JPH0412476B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to key touch control in an electronic musical instrument and musical sound waveform synthesis of an attack portion in response to key touches.

In musical sound synthesis or generation, it is known that the musical sound waveform in the rising part (attack part) of a sound contributes to the timbre with a considerable degree of weight. It has been difficult to synthesize this with conventional electronic musical instruments. This is because the attack part is a transient part that is quite unstable, so its musical sound waveform becomes complex. This is also due to the fact that the complexity of the waveform at the attack portion of a piano or the like changes depending on the strength with which the string is struck (in the case of a wind instrument, the strength at the beginning of the string). On the other hand, after passing this attack portion, the musical sound waveform exhibits stable characteristics. The reality of conventional electronic musical instruments is that musical sound waveforms in this stable region are synthesized or generated even in the attack portion. In order to eliminate this drawback, JP-A-52
No. 121313 discloses that the entire waveform of the attack part and a part of the waveform of the sustaining part are each stored in a memory, and after reading the entire waveform of the attack part once, the stored waveform of the sustaining part is read out repeatedly. There is.
It is also shown that a plurality of all waveforms of the attack section are stored in response to a plurality of different key touches, and one is selected and read out in response to the key touch.

However, there was a problem in that the key touch control of the attack part was possible, but the key touch control of the continuation part was not possible. Another problem is that a memory is required to store all waveforms of the attack section in a number corresponding to the number of controllable key touch stages.

The present invention has been made in view of the above-mentioned points, and is an electronic musical instrument that generates or synthesizes a complex musical sound waveform that exhibits different characteristics depending on the touch of a key in the attack part, and also enables key touch control over the entire sound generation period of a musical tone. This is what we are trying to provide.

The electronic musical instrument according to the present invention includes: a key touch detection means for detecting a touch of a key operated on a keyboard; and a plurality of first memories each storing musical waveforms of a rising part of a sound over a plurality of cycles with different characteristics. , a second memory that stores the musical waveform of the sustained part of the sound; and one of the first memories is selected according to the touch of a key when a key is operated on the keyboard, and the selected first memory is selected. reads out the musical waveform of the rising portion from the memory at a rate corresponding to the operated key;
Subsequently, readout control means reads out a musical sound waveform from the second memory at a rate corresponding to the operated key, and control means controls the musical sound waveform read from the second memory in accordance with the touch of the key. It is equipped with the means.

One of the first memories is selected in response to the touch of a key, and a tone waveform of a rising portion having characteristics corresponding to the touch of the key is read out from the memory. Subsequently, the tone waveform of the continuous portion is read out from the second memory, and this tone waveform is controlled in accordance with the touch of the key. Therefore, not only can high-quality key touch control be performed on the tone waveform of the rising portion, but also key touch control can be performed on the tone waveform of the sustaining portion.

According to another aspect of the invention, at least two first memories each storing musical waveforms of a rising part of a sound over a plurality of periods with different characteristics are provided, coefficient supply means for supplying coefficients correspondingly to each of the first memories in accordance with the touch; and coefficients supplied from the coefficient supply means correspondingly to the amplitudes of the plural tone waveforms read from the first memory. A control means is provided for controlling the respective conditions according to the conditions.

Coefficients corresponding to key touches are supplied to each of the first memories, and the amplitudes of the plural tone waveforms read from the first memories are respectively controlled by these coefficients. In this manner, the amplitudes of the multiple tone waveforms read from the first memory are controlled at the rising portion of the sound, so that these controlled multiple tone waveforms are mixed and produced at the rising portion of the sound. Following this, the tone waveform read from the second memory is sounded. In this case, since the number of first memories is smaller than the number of stages of key touch control that can be realized, the memory configuration can be simplified.

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

In FIG. 1, the key switch circuit 11 is composed of key switches corresponding to each key of the keyboard 10, and is composed of, for example, a two-stage key switch in order to detect the touch of each key. That is, the first key switch K1 for each key
and a second key switch K2, respectively, so that when a certain key is pressed, the first key switch K1 corresponding to that key closes first, and then the second key switch K2 closes. The key assigner 12 scans the key switch circuit 11 to detect a pressed key, and assigns the pressed key to one of a predetermined number (for example, 8) of musical tone generation channels. For more information,
When it is detected that the first key switch K1 of a certain key has been switched from off to on, the key is assigned to an arbitrary channel, and then the second key switch K2 of that key is switched from off to on. When the key is broken, the key code KC of the key is output corresponding to the assigned channel. A time division time slot is predetermined for each channel, and a key code assigned to each channel.
KC is outputted in a time-division manner using the respective time slots. The time slot of each channel is set with a width corresponding to one period of the clock pulse φ.

