JPH04174498A - Musical sound signal generating device - Google Patents

Abstract

PURPOSE:To reproduce a change of a waveform in increasing or decreasing a sound volume gradually by controlling a sound in a musical sound signal based on a degree of the change in addition to control of a sound based on a control waveform signal. CONSTITUTION:A key on signal KON, a key touch signal KT, and a pedal signal EP are outputted from an operator control part 30 in addition to a key code signal KC, and each signal is inputted to an envelope generator 110. A level of a volume is controlled by an envelope signal ED outputted from this envelope generator 110, so a memory control part 100 becomes data in a lateral axis direction for selecting a musical sound waveform group by the envelope signal ED. In the meanwhile, the envelope signal ED is inputted to a level change ratio calculating part 120, where a change ratio of the envelope signal ED per unit time is calculated. The memory control part 100 takes this level change ratio signal DELTAED for data in vertical axis direction for selecting the musical sound waveform group, and a waveform group selection signal SL is outputted.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a musical tone signal generating device that generates a musical tone signal while controlling the volume of the musical tone to be generated.

[Prior art]

Conventional musical tone signal generators store multiple waveform sample data groups with slightly different waveforms corresponding to the volume in a waveform memory, and during a performance, switch the waveform memory in real time according to the volume. The tone was changing. Further, as a musical tone signal generating device that switches the waveform of the sustaining portion while cross-fading one after another as time passes, there is known a device shown in Japanese Patent Publication No. 2-30033.

[Problem to be solved by the invention]

By the way, in an actual performance, when the volume is gradually increased or decreased, even if the volume is the same, the waveform will slightly differ depending on the degree of change in the volume. However, although the above-mentioned conventional musical tone signal generators can change the tone in response to the passage of time or the volume, they cannot respond to the degree of change in volume, so they can gradually increase the volume or reduce the volume. It was not possible to reproduce the change in waveform when gradually decreasing the amount. The present invention was made to solve this problem, and the volume changes over time by real-time operations such as aftertouch, express pedal, and press sensor (it is possible to reproduce subtle waveforms in the process). An object of the present invention is to provide a musical tone signal generating device capable of generating a musical tone signal.Means for Solving the Problems] In order to achieve the above object, the structural feature of the invention according to claim 1 is to form and output a musical tone signal. and a musical tone signal forming means for forming a control waveform signal that changes over time to control the volume and timbre, and outputting the control waveform signal to the musical tone signal forming means to control the volume and timbre of the musical tone signal. In addition to detecting the degree of change in the control waveform signal and controlling the timbre based on the control waveform signal, the musical tone signal generation device includes control waveform signal generation means. The musical tone signal is configured to control the timbre of the musical tone signal.Another structural feature of the invention according to claim 2 is that in the musical tone signal generating device according to claim 1, the musical tone signal forming means comprises: A plurality of sets of musical sound waveform sample data groups corresponding to the timbres are stored, and a musical sound signal is formed based on the musical sound waveform sample data groups corresponding to the control waveform signal and the detected degree of change. Furthermore, a structural feature of the invention according to claim 3 is that in the musical tone signal generation device according to claim 1, the musical tone signal forming means divides each harmonic component of the musical tone into a predetermined ratio. and a harmonic synthesized musical tone signal forming means for synthesizing the harmonics to form a musical tone signal, and changing the ratio according to the control waveform signal and the degree of change in the detected tone to correspond to a desired tone. Furthermore, the structural feature of the invention according to claim 4 is that it is configured to form a musical tone signal that is
In the musical tone signal generating device described in , the musical tone signal forming means includes a filter that controls the tone of the musical tone signal, and the tone of the filter is adjusted according to the control waveform signal and the degree of change in the detected tone. The present invention is constructed so that a musical tone signal corresponding to a desired timbre is formed by changing filter characteristics. The structural feature of the invention according to claim 5 is that in the musical tone signal generating device according to claim 1, the musical tone signal forming means includes a modulation calculation musical tone synthesis means for forming a musical tone signal by a modulation calculation. The musical tone signal corresponding to a desired timbre is formed using the calculation parameters corresponding to the timbre control waveform signal and the detected degree of change in timbre.

[Action of the invention]

In the invention according to claim 1 configured as above,
When the musical tone signal forming means forms and outputs a musical tone signal corresponding to the volume and timbre corresponding to the control waveform signal outputted from the control signal generating means, it detects the degree of change in the control waveform signal and converts it into the control waveform signal. In addition to controlling the tone color based on the detected change degree, the tone color in the musical tone signal is controlled based on the detected degree of change. In other words, when generating a musical tone signal while changing the volume and timbre based on the control waveform signal, the timbre is changed and controlled in response to the degree of change in volume, and the tone is changed in accordance with the degree of change in volume during performance. Reproduce change. Further, in the invention according to claim 2, the storage means in the musical tone signal forming means stores a plurality of sample data groups of musical waveforms corresponding to timbres, and the storage means of the musical tone signal forming means stores a plurality of sample data groups of musical waveforms corresponding to timbres, and A musical tone signal is formed based on a group of sample data of musical waveforms corresponding to the degree of change. Furthermore, in the invention according to claim 3, when the harmonic synthesized musical tone signal forming means in the musical tone signal forming means synthesizes each harmonic component of a musical tone at a predetermined ratio to form a musical tone signal, the control waveform The ratio is changed according to the signal and the detected degree of change in tone color to form a musical tone signal corresponding to a desired tone color. Further, in the invention according to claim 4, when the musical tone signal forming means controls the tone of the musical tone signal by the filter, the filter of the filter is adjusted according to the control waveform signal and the detected degree of change in tone. A musical tone signal corresponding to a desired tone is formed by changing the characteristics. In the invention according to claim 5, when the modulation calculation musical tone synthesis means in the musical tone signal forming means forms the musical tone signal by modulation calculation, the timbre control waveform signal and the detected degree of change in timbre are combined. A musical tone signal corresponding to a desired tone is formed using the corresponding calculation parameters.

【Effect of the invention】

As described above, according to the present invention, the timbre of the musical tone signal can be changed in accordance with the volume and the degree of change in the same volume,
It is possible to further improve expressiveness by reproducing subtle waveform changes that occur during performances in which the volume is gradually increased or decreased. Further, in the inventions according to claims 2 to 6, according to the invention according to claim 2, it can be applied to a device that forms a musical tone signal based on waveform sample data, and according to the invention according to claim 3, it is possible to use harmonic synthesis. The present invention can be applied to a device that forms a musical tone signal, and according to the invention according to claim 4, it can be applied to a device that forms a musical tone signal by changing filter characteristics, such as a so-called digital filter, and the invention according to claim 5 Accordingly, the present invention can be applied to a device that forms a musical tone signal by calculation, such as a so-called frequency modulation method.

【Example】

Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 schematically shows the entirety of an electronic musical instrument to which the present invention is applied.
Generates musical tone signals using the CM method. This electronic musical instrument has a performance operator group 10 consisting of a keyboard, switches, etc., and an adjustable pedal 20 consisting of a pedal and a sensor. The performance operator group 10 outputs keyboard performance data KD to the operator controller 30 in response to performance operations on the keyboard, and the operator controller 30 outputs the keyboard performance data KD from the performance operator group 10.
When input, the key code representing the pressed key, the ten-on indicating that the key is being pressed, and the key touch indicating the speed at which the key was pressed are detected from the same data KD, and the key code is detected for each sound channel. A key code signal KC, key-on signal KON, and key touch signal KT representing each are output. In addition, the express pedal 20 outputs pedal performance data PD to the operator control section 30 in response to the pedal operation by the performer to instruct the volume level. Same data P from pedal 20
When D is input, a pedal signal EP representing the volume level is output from the same data PD. A timbre selection setting operator group 40 is also connected to the operator control unit 30, and the operator group 40 is provided with various operators for selecting a pre-prepared timbre or setting an arbitrary timbre. . The timbre selection setting operator group 40 outputs timbre selection setting data SET to the operator controller 30 in accordance with the operation status of the operator. Among the various signals output from the operator control section 30, the key food signal 'C is input to the frequency number conversion section 50, which converts the frequency corresponding to the pitch represented by the cow chord signal KC. The number signal FN is output to the address generator 60. This frequency number signal FN represents a phase increment value for reading out the sample waveform, and has a value less than 1. The address generation section 60 accumulates the phase increment values represented by the frequency number and XH signals FN according to the timing of the clock signal, and stores sample waveform data corresponding to the discrete phases (\music waveform memory 1) 70 The integer part of the accumulated value is output as the address signal ADR3, and the decimal part of the accumulated value is sent to the interwaveform interpolator 90 as interpolated data FRA.
Output as C. The musical tone waveform memory 70 stores waveform sample data corresponding to a plurality of waveforms, as shown in FIG. Each waveform corresponds to a state in which the volume level and the degree of change in volume are different, and the phase of each waveform sample data is specified by the address signal ADR8 representing the integer part of the phase value output from the address generator 60. Then, one of the waveforms is selected by the waveform group selection signal SL output from the memory control section 100. Waveform group selection signal SL output by this memory control unit 100
In order to explain this, processing of data related to volume, which is the basis for forming the same waveform group selection signal SL, will be described. In addition to the key-coded MKC described above, the operator control unit 30 outputs a key-on signal KON, a key touch signal KT, and a pedal signal EP, and each signal is sent to the envelope generator 1.
10 is entered. The envelope generator 110 is
It outputs an envelope signal ED representing the envelope of the musical sound waveform shown in FIG. 3, and at this time, the overall volume level is controlled by the pedal signal EP. In addition,
For sustained musical tones, the output start point and decay start point of the envelope signal are controlled based on the key-on signal KON, and the angle (AR) and level (AL) at the attack part are controlled based on the key touch signal KT. The angle (DIR) and level (DIL) of the first decay part, the angle (D 2 R) and level (D2L) of the second decay part, and the angle (RLR) and level (RLL) of the release part are controlled. Since the volume level is controlled by the envelope signal ED output from the envelope generator 110, the memory control unit 100 uses the envelope signal ED to generate data in the horizontal axis direction for selecting the waveform group shown in FIG. shall be. On the other hand, the envelope signal ED is processed by the level change rate calculating section 12.
0, and the same level change rate calculation unit 120 calculates the change rate of the envelope signal ED per unit time. The level change rate calculating section 120 is configured as shown in FIG. The value held in the register 121 from a predetermined time