therefore, the forced dump pulse FDP2 of the second channel
(block 30), and based on this, if the current level of the second channel envelope signal VL2 generated by the envelope generator 15 is not zero, it is forced to be rapidly dumped and made to be zero. This allows a new sound to be assigned to the second channel.

In block 32, a first channel key-off pulse KFP1 is output, and based on this, the first channel envelope signal VL1 generated by the envelope generator 15 enters the decay state. In this example, the microcomputer section 13
Since the key-off pulse KFP1 and other pulses are created by decoding the address data output from the interface 14 inside the interface 14, the pulse output processing in the block 32 and other blocks means that the pulse itself is sent from the microcomputer section 13. Rather than outputting the pulse, it is to designate the address corresponding to the pulse to be output by writing some data to the address corresponding to the pulse. In this way, the address code of the specified address is given from the microcomputer section 13 to the interface 14 via the address bus, and by decoding this address code, a desired pulse can be obtained.

In block 33, the key code selected with priority along with the key-on pulse KOP2 of the second channel is
Output NKC. In other words, as mentioned above, the pulse
Write the key code NKC to the address corresponding to KOP2, output the address code to the address bus, and output the key code NKC to the data bus.

Block 34 checks again to see if any key is on. NO In other words, if it is a legato type key press operation, the key code currently being sounded
Write the prioritized key code NKC to the internal register that stores PKC and set "PKC=NKC"
(block 35). Then slur register SL
Check if is “1” (block 36), and check if YES
If so, output the slur mode pulse SLP (block 37), and finally switch the channel plug CHFLG to the content indicating the second channel (block 3).
8). When the slur register SL is "0", the process advances to block 38 via block 39. In block 39, the first channel forced dump pulse
FDP1 is output, and based on this, when the slur is off, the level VL1 of the envelope signal of the first channel (that is, the envelope signal of the previous sound) is forcibly ramped up.

On the other hand, if Any Key On is YES (that is, a staccato type key press operation), the process advances from block 34 to block 40, and the key code currently being sounded is
Write the selected key code NKC into the internal register that stores PKC, and set "1" in the internal key-on register KONR to indicate key press. Then proceed to block 38.

1 when it is determined in block 28 that the channel flag CHFLG indicates the second channel.
A routine 41 surrounded by a dotted chain line is performed. In this routine 41, processing similar to the routine from blocks 29 to 38 described above is performed for the opposite channel. That is, the pulses or signals FDP2, VL in the routine from blocks 29 to 38
2. Processing regarding KFP1, KOP2, and FDP1 (processing of blocks 30, 31, 32, 33, and 39) is performed on channels opposite to each of them, that is, forced dump pulse FDP1 of the first channel, envelope signal VL1, and envelope signal VL1 of the second channel. Key-off pulse KFP2, 1st channel key-on pulse KOP1, 2nd channel forced dump pulse
Concerning FDP2. Finally, the channel flag CHFLG is switched to the first channel (block 42).

In the key-off event program shown in FIG. 4, first, the key code KC given from the key press detection circuit 12 is fetched together with the key-off event signal KOFEV, and the key code in the pressed key code register KCR corresponding to the fetched key code KC is stored. (block 43). In block 44, the third
The same priority selection process as block 21 in the figure is performed.
This is to review the priority order of the key codes in register KCR, which may change due to key-off. If all the keys are off, the priority selection key code NKC is a predetermined code indicating that, for example, all bits are “0”.
The code for is stored. Block 45 checks whether all keys are off. If YES, key-off pulse KFP1 or KFP corresponding to the channel indicated by channel flag CHFLG
2 (block 46), key-on register
Reset KONR to “0” (block 47),
Exit this program. If all keys are not off, it is checked in block 48 whether "NKC=PKC", and if YES, this key-off event is not caused by the key that is currently sounding, so this program ends. (return).

If the key that is currently sounding, that is, the key corresponding to the key code PKC, is turned off, the newly detected priority selection key code NKC and PKC do not match, and block 48 becomes NO. Next, it is checked whether the slur switch 17a is on, and the slur register SL is set to "1" or "0" accordingly (blocks 49, 50, 51).

Block 52 examines the channel flag CHFLG and proceeds to block 53 or 54 accordingly. If CHFLG indicates the first channel, proceed to block 53 and proceed to block 2 of FIG.
The same processing as 9, 30, 31 is performed in blocks 53, 5.
5,56. Next, block 57 outputs the first channel key-off pulse KFP1, and further outputs the second channel key-on pulse.
Outputs the priority selected key code NKC along with KOP2, and then writes NKC to the PKC register. Next, check whether the slur register SL is "1" or not, and if SL = "1", the channel flag is set after sending the slur mode pulse SLP.
CHFLG is switched to the second channel (blocks 58, 59, 60). If SL="1",
SLP outputs forced dump pulse FDP1 of the first channel without sending out (block 61), and then
Switch CHFLG to the second channel.

On the other hand, if it is determined in block 52 that CHFLG indicates the second channel, the process proceeds to block 54. Blocks 54, 62 and 63 perform the same processing as blocks 53, 55 and 56 for signals FDP1 and VL1 of opposite channels.
Next, in block 64, the same process as in block 57 is performed on pulses KFP2 and KOP1 of the opposite channel.
, and sends the key code NKC and writes the NKC to the PKC register. When adding a slur, a slur mode pulse SLP is output, and then the flag CHFLG is switched to the first channel (blocks 65, 66, 67). If no slur is to be added, a forced dump pulse FDP2 of the second channel is output (block 68).

FIG. 5 shows a detailed example of the interface 14. An address code is given from the microcomputer section 13 to this interface 14 via an address bus 70 in accordance with the output timing of each pulse KOP1 to SLP described above, and this is applied to a decoder 71 corresponding to each pulse KOP1 to SLP. decoded to the output line. Further, data (particularly the key code NKC) outputted from the microcomputer section 13 along with the address code is provided to the latch circuit 73 via the data bus 72.

The key-on pulse KOP1 and key-off pulse KFP1 of the first channel are input to the set input S and reset input R of the flip-flop 74, respectively.
Its output Q is the key-on signal for the first channel.
It is applied to the A input of the selector 75 as KON1. Similarly, the key-on pulse of the second channel
KOP2 and key-off pulse KFP2 are input to flip-flop 76, and its output Q is sent to selector 75 as key-on signal KON2 for the second channel.
is applied to the B input of

Forced dump pulse FDP1 for each channel,
FDP2 is applied to one-shot circuits 77 and 78, respectively. One shot circuit 77 is pulse FDP
In response to the rising edge of channel timing pulse CH1, one pulse FD1 is output with the pulse width of channel timing pulse CH1. One shot circuit 78 generates an inverted channel timing pulse in response to the rising edge of pulse FDP2.
Outputs one pulse FD2 with the pulse width of CH1. The channel timing pulse CH1 is synchronized with the channel time division timing in the tone generator 10, and becomes "1" at the timing of the first channel and "0" at the timing of the second channel. Conversely, the inverted channel timing pulse 1 becomes "0" at the timing of the first channel and becomes "1" at the timing of the second channel. Each pulse FD1, FD2 is multiplexed via an OR circuit 79 and is a time division multiplexed dump signal.
FD is output.

A slur mode pulse SLP is applied to the set input S of flip-flop 80, and a slur off pulse SLOFP is applied to its reset input R.
The slur mode signal SLM output from the flip-flop 80 is "1" from the time when the slur mode pulse SLP is generated until the slur off pulse SLOFP is generated. Note that the clock pulse φ _M for synchronously controlling the operations of the flip-flops 74, 76, and 80 is a high-speed clock pulse synchronized with the clock of the microcomputer section 13.

The latch circuit 73 is latch-controlled by key-on pulses KOP1 and KOP2 applied via an OR circuit 81. Therefore, block 33 in FIG.
Or key-on pulse by processing of routine 41.
Only when the key code NKC is sent out together with KOP1 or KOP2, the latch contents are rewritten and the key code NKC is latched in the latch circuit 73. The output of the latch circuit 73 is synchronized via the delay flip-flop 82 by the channel timing paros CH1, and is output as a key code KC indicating the musical tone to be generated.

In FIG. 6a, key KC1 is pressed for the first time at time t1, key KC2, which has priority over key KC1, is pressed at time t2, and key KC2 is released at time t3.
The pressing of KC1 continues, and the key KC1 is pressed at time t4.
The signal states of each part in FIG. 5 are illustrated when the key is released. The priority selection key code NKC given to the data bus 72 is KC from t1 to t2.
1, and KC2 from t2 to t3, and t3
KC1 is shown from t4 to t4. In addition, the channel flag CHFLG in the microcomputer section 13
An example of the content of is also shown in Figure 6a, and at time t
It is assumed that before 1, CHFLG indicates the first channel. Then, the flag is set such that the second channel is used in the period t1 to t2, the first channel is used in the period t2 to t3, and the second channel is used in the period t3 to t4.
CHFLG is switched. Depending on this flag CHFLG, key-on pulse KOP1 of each channel,
KOP2 and key-off pulses KFP1 and KFP2 are generated at each time point t1 to t4 as shown in the figure. Therefore, the key-on signal KON2 of the second channel is generated during the period from t1 to t2.

FIG. 6b shows the states of the dump signal FD, slur mode pulse SLP, slur mode signal SLM, and others corresponding to FIG. 6a when the slur switch 17a is turned on. In this case, it is determined that any key is on at time t1, and the dump signal FD is not generated. Furthermore, it is determined that there is a key press operation in legato format at time points t2 and t3 (it is determined by the key-on event processing in FIG. 3 that a new priority key has been pressed in legato format), and the slur mode pulse SLP is generated.
Also, at time t2, the forced dump pulse FDP of the first channel is generated by the processing of routine 41 in FIG.
1 is generated, and a pulse FD1 is generated in response. Furthermore, at time t3, a forced dump pulse FDP2 of the second channel is generated by the process of block 30 in FIG.
FD2 is generated. The slur mode signal SLM becomes “1” in response to the slur mode pulse SLP,
During this time, the levels of the envelope signals VL1 and VL2 of the first channel and the second channel change in a cross-fade manner (one rises and the other falls). The change rate of the envelope signal at this time is the attack rate and decay rate for the slur. Envelope signal during slur period
When VL2 or VL1 finishes falling, a slur-off pulse SLOFP is generated and the flip-flop 8
0 (Figure 5) is reset and the slur mode signal
SLM becomes “0”. On the other hand, Pituchi Information SKC,
In response to a change in the key code, the pitch gradually changes from the pitch corresponding to the previously pressed key key code KC1 (or KC2) to the pitch corresponding to the currently pressed key key code KC2 (or KC1).

FIG. 6c shows the signal state corresponding to FIG. 6a when the slur switch 17a is off. In this case, slur mode pulse SLP is not generated. Further, at time t2, a forced dump pulse FDP2 of the second channel is generated by the processing of the routine 41 in FIG. 3, and a pulse FD2 is generated in response to this. At time t3, the first channel forced dump pulse FDP1 is generated by the process of block 39 in FIG. 3, and the pulse FD1 is generated in response. Therefore, during a predetermined period from time t2, the envelope signal VL2 of the second channel falls rapidly at a predetermined damping rate,
During a predetermined period from time t3, the envelope signal VL1 of the first channel rapidly falls at a predetermined damping rate. Further, the pitch information SKC immediately changes to the pitch corresponding to the musical tone to be generated.

An example of the envelope generator 15 will be explained with reference to FIG.

Two-stage shift register 83 shift-controlled by channel timing pulse CH1
is the envelope signal VL1 for 2 channels,
The instantaneous value of VL2 is dynamically stored, and its output is given to the tone generator 10 (FIG. 2) as an envelope signal VL (that is, a time division multiplexed signal of VL1 and VL2). Further, the output of this shift register 83 is given to an arithmetic circuit 84, and added by change width data ΔV (with positive and negative signs) indicating an increase value or a decrease value per unit operation time given via a selector 85. or subtracted. The output of this arithmetic circuit 84 is input to the shift register 83.

The target value generator 86 supplies data indicating the attained target value TG of the addition/subtraction operation result in the loop of the shift register 83 and the arithmetic circuit 84 to the A input of the comparator 87 . The comparator 87 compares the output signal VL of the shift register 83 applied to the B input, that is, the instantaneous value of the envelope signals VL1 and VL2 of each channel, with the target value TG of the A input, and determines the relationship between the values of both inputs A and B. produces an output signal in response to

The state control circuit 88 is for controlling the calculation state for envelope signal formation, and by way of example, the calculation state is for four typical parts of the envelope waveform: attack, sustain, decay, dump, and slur. It corresponds to two parts specially prepared for this, namely the slur attack and slur decay parts. The state control circuit 88 outputs a state signal ST in a time-division multiplexed manner for each channel, instructing which of the above-mentioned six parts of the envelope waveform should be subjected to calculation. Here, regarding attacks, normal attacks and slur attacks are used selectively, and slur attacks are used during the slur application period, and normal attacks are used at other times. Further, regarding the decay, normal decay and slur decay are selectively used. For state control, the key-on signal KON, forced dump signal FD, slur mode signal SLM, and timbre selection information TC for each channel, which are given from the interface 14 (FIG. 5) in a time-division manner, are used.

The target value generator 86 stores in advance the level information at the switching points of each part of the envelope waveform for each tone in a state that is key scaled according to each key (or range), and uses the state signal ST. Depending on the content, the level information of a predetermined switching point is read out and output as target value data TG. Therefore, even if the contents of the state signal ST are the same, the value of this level information is different from the tone selection information TC.
Or it will be different if the key code KC is different. To simplify the explanation, it is assumed that the level information, that is, the target value TG, has two types: a sustain level and a zero level. State signal ST is initial state S0, normal day state S4,
Dump State S5, Slurday Kay State S
6, the zero level is used as the TG, and when the slur attack state S1, normal attack state S2, and sustain state S3, the sustain level is used as the TG. The zero level is unrelated to key scaling, and the sustain level is subject to key scaling.

The change width data generator 89 includes each state S0.
~ Change width data indicating the change rate (slope) of the envelope signal in S6 is stored in advance as a key-scaled value corresponding to each tone, and is stored as a state signal ST, tone selection information TC, and key. Read according to code KC. For example, in normal attack state S2, change width data indicating the attack rate value is read out,
In the normal decay state S4, change width data indicating the decay rate value is read out, in the dump state S5, change width data indicating the dump rate value is read out, and in other states S0,
Change width data is not read in S1, S3, and S6. The change width data generated from the generator 89 is
At the gate 90, the selector 8 is selected intermittently at predetermined time intervals according to the calculation timing signal given from the calculation timing signal generator 91.
5 is applied to the A input. The calculation timing can also be controlled by the tone color selection information TC and the state signal ST.

The slur change width data generator 92 is a device that exclusively generates slur rate values in slur attack state S1 or slur decay state S6, and generates key-scaled slur rate values according to the tone and slur speed. Read out as change width data according to tone selection information TC, slur speed data SSD, and key code KC.
The slur calculation timing generator 93 generates a calculation timing signal at time intervals according to the tone selection information TC and slur speed data SSD, and sends this timing signal to the gate 95 via an AND circuit 94.
Add to. Other inputs of the AND circuit 94 receive a slur attack state signal ST1 via an OR circuit 96.
and a slurry decay state signal ST6.

The slur change width data generated from the generator 92 is applied to the B input of the selector 97 and also to the negative conversion circuit 98, where it is converted into negative data. The output of the negative conversion circuit 98 is given to the A input of the selector 97. The selector 97 selects the B input when the slur attack state signal ST1 is "1" and gives positive change width data to the gate 95, but when the slur attack state signal ST6 is "1"
When , the A input is selected and negative change width data is applied to the gate 95. The gate 95 is intermittently opened by a timing signal given from the AND circuit 94, and the positive or negative change width data is sent to the selector 8.
Give it to the B input of 5. Note that the signal ST1 becomes "1" when the state signal ST indicates the slur attack state S1, and the signal ST6 becomes the state signal
The signal becomes “1” when ST indicates the slurry decay state S6, and the state control circuit 8
Generated from 8. In this way, the change width data for the slur attack and the change width data for the slur decay have the same absolute value and opposite signs. As a result, slur attack and slur decay have the same rate of change, but have opposite slopes.

The selector 85 selects the B input when the output of the OR circuit 96 is "1", that is, when it is in the slur attack state S1 or the slur decay state S6,
When “0”, that is, other states S0, S
2 to S5, A input is selected. Therefore, when the slur attack state S1 or the slur decay state S6 is in effect, the change width data ΔV indicating the slur rate value is given to the arithmetic circuit 84, but otherwise it is given from the generator 89 via the gate 90. The change width data ΔV is given to the arithmetic circuit 84.

FIG. 8 shows an example of a state switching operation for one channel in the state control circuit 88, and similar processing is performed for two channels in a time-division manner. Initially, it is set to the initial state S0 (block 100). When the key-on signal KON rises to "1", if the slur mode signal SLM is "1", the state becomes the slur attack state S1, and if SLM is "0", the state becomes the normal attack state S2. This state S1 or S2
Then, the level of the envelope signal VL increases according to the slur rate value or attack rate value, and the sustain level is used as the target value TG.
Eventually, “VL=TG” (block 101)
(YES), changes to sustain state S3. In this state S3, it is checked whether the timbre selection information TC indicates the timbre of a sustained tone (block 102),
If so, this state S3 is maintained and the envelope signal VL is maintained at the sustain level.
Otherwise, proceed immediately to block 103;
The state changes to the normal decay state S4 or the slur decay state S6 depending on whether the slur mode signal SLM is "1" or not. Furthermore, when the key-on signal KON changes to "0" in state S3, the state changes from NO in block 104 to block 103.
and changes to daylight state S4 or S6.

In the normal decay state S4, the level of the envelope signal VL decreases according to the decay rate value, and when VL eventually reaches zero level (YES in block 105), the state is switched to the initial state S0. On the other hand, Slurday K-State S
6 the envelope signal according to the slur rate value
When the level of VL decreases and VL eventually reaches zero level, the program proceeds from YES in block 106 to block 107, outputs a slur-off pulse SLOFP, and then switches to the initial state S0.

Also, it is checked whether the forced dump signal FD becomes "1" in the middle of each state S1 to S4, S6,
(blocks 108, 109, 110, 111),
If YES, the state switches to dump state S5. In this dump state S5, the envelope signal VL
When the level of VL decreases according to the dump rate value and eventually reaches the zero level (block 11)
2 (YES), the state changes to the initial state S0. Note that typically the normal attack rate is faster than the slur attack rate, the normal decay rate is slower than the slur decay rate, and the dump rate is faster than the decay rate.

Returning to FIG. 6b, the envelope signal VL (that is, VL1, VL2) and pitch information SKC corresponding to FIG. 6a when the slur switch 17a is turned on will be described. During the period from t1 to t2, after the envelope signal VL2 of the second channel rises at a normal attack rate, it maintains the sustain level L1 corresponding to the key KC1, and during that period, the pitch information SKC maintains a constant pitch corresponding to KC1. When the key-on pulse KOP1 of the first channel, the key-off pulse KFP2 of the second channel, and the slur mode pulse SLP are generated at time t2, the key-on signal KON1 of the first channel is generated.
is “1”, second channel key-on signal KON2
becomes “0” and the slur mode signal SLM becomes “1”. As a result, block 103 in FIG. 8 regarding the second channel becomes YES, and the slurry decay state S6 becomes YES, and at the same time, block 113 in FIG. 8 regarding the first channel becomes YES.
The slur attack state becomes S1. Therefore, the envelope signal VL2 is attenuated at the slur rate, and VL
1 rises at a slur rate. On the other hand, at time t2, the content of key code KC changes from KC1 to KC2,
As a result, the pitch information SKC gradually changes from the pitch corresponding to KC1 to the pitch corresponding to KC2, giving a slur effect. During this slur application period, one envelope signal VL
The slur decay portion of the second envelope signal VL1 intersects with the slur attack portion of the other envelope signal VL1. Second
When the channel's slur day Kay ends, the 8th
Block 106 in the diagram becomes YES, and block 1
At 07, slur off pulse SLOFP is output. As a result, the flip-flop 80 shown in FIG. 5 is reset, and the slur mode signal SLM becomes "0".
In the period from t3 to t4, the same operation as in the period from t2 to t3 described above is performed with the channels reversed. Also, at time t4, the slur mode signal
Since SLM does not become "1", block 103 in FIG. 8 becomes NO, the normal decay mode S4 is entered, and the envelope signal VL2 attenuates at the normal decay rate.

Referring to Figure 6c, the envelope signal corresponding to Figure 6a in the case where no slur effect is applied.
VL (VL1, VL2) and pitch information SKC will be explained. In this case, when the key-off pulse KFP2 of the second channel is generated at time t2, the forced dump pulse FDP2 is simultaneously generated by the processing of the routine 41 of FIG. 3, and as a result, block 109 of FIG. , becomes dump state S5. Also, since the slur mode pulse SLP is not generated, SLM is "0", and block 113 in FIG. 8 regarding the first channel becomes NO, resulting in a normal attack state S2. Therefore the envelope signal
VL2 rapidly decays at the damping rate, and VL1 rises at the normal attack rate. At time t3, the same operation as described above is performed with the channel reversed. At time points t2 and t3, the dump pulses FDP2 and FDP1 are generated because the legato type key press is changed, but at time point t4, the dump pulse FDP2 is not generated because the key is normally released. Therefore, envelope signal VL2 decays at the normal decay rate at time t4. Further, at time points t2 and t3, the pitch information SKC is suddenly switched from KC1 to KC2 or from KC2 to KC1. In this way, providing a dump period for the previous note when changing the legato key press brings about a desirable effect in the sense of adding excitement to the transition from the previous note to the new note, even when no slur effect is applied.

In addition, the change point of the envelope signal VL and pitch information SKC during the slur application period (this term does not necessarily refer only to the period in which the pitch is actually changing to the slur, but refers to the entire period in which this operation is performed) A, B in Figure 6b,
C, D, E, F) need not have a special correlation and may be set arbitrarily. However, in general, it is preferable that the point A where the envelope of the previous note starts to decay corresponds to the change time t2 of the key code NKC, and the point E where the slur change of the pitch information SKC starts corresponds to the point B where the envelope of the new note starts rising. It is preferable that it corresponds to The other points B, C, D, and F may be set completely arbitrarily. By variably controlling these points B, C, D, and F, a composite envelope (key-scaled first level L1
to the second level L2)
can be easily set to various characteristics.

By the way, according to the embodiment shown in FIG. 7, since the change rate of slur attack and slur decay is the same, the envelope signals VL1 and VL1 of each channel are
VL2 has completely opposite characteristics as shown in Figure 9a,
Arithmetically speaking, this composite level (VL1+VL2) should be flat as shown by the dotted line in FIG. 9a. However, to the auditory sense, it sounds like the level drops a little at the intersection, as shown by the dotted line in FIG. 9b. Since such a drop in volume during the slur application period is undesirable, the slur attack rate is made steeper than the slur decay rate, as shown by the dashed line in Figure 9c, so that the composite volume level is enhanced. It's good to do that. To do this, the slur speed data SSD given to the slur change width data generator 92 and/or slur calculation timing generator 93 in FIG. 7 is changed by a circuit as shown in FIG. data
SSD' instead of SSD generator 92 and/or
All you have to do is give it to 93.

In FIG. 10, a predetermined correction value is generated from the correction value generation circuit 115 (this correction value is based on the timbre selection information).
(may be varied depending on the TC), this correction value is added to the slur speed data SSD in an adder 116. The adder 11 is connected to the A input of the selector 117.
6 output is given, and the B input is given slur speed data SSD. The selector 117 selects the A input when the slur attack state signal ST1 is "1", and selects the A input when the slur attack state signal ST1 is "1".
When ST6 is "1", select B input. The output of the selector 117 is input as corrected slur speed data SSD' to the slur change width data generator 92 and/or the slur calculation timing generator 93 in FIG. The value of the change width data for the slur to be used and/or the period of the calculation timing signal are the data
Controlled according to SSD′. When doing a slur attack,
The data is the data SSD plus the correction value.
It is selected by the selector 117 as SSD'. At the time of slow decay, the data SSD itself is
It is selected by the selector 117 as SSD'. Therefore, the change rate of the envelope signal level during slur attack is faster than that during slur decay, and as shown in Figure 9c, the rise of the slur attack envelope is made steeper than the fall of the slur decay envelope. can do.

In Fig. 10, the slur attack rate is made faster than the slur decay rate by increasing the value of the slur speed data SSD at the time of slur attack, but the point is that the slur attack rate is faster than the slur decay rate. Any means can be used as long as it can be made faster than the kerate. For example, the change width data generated from the slur change width data generator 92 may be increased and corrected using a circuit as shown in FIG. 10 at the time of a slur attack. Also, how to correct
Rather than increasing the slurry attach rate, it is also possible to decrease the slurry decay rate.

FIG. 11 shows an example of the tone generator 10. In this example, a musical tone signal is generated by reading a musical waveform stored in the waveform memory 120. Specifically, as shown in FIG. 12, the waveform memory 120 stores musical waveform data for M timbre types from timbre 1 to M, and N musical waveform data from 1 to N for each timbre type. The tone waveform data for each key range is stored, and the tone waveform data for an attack part (rising part) and a sustain part are stored for each key range. As an example, the stored musical sound waveform data of the attack section consists of musical sound waveform data for a plurality of continuous periods in the rising part of a normal musical tone, and the amplitude envelope is standardized so that it is at a constant level. The timbre characteristics of the musical sound waveform correspond to the characteristic timbre characteristics of the rising part of a normal musical tone. Also, as an example,
The stored musical sound waveform data of the sustaining part consists of musical sound waveform data for a plurality of cycles in the sustaining part (sustain part) of the musical tone, and its timbre characteristics are different from those of the musical sound waveform of the rising part. Since the amplitude envelope of the sustaining part is originally at a substantially constant level, it is not necessary to perform special level normalization processing when storing waveform data, although it is of course possible.

Tone waveform data for each key range for each tone color type is stored at different absolute addresses in the waveform memory 120, respectively. For example, the tone waveform data of the attack part of key range 1 of tone 1 is stored at addresses from a predetermined attack start address AS1 to attack end address AE1, and the tone waveform data of the sustaining part is stored at a predetermined repeat start address. It is stored at addresses from address RS1 to repetition end address RE1. Similarly for other key ranges, from the respective predetermined attack start addresses AS2 and ASn to the attack end address AE.
2. The musical waveform data of the attack section is memorized by AEn, and the start addresses RS2 and RSn are repeatedly
The musical sound waveform data of the continuous part is stored from to the repetition end addresses RE2 and REn.

In FIG. 11, latch circuits 121, 122
A key code NKC is given to the microcomputer unit 13 (FIG. 2) via the data bus 72. The latch circuit 121 is latch-controlled by the key-on pulse KOP1 of the first channel.
The latch circuit 122 is latch-controlled by the key-on pulse KOP2 of the second channel. Therefore,
The key code of the key assigned to the first channel is latched in the latch circuit 121.
At 22, the key code of the key assigned to the second channel is latched. Key range detection circuit 123,1
24 inputs the output key codes of the latch circuits 121 and 122, respectively, and detects the key range to which the key corresponding to each key code belongs. Assuming that the way the key range is divided differs depending on the tone type, the tone selection information TC
is input to each circuit 123, 124, and the key range of the pressed key is detected according to the division according to the selected tone color type.

The key range detection circuits 123 and 124 output a key range code indicating the detected key range, and this is sent to the selector 12.
Give to 5. The selector 125 selects the key range code (output of the circuit 123) of the key assigned to the first channel at the time division timing of the first channel according to the channel timing pulse CH1, and selects the key range code (output of the circuit 123) at the time division timing of the second channel. Now for the second
Select the key area code (output of circuit 124) of the key assigned to the channel. In this way, the key area code KS of each channel time-division multiplexed according to the channel time-division timing is determined by the selector 125.
is output from. This key area code KS is applied to an attack start address generation circuit 126, a repetition start address generation circuit 127, an attack end detection circuit 128, and a repetition end detection circuit 129, respectively.

The attack start address generation circuit 126 is
Based on the input key range code KS and tone selection information TC, an attribute indicating an attack start address (for example, AS1 in the case of key range 1 of tone 1) according to the key range of the pressed key in the selected tone type. Generates start address data AS. The repeat start address generation circuit 127 generates a repeat start address (for example, for key range 1 of tone 1) according to the key range of the pressed key in the selected tone type, based on the input key range code KS and tone selection information TC. In this case, repeating start address data RS indicating RS1) is generated. Each data AS, RS is generated in a two-channel manner in a time-division manner according to the time-division multiplexed key range code KS of each channel.

Data AS is applied to the A input of the selector 130, data RS is applied to the B input, and either one is selected and applied to the adder 131.
The other input to the adder 131 is the phase accumulator 1.
32 provides phase angle data (in other words, relative address data for reading the waveform memory) that changes over time at a rate according to the pitch information SKC. This phase angle data and start address data AS or RS are added, and the address data indicating the absolute address of the musical waveform data for one sample point to be read from the waveform memory 120 is sent to the adder 1.
31.

In the waveform memory 120, musical waveform data for each sample point is read out in accordance with the address data given from the adder 131, and is given to a multiplier 133 for applying an envelope. The multiplier 133 controls the amplitude level of the musical waveform data using the envelope signal VL, and provides an amplitude envelope.

The attack end detection circuit 128 specifies an attack end address (for example, AE1 in the case of key range 1 of tone 1) based on the input key range code KS and tone color selection information TC, and uses the address data given from the adder 131. When the current value matches this attack end address, this is detected and a signal "1" is output as the attack end detection signal AED. The repeat end detection circuit 129 detects the repeat end address (for example, in the case of key range 1 of tone 1) based on the input key range code KS and tone selection information TC.
RE1), and when the current value of the address data given from the adder 131 matches this repetition end address, it is detected and a signal "1" is output as the repetition end detection signal RED. Both detection circuits 128 and 129 do not receive the output of the adder 131, but the output of the phase accumulator 1 as shown by the broken line.
It is also possible to input the output of 32 and detect the attack end or repetition end based on this.

The phase accumulator 132 is shift-controlled by channel timing pulse CH1.
A stage shift register 134, an adder 135 that adds the output of this shift register 134 and pitch information SKC, and a gate 136 that controls whether or not to give the output of the adder 135 to the input of the shift register 134. Contains pitch information
Accumulate SKC on two channels in a time-sharing manner.

The key-on signal KON1 for each channel output from the flip-flops 74 and 76 in FIG.
KON2 is the one-shot circuit 137 in Fig. 11,
138 respectively. One shot circuit 13
7 outputs one key-on pulse (at the time slot of the first channel) with the pulse width of the channel timing pulse CH1 in response to the rise of the signal KON1. One shot circuit 138 is a signal
In response to the rising edge of KON2, one key-on pulse is output with the pulse width of the inverted channel timing pulse 1 (at the time slot of the second channel). These key-on pulses, attack end detection signal AED and repeated end detection signal
RED is input to the OR circuit 139, and the signal obtained by inverting its output by the inverter 140 is sent to the gate 136.
is given to the control input of Therefore, when starting to generate a new musical tone in the first channel or the second channel, the output of the one-shot circuit 137 or 138 becomes "1", which is inverted by the inverter 140 via the OR circuit 139, and the gate 136 is closed. It will be done. As a result, the phase accumulator 1
The old calculation result of the channel in 32 is cleared, and the pitch information of the new musical tone to be generated is
SKC's accumulation rate begins. Similarly, when the attack end detection signal AED or the repeated end detection signal RED is generated, the old calculation result of the channel in the phase accumulator 132 is cleared, and a new accumulation rate of the pitch information SKC is started.

When attack start address data AS is selected by selector 130, adder 131
The address data given to the waveform memory 120 from the start address changes successively at a rate corresponding to the pitch information SKC with the attack start address as the initial value. Therefore, the musical waveform data of the attack section is sequentially read out from the waveform memory 120.

When the repeat start address data RS is selected by the selector 130, the address data given from the adder 131 to the waveform memory 120 changes successively at a rate corresponding to the pitch information SKC with the repeat start address as the initial value.
Therefore, the tone waveform data of the continuation part is read out from the waveform memory 120.

The selection switching of the selector 130 is performed using the selector 14.
2. The waveform switching control unit 141 includes flip-flops 143, 144, AND circuits 145, 146, and OR circuits 147, 148. This waveform switching control section 141 has the function of performing temporal waveform switching control such as reading out the musical sound waveform of the attack part and then reading the musical sound waveform of the sustaining part during normal sound generation, and also performs temporal waveform switching control such as reading out the musical sound waveform of the sustaining part during normal sound generation. By reading out the musical sound waveform of the sustained part from the beginning instead of reading out the musical sound waveform, it also functions as a "timbre control means" that makes the timbre of the rising part of the musical sound different between slurs and non-slurs. .

The set input S of the flip-flops 143 and 144 is connected to the decoder 71 (the fifth one) of the interface 14.
Key-on pulse KOP1 output from
KOP2 is applied, and the key-off pulse KFP is applied to the reset input R via OR circuits 147 and 148.
1, KFP2, slur mode pulse SLP, and output signals of AND circuits 145 and 146 are applied. The AND circuit 145 receives the channel timing pulse CH1, and the AND circuit 146 receives the inverted channel timing pulse 1. The attack end detection signal AED is applied to the other input of each AND circuit 145, 146. The output of flip-flop 143 is given to the A input of selector 142, and the output of flip-flop 143 is applied to the A input of selector 142.
The output of 44 is given to the B input of selector 142. The selector 142 selects the A input when the channel timing pulse CH1 is "1", and selects the A input when the channel timing pulse CH1 is "0".
Select B input when . The output signal of the selector 142 is applied to the selection control input of the selector 130. When the signal is "1", the A input, that is, the attack start address data AS is selected, and when it is "0", the B input, that is, the repeated start address data is selected.
Select RS.

When in normal mode (second mode) that does not apply slurs, key-on pulse KOP1 or KOP
Flip-flop 143 or 144 is set in response to the occurrence of 2. This flip-flop 1
43 or 144 output signal “1” is selector 1
42 in a time-sharing manner, and the selector 130 is set to A.
Set to input selection state. Therefore, attack start address data AS is selected initially, and adder 1
31. Based on this, as mentioned above,
The musical waveform data of the attack section is stored in the waveform memory 120.
is read from. When the reading of the attack part is completed, the attack end detection signal AED becomes “1”.
becomes. If this signal AED becomes "1" at the timing of the first channel, the AND circuit 145 outputs a signal "1" and resets the flip-flop 143. Further, if the signal AED becomes "1" at the timing of the second channel, the AND circuit 146 outputs the signal "1" and resets the flip-flop 144. flipflop 143
144 is reset, its output signal "0" is applied to the selector 130 via the selector 142 at the corresponding channel timing, and the selector 130 enters the B input selection state. In this way, after reading the attack part,
The start address data RS is repeatedly selected, and after the phase accumulator 132 is once cleared as described above, the accumulation rate of the pitch information SKC is resumed. As a result, the tone waveform data of the continuation section is read out from the waveform memory 120. When the reading of the continuation part is completed, the repetition end detection signal RED becomes "1", and as described above, the phase accumulator 132 is cleared once, and then the accumulation rate of the pitch information SKC is restarted. On the other hand, repeated start address data
RS continues to be selected by selector 130. Thus, when reading the continuation part, the address data output from the adder 131 changes sequentially from the repeat start address to the repeat end address, and when it reaches the repeat end address, returns to the repeat start address and repeats this change. Therefore, during normal sound production, the tone waveform of the attack part is read out once, and then the tone waveform of the sustaining part is read out repeatedly. In addition, if key-off pulses KFP1 and KFP2 are generated before the end of the attack section,
The flip-flops 143 and 144 are reset without the occurrence of the attack end detection signal AED, and the process moves on to reading the continuation part.

In the slur mode (first mode), a slur mode pulse SLP is generated simultaneously with the key-on pulse KOP1 or KOP2 (see FIG. 6b).
The flip-flops 143 and 144 are reset priority type, and are set to the reset state by applying "1" to the reset input R via the OR circuits 147 and 148 in response to the slur mode pulse SLP. As a result, the selector 130 repeatedly selects the start address data RS from the beginning. Therefore, in the slur mode, the normal tone waveform of the attack section is not read out, but the tone waveform of the sustain section is read out from the beginning. In this way, the musical sound waveform of the sustained part is read out at the beginning of the pronunciation, but during that time, the pitch information SKC gradually shifts from the value corresponding to the pitch of the prenote to the value corresponding to the pitch of the new note, and the pitch changes. Granted. In this way, the timbre of the musical waveform that rises in a slur-like manner with pitch changes is not the timbre of the normal rising part, but the timbre of the sustaining part.

In the embodiment shown in FIG. 11, the timbre of the sustaining part is used as the timbre of the rising part of the musical tone during slurs, but the present invention is not limited to this, and a timbre exclusively for slurs may be used. FIG. 13 shows one example, and the same circuits as those shown in FIG. 11 are given the same reference numerals. In the waveform memory 120', for each key range of each tone type,
It stores not only the tone waveforms of the attack part and the sustaining part, but also the tone waveform of the rising part exclusively for slurs consisting of multiple cycles.

Along with storing the rising part musical waveform exclusively for slurs, the slur start address generation circuit 14
9 and a slur end detection circuit 150 are provided. The circuit 149 generates data SLS indicating the first address of the rising edge tone waveform exclusively for slurs, that is, the slur start address, in accordance with the key range code KS and the tone color selection information TC. The circuit 150 is
In accordance with the key range code KS and the tone selection information TC, the last address of the slur-specific rising part musical sound waveform, that is, the slur end address, is specified and stored in the memory 120'.
When the input address data reaches the slur end address, a slur end detection signal SED is output. This slur end detection signal SED is input to the OR circuit 139 together with other signals AED and RED, and is also input to the OR circuit 151 as a signal.
It is input to AND circuits 145 and 146 similarly to the AED.

The tone control means 152 includes a selector 153, a flip-flop 154, and AND circuits 155, 15.
6,157, and an OR circuit 158. The flip-flop 154 is a slur mode pulse SLP.
The slur end detection signal is set by
Reset by SED. The output of flip-flop 154 is applied to AND circuit 155, and the output of AND circuit 155 is applied to the control input of selector 153. When the output signal of the AND circuit 155 is "0", the selector 153 enters the B input selection state, selects the attack start address data AS output from the circuit 126, and applies it to the A input of the selector 130. However, AND circuit 1
When the output signal of 55 is "1", the A input selection state is entered, the slur start address data SLS output from the circuit 149 is selected, and the selector 130
to the A input of Channel flag CHFLG inverted signal and channel timing pulse CH
1 is input to the AND circuit 156, and a signal corresponding to the channel flag CHFLG and a signal obtained by inverting the channel timing pulse CH1 are input to the AND circuit 1.
57 is given. The channel flag CHFLG is "0" when indicating the first channel, and "1" when indicating the second channel. AND circuit 156,
The output of 157 is given to the other input of AND circuit 155 via OR circuit 158. Flip-flop 1 when slur mode pulse SLP is generated
54 is set, but this slur mode pulse
If SLP is generated corresponding to the first channel, the channel flag CHFLG is "0", the AND circuit 156 is enabled, and the AND circuit 155 performs time division in the time slot of the first channel in response to the pulse CH1. is made possible. On the other hand, if pulse SLP corresponds to the second channel
CHFLG is "1", the AND circuit 157 is enabled, and the second
AND circuit 155 in channel time slot
is made possible in a time-sharing manner. In this way, the AND circuit 155 is enabled in a time-division manner corresponding to the channel of the musical tone that should rise in the slur mode, and provides the output of the flip-flop 154 to the selector 153.

At the rise of a normal musical tone, the output of the AND circuit 155 is "0", and the selector 153 selects the attack start address data AS. Therefore, as described above, the musical sound waveform of the attack section is read out at the rise of the musical tone.

On the other hand, in the slur mode, the flip-flop 154 is set by the pulse SLP, and the output of the AND circuit 155 becomes "1" at the timing of the channel to which the musical tone to rise in the slur mode is assigned. As a result, the selector 153 selects the slur start address data SLS at the channel timing, and the waveform memory 12
Reading out the rising part tone waveform exclusively for slurs starts from 0'. When one reading of this slur-specific rising edge tone waveform is completed, a slur end detection signal SED is generated and the flip-flop 154 is reset. Also, OR circuit 15
1. Flip-flops 143 and 144 are reset via AND circuit 145 or 146. Therefore, the selector 130 switches to the B input selection state and repeatedly selects the start address data RS. In this manner, in the slur mode, the rising part tone waveform exclusively for slurs is read out first, and thereafter, the sustaining part tone waveforms are repeatedly read out, as described above.

In the above embodiment, the waveform memories 120, 120'
Although the level of the tone waveform stored in the memory is standardized to a certain level, the present invention is not limited to this, and tone waveforms to which a predetermined rise envelope or the like has been added may be stored. In addition, as a way of storing the musical waveform data of the sustained part,
It is also possible to use a method of smoothing the connection between the end part and the start part of the continuation part waveform that is repeatedly read out, as shown in the above issue. Furthermore, the tone waveform of the sustaining part stored in the memory is not limited to a plurality of cycles, but may be one cycle, 1/2 cycle, or the like. Further, in the above embodiment, different tone waveforms are stored for each key range, but different tone waveforms may be stored for each individual key. Furthermore, if timbre key scaling is not performed, it is sufficient to simply store a tone waveform common to all keys.

In addition, instead of storing all the waveform data at each sample point of the waveform to be stored in the waveform memory, only the waveform data at discrete sample points are stored, and the waveform data at intermediate sample points is processed by interpolation. It may be calculated. Further, the multi-period waveform stored in the waveform memory may consist not only of continuous plural periods but also of discontinuous plural periods. For example, the period from the rise to the fall of a musical tone may be divided into multiple frames, and only one or two representative cycles of waveform data may be stored for each frame, and this waveform data may be sequentially switched and read out repeatedly. Furthermore, if necessary, at the time of this waveform switching, the previous waveform and the next new waveform may be interpolated to form waveform data that changes smoothly.

In addition, the encoding method for waveform data stored in the waveform memory is not limited to the pulse code modulation method (PCM method), but also the differential PCM method, delta modulation method (DM method), adaptive PCM method (ADPCM method), and adaptive delta modulation method (DM method). Other appropriate methods such as a modulation method (ADM method) may also be used.

Furthermore, the two musical tone generation channels may be formed in parallel instead of time-sharing. moreover,
In the above embodiment, the previous sound and new sound are generated in separate channels during slurs, and the falling and rising envelopes of both sounds are cross-faded. The present invention can also be applied to the case of providing.

Furthermore, the tone generation method in the tone generator 10 is not limited to the simple waveform memory reading method as described above, but may also be an appropriate method such as a harmonic synthesis method, a frequency modulation calculation method, an amplitude modulation calculation method, or a digital filter method. May be used. In this case, the timbre switching control at the rising edge of a musical tone is performed by controlling calculation parameters such as harmonic coefficients or modulation indexes, or filter coefficients.

Furthermore, this invention is not limited to single-note electronic musical instruments;
Of course, the present invention can also be applied to multitone electronic musical instruments.

〔Effect of the invention〕

As described above, according to the present invention, the timbre of the rising part of a musical tone is made different depending on when a pitch change is applied and when it is not, so that the unique timbre change at the time of a slur as seen in natural musical instruments can be realized. can be realized.

[Brief explanation of the drawing]

FIG. 1 is a waveform diagram schematically showing an embodiment of the present invention, FIG. 2 is a block diagram of the overall configuration of an electronic musical instrument according to an embodiment of the present invention, and FIG. 3 is a diagram showing the microcomputer section of FIG. FIG. 4 is a flowchart showing an example of a key-off event program executed by the microcomputer, and FIG. 5 is a flowchart showing an example of a key-off event program executed by the microcomputer section. 6 is a timing chart showing an example of generation of output signals, envelope signals, and pitch information of each part in FIG. 5. FIG. 7 is a block diagram showing an example of the envelope generator in FIG. Figure 8 is a flowchart showing an example of the processing executed by the state control circuit in Figure 7, and Figure 9 shows the synthesis level of the envelopes of the falling and rising parts of the front and new sounds during slurs. 10 is a block diagram showing an example of a circuit added to make the slur attack rate steeper than the slur decay rate in the example of FIG. 7, and FIG. 11 is a block diagram of the tone generator in FIG. 2. Block diagram showing an example of the 12th
11 is a diagram showing an example of a tone waveform stored in the waveform memory of FIG. 11, and FIG. 13 is a block diagram showing an example of a modification of the tone generator shown in FIG. 11. 10...Tone generator, 11...Keyboard,
12...Key press detection circuit, 13...Microcomputer unit, 14...Interface, 15...Envelope generator, 16...Block information generation and slur adding circuit, 17...Slur control operator,
18... Tone color selection circuit, 120, 120'... Waveform memory, 141... Waveform switching control section (tone control means), 152... Tone color control means.

Claims (1)

[Scope of Claims] 1. A first mode in which the pitch of the musical tone to be generated is gradually changed from a predetermined first pitch to a second pitch at the rise of the musical tone, or a mode in which the pitch of the musical tone to be generated is gradually changed from a predetermined first pitch to a second pitch, or the second pitch is changed from the beginning. pitch control means for controlling the pitch of the musical tone in accordance with one of the modes; a mode selection means for selecting the first mode or the second mode; a musical sound generating means capable of selectively generating a musical sound waveform and a musical sound waveform for a sustained part, and generating a musical sound signal corresponding to the selected musical sound waveform at a pitch controlled by the pitch control means; , the musical sound waveform of the rising portion of the musical tone signal generated by the musical tone generating means is controlled to be different depending on the first or second mode, and in the first mode, the musical sound waveform of the rising portion of the musical tone signal generated by the musical tone generating means is controlled to be different depending on the first or second mode. and timbre control means for generating a musical sound waveform for the sustaining part in the second mode, and generating a musical sound waveform for the rising part at a rising part of the musical tone signal when in the second mode. 2. The mode selection means selects the first mode or the second mode in accordance with a key operation method on a keyboard for specifying the pitch of a musical tone to be generated. Electronic musical instruments listed in section. 3. The electronic musical instrument according to claim 1, wherein the predetermined second pitch corresponds to the pitch of the musical tone to be generated. 4. The electronic musical instrument according to claim 1, wherein the predetermined first pitch corresponds to the pitch of a musical tone that has been generated immediately before. 5. A first mode in which the pitch of the musical tone to be generated is gradually changed from a predetermined first pitch to a predetermined second pitch at the rise of the musical tone, or a second mode in which the pitch is set at the second pitch from the beginning. , a pitch control means for controlling the pitch of the musical tone according to one of the modes, a mode selection means for selecting the first mode or the second mode, and a musical sound waveform for a rising part and a pitch for a slur. musical sound generating means capable of selectively generating a musical sound waveform and generating a musical sound signal corresponding to the selected musical sound waveform at a pitch controlled by the pitch control means; The musical sound waveform for the slur is controlled to be different at the rising edge of the musical tone signal depending on the first or second mode, and in the first mode, the musical waveform for the slur is changed at the rising edge of the musical tone signal. and timbre control means for generating a musical sound waveform for the rising edge of the musical tone signal at the rising edge of the musical tone signal when in the second mode.