From the key assigner 12, in addition to the key code KC, the start pulse S, the first key-on signal TK
1. 2nd key-on signal TK2, key-off signal TD
0 is output. The start pulse S is generated only once when the second key switch K2 is switched from OFF to ON in synchronization with the time slot of the channel to which the corresponding key is assigned. 1st key-on signal TK1
When a new assignment process is performed in response to turning on the first key switch K1, it immediately becomes "1" in synchronization with the time slot of the assigned channel.
This is a signal that thereafter becomes "1" repeatedly in synchronization with the time slot. 2nd key-on signal TK
2 rises to "1" corresponding to the time slot of the channel to which the key is assigned when the second key switch K2 is switched from OFF to ON, and thereafter repeatedly rises to "1" in synchronization with the time slot. This is a signal that rises to "1" and then repeatedly becomes "1" in synchronization with the time slot. key off signal
TD0 is a signal that becomes "1" when the first key switch K1 is switched from on to off, corresponding to the time slot of the channel to which the key is assigned. These key-on signals TK1, TK
2 and the key-off signal TD0 fall to "0" when the decay end pulse DF of the corresponding channel is applied to the key assigner 12 as described later.
An example of the generation of the above-mentioned signals TK1, TK2, TD0, S, and DF for one channel is shown in FIG. 2. In the same figure, signals TK1, TK2, TD0
is shown as a continuous signal, but as mentioned above, it occurs in a time-division manner.

Key code output from key assigner 12
KC is applied to a frequency number generator 13, which generates a frequency number F corresponding to the tone frequency of the key indicated by the key code KC. The frequency number F of each channel output from the generator 13 in a time-division manner is repeatedly added for each channel in a phase accumulator 14, and the instantaneous phase angle data of the musical tone signal to be generated in each channel is added from the accumulator 14 in a time-division manner. is output as follows. That is, the frequency number F indicates a phase increment value per sampling time, and instantaneous phase angle data is obtained by repeatedly adding this in the accumulator 14 for each sampling time. The accumulator 14 is capable of time-division accumulation for each time slot corresponding to each channel, and a start pulse S is applied from the key assigner 12 to its clear input. When a new press key is assigned to a certain channel,
That is, when the sound generation of musical tones is to be started on that channel, a start pulse S is applied, and the contents of the accumulator 14 regarding that channel are cleared. In this way, the accumulation of frequency numbers F for the newly assigned key starts from zero.

The output of the accumulator 14 is the musical sound waveform memory 1
It is used as an address signal for reading 5 and 16. The attack section memory 15 includes a plurality of memories 15-1 to 15-8 (eight in this example) in which musical sound waveforms of an attack section consisting of a plurality of periods are stored in advance.
The characteristics (waveform shape, amplitude level, period, etc.) of the musical sound waveforms stored in -8 are different from each other. The sustain part memory 16 stores musical sound waveforms of sustain parts (stable parts) of musical tones in advance over a plurality of cycles, and includes a plurality of memories 16-1 to 16-N storing sustain part waveforms with mutually different characteristics. Equipped with:

Individual memory 15-1 of attack section memory 15
15-8 correspond to key touches of different intensities, and each has a complex attack musical sound waveform (a musical sound waveform over multiple cycles with amplitude characteristics at the rise of the sound) having characteristics corresponding to each key touch. I remember. For example, the memory corresponding to a stronger key touch stores a waveform with more harmonic components and larger amplitude, and the memory corresponding to a weaker key touch stores a waveform with fewer harmonic components and smaller amplitude. In order to select one attack unit memory (one of 15-1 to 15-8) in response to the touch of the pressed key, the key touch detection signal obtained by the touch counter 17 is used. The touch counter 17 has
A first key-on signal TK1 and a second key-on signal TK2 of each channel are output from the key assigner 12 in a time-division manner, and the time difference between the rise times of the signals TK1 and TK2 is counted for each channel. As a result, the touch (keystroke speed) of the keys assigned to each channel is individually detected. The state of the output of the touch counter 17 regarding a certain channel is determined by the first key-on signal TK1.
becomes “0” when it rises to “1”, and from then on the second
The count value increases gradually until the key-on signal TK2 rises to "1", and when the signal TK2 rises to "1", counting is stopped and the count value up to that point is held thereafter. This held count value corresponds to the strength of the key touch. Such count value data, that is, the key touch detection signal is sent to the touch counter 17 in a time-division manner for each channel.
is output from.

The upper three bits of the count value data output from the touch counter 17 are input to the decoder 18, and outputs corresponding to eight different key touch strengths are obtained from the decoder 18. The eight output lines of the decoder 18 are connected to each attack section memory 15-1 to 1.
5-8 are individually connected to enable input E.
With this configuration, 1 corresponding to the key touch of the pressed key
Two attack section memories (one of 15-1 to 15-8) are selectively set to a readable state. The enabling operation of the memories 15-1 to 15-8 is performed in a time-division manner for each time slot of each channel in accordance with the output of the decoder 18.

The output of the phase accumulator 14 is commonly applied via a distributor 19 to the address inputs of the attack section memories 15-1 to 15-8. Distributor 1
Reference numeral 9 indicates that the phase accumulator 1 is activated only for the period required to read out all the addresses of the attack section memories 15-1 to 15-8 once when the sound should be started.
4 to the attack section memories 15-1 to 15-
8, otherwise the corresponding accumulator 1
4 to the persistence section memory 16. Assuming that the number of bits of the address signal for each memory 15-1 to 15-8 is 8 bits,
The accumulator 14 outputs a 9-bit signal, which is at least one more bit than that, and uses the 9th bit output signal as a signal for identifying the period required to read all addresses of each memory 15-1 to 15-8.

For controlling the distributor 19, the clock pulse φ
Accordingly, an 8-stage/1-bit shift register 20, an AND circuit 21, and an OR circuit 22 are provided, which are shift-controlled in synchronization with the time-division time slots of each channel. The output of the shift register 20 is applied to one input of the AND circuit 21, and the start pulse S outputted from the key assigner 12 inverted by an inverter 23 is applied to the other input. The most significant bit MSB, that is, the 9th bit, of the output of the AND circuit 21 and the output of the accumulator 14 is input to the OR circuit 22, and the output thereof is input to the shift register 20. The shift register 20 is for time-divisionally supplying the control signal of the distributor 19 corresponding to each channel, and the control signal of each channel is outputted in a time-division manner and applied to the control input of the distributor 19. Distributor 1
9 distributes the output signal of the accumulator 14 to the attack section memory 15 when the control signal given from the shift register 20 is "0", and distributes the output signal to the sustain section memory 16 when it is "1".

When a newly pressed key is assigned to a certain channel, the start pulse S corresponding to that channel becomes "1" only in the first time slot, as described above. In response to this start pulse S being "1", the AND circuit 21 is disabled and the stored contents of the shift register 20 corresponding to the channel are cleared to "0". On the other hand, as mentioned above, the start pulse S causes the accumulator 1 to
The contents of the corresponding channel of No. 4 are once cleared, and the corresponding channel of the accumulator 14 starts accumulating the frequency number F starting from all bits "0". Therefore, from the time when the start pulse S is generated, the accumulator 1 for the channel concerned is
The value of the 9th bit of the output signal remains "0" until the contents of the lower 8 bits of the output signal of No. 4 go through one cycle, and becomes "1" when a carry signal from the 8th bit is given. Switch to . When the 9th bit signal switches to "1", "1" is loaded into the shift register 20 via the OR circuit 22, and from then on, this "1" is loaded into the shift register 20, the AND circuit 21, and the OR circuit 22. The signal is circulated through a loop and held in the shift register 20. Therefore, the storage contents of the shift register 20 corresponding to the channel in which the start pulse S is generated are stored only from the time when the pulse S is generated until the lower 8 bits of the output signal of the accumulator 14 regarding the channel go through one cycle. Retains 0” and thereafter becomes 1
hold. Therefore, when the output state of the shift register 20 is shown for only one channel, it is as shown in FIG.

By controlling the distributor 19 as described above in accordance with the output signal "0" or "1" of the shift register 20, the lower 8 bits of the output signal of the accumulator 14 complete one cycle from the time when the start pulse S is generated. Until then, the output signal of the accumulator 14 is applied to the attack section memory 15 as an address signal, and the musical tone waveform is read out from the attack section memory 15. Since the lower 8 bits of the output signal of the accumulator 14 are used as an address signal, reading of all addresses in the attack section memory 15 is completed during this time (see the column of the attack section in FIG. 2). Specifically, from the only attack section memory (one of 15-1 to 15-8) enabled by the output signal of the decoder 18,
The amplitudes of all sample points of the multi-period attack music waveform are sequentially read out. The read musical tone signal is converted into an analog signal by a digital/analog converter 24 and provided to a sound system 25. At the same time as the reading of the attack section memory 15 is completed,
The distribution of the output signal of the accumulator 14 is switched, and from now on, the address signal is given to the persistence section memory 16. In the sustain section memory 16, the tone waveform signal stored therein is repeatedly read out in accordance with the address signal given from the accumulator 14. The read musical waveform signal is subjected to processing such as adding an amplitude envelope, and then provided to the digital/analog converter 24.

At least one of the memories in the sustain part memory 16, for example memory 16-N, stores a musical waveform with simple sustain part characteristics (constant amplitude characteristics) without any modulation effect. The other memories 16-1, 16-2, . . . prestore musical sound waveforms to which different modulations have been applied. Examples of modulation include frequency modulation (vibrato modulation), amplitude modulation (tremolo modulation), and timbre modulation (modulation of tone waveform shape), and tone waveforms to which one or more of these are applied are stored in advance. As an example, a musical sound waveform mainly composed of vibrato modulation and further subjected to appropriate amplitude modulation or timbre modulation is stored. Further, as an example, each memory 16-
1, 16-2, . . . store in advance musical sound waveforms modulated with different depths. Similar to the memories 15-1 to 15-8, each memory 16-
1 to 16-N, musical sound waveforms having a plurality of cycles are stored, and the plurality of cycle waveforms are repeatedly read out in accordance with an 8-bit address signal given from the accumulator 14 via the distributor 19.

The selector 26 selects each memory 16-1 to 16-N.
For example, it is linked to the vibrato depth selection switch placed on the panel of the electronic musical instrument, and the memory (16-1 to 16-N) corresponding to the selected depth is selected. Select the output signal. The selected signal is applied to multiplier 27. The multiplier 27 applies an amplitude envelope to the musical waveform signal read out from the sustain section memory 16. Second key-on signal TK from key assigner 12 to envelope generator 28
2 and a key-off signal TD0, and based on these signals, the envelope generator 28 generates an envelope signal having sustain characteristics and decay characteristics as shown in FIG. 2 in a time-division manner for each channel. . This envelope signal is applied to a multiplier 27, which controls the amplitude of the musical waveform signal in accordance with the envelope signal (see the column of the sustain section in FIG. 2). When the decay envelope ends, the envelope generator 28 outputs a decay end pulse DF, and the key assigner 12
given to.

The output of multiplier 27 is given to multiplier 29.
The multiplier 29 is for controlling the amplitude of the musical waveform signal in accordance with the touch of a key. As mentioned above, since the musical waveform signal of the attack part outputted from the attack part memory 15 has an amplitude level corresponding to the touch of the key, the amplitude of the musical waveform signal of the sustaining part is controlled according to the touch of the key at the time of keystroke. Otherwise, there is a risk that the connection between the musical waveform signals of the attack part and the sustain part will be poor. Multiplier 29 serves to eliminate such disadvantages. The key touch detection signal output from the touch counter 17 is given to the coefficient ROM 30, and coefficient data corresponding to the keystroke strength is stored in the ROM.
30. Coefficient data read from ROM 30 is given to multiplier 29 . As a result, the amplitude of the tone waveform signal of the sustain part is controlled in accordance with the keystroke strength. The output signal of the multiplier 29 is applied to the digital/analog converter 24 and is converted into an analog signal before being sent to the sound system 25.
Note that the musical waveform signal read from the memory 16 and the envelope signal and coefficients generated from the generator 28
If the coefficient data read from the ROM 30 is expressed logarithmically, the multipliers 27 and 29 can be constructed by simple adders. In that case, a counting/linear converter is provided to convert the continuous Buraku sound waveform signal for which calculation has been completed into a linear expression, and then to provide the signal to the digital/analog converter 24.

Another embodiment of the attack section memory 15 shown in FIG. 1 is shown in FIG. The memories 31 and 32 include
Tone waveforms of attack sections having mutually different characteristics are stored in advance over a plurality of cycles. For example, the memory 31 stores a tone waveform corresponding to the strongest key touch, and the memory 32 stores a tone waveform corresponding to the weakest key touch. Memory 31,3
Similarly to the memories 15-1 to 15-8 in FIG. 1, the 8-bit address signal is applied to the memory 2 from the distributor 19 (FIG. 1). Tone waveform signals are simultaneously read out from both memories 31 and 32 in accordance with this address signal. The outputs of the memories 31 and 32 are given to multipliers 33 and 34, respectively, and multiplied by coefficients P1 and P2 given from the coefficient ROM 35. In this way, the two musical waveform signals whose amplitude levels are controlled according to the coefficients P1 and P2 are transmitted to the multipliers 33 and 34.
are outputted from each of them, and the adder 36 adds them together. The output signal of the adder 36 is sent to the digital/analog converter 24 as a musical waveform signal of the attack section.
(Figure 1).

The coefficient ROM 35 stores in advance coefficients P1 and P2 of various values corresponding to the key touch and the elapse of time of the attack section. The output of the decoder 18 (FIG. 1) and the upper bits of the address signal applied from the distributor 19 to the memories 31 and 32 are applied to the address input of the ROM 35. A set of coefficients corresponding to the key touch is selected by the output of the decoder 18, and coefficients P1 and P2 are read out from the selected set according to the upper bits of the address signal of the musical waveform memory (that is, according to the passage of time). . Furthermore, the decoder 18
Only the output signal may be used as the address signal of the ROM 35, and the coefficients P1 and P2 may not change over time. The number of memories 31 and 32 is not limited to two, but may be more than two. Also, one of the attack section memories 31 and 32 is used as the sustain section memory 1.
6 may be shared.

In FIG. 1, there may be only one sustaining section memory 16, and only one musical sound waveform may be stored there.
It may be only for the period. Alternatively, the output of the attack section memory 15 and the output of the sustain section memory 16 may be opened and closed in accordance with the output signal of the shift register 20 without providing the distributor 19.

A detailed example of an electronic musical instrument that detects key touches using the two-stage key switches K1 and K2 as shown in the above embodiment is shown in Japanese Patent Laid-Open No. 139520/1983. However, the method of detecting a key touch is not limited to this, and any other method may be used. For example, it is also possible to use the one shown in Japanese Patent Publication No. 56-793.

As explained above, according to the present invention, it is possible to make the waveform shape, amplitude, pitch, etc. for each cycle different in the multi-cycle musical sound waveform of the attack part, so that a complex musical tone signal can be generated or synthesized in the attack part. Furthermore, the musical waveform characteristics (waveform shape, amplitude, pitch, etc.) can be further varied depending on the touch of the key. Therefore, it is possible to generate high-quality musical tones that exhibit complex characteristics in the attack portion, such as the sounds of natural musical instruments such as a piano. In addition, not only can high-quality key touch control be achieved in the attack part, but also the key touch control can be applied to the musical sound waveform in the sustained part without expanding the memory configuration, which is an excellent effect. . Furthermore, it is possible to realize high-quality key touch control in the attack portion without expanding the memory configuration, which is an excellent effect.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an embodiment of an electronic musical instrument according to the present invention, Fig. 2 is a timing chart explaining the operation of the embodiment, and Fig. 3 shows an example of modification of the attack section memory in the embodiment. This is a block diagram. 10...Keyboard, 11...Key switch circuit, 1
2...Key assigner, 15...Attack section memory, 16...Continuation section memory, 17...Touch counter, 19...Distributor, 31, 32...Attack section memory, 33, 34...Multiplier, 35... …coefficient
ROM, 36... Adder.

Claims (1)

[Scope of Claims] 1. A keyboard, a key touch detection means for detecting the touch of a key operated on the keyboard, and a plurality of first keys each storing a musical sound waveform of a rising portion of a sound over a plurality of cycles with different characteristics. a second memory that stores musical waveforms of sustained parts of notes, and selects one of the first memories according to the touch of a key when a key is operated on the keyboard, and selects one of the first memories. readout control that reads out the musical sound waveform of the rising portion from the first memory at a rate corresponding to the operated key, and subsequently reads the musical sound waveform from the second memory at a rate corresponding to the operated key; An electronic musical instrument comprising: means for controlling a musical sound waveform read from the second memory in accordance with the touch of the key. 2. A keyboard, a key touch detection means for detecting the touch of a key operated on the keyboard, and at least two first keys each storing a musical sound waveform of a rising part of a sound over a plurality of cycles with different characteristics.
a second memory storing musical sound waveforms of sustained parts of sounds; and coefficients for supplying coefficients to each of the first memories in response to key touches detected by the key touch detection means. supply means; when a key is operated on the keyboard, each of the musical sound waveforms is read out from the first memory at a rate corresponding to the key, and subsequently from the second memory at a rate corresponding to the operated key; readout control means for reading out musical sound waveforms at a rate; and control means for controlling the amplitudes of the plurality of musical sound waveforms read from the first memory in accordance with coefficients respectively given from the coefficient supply means. . An electronic musical instrument, wherein a plurality of tone waveforms controlled by the control means are mixedly sounded at a rising portion of a sound, and subsequently a tone waveform read from the second memory is sounded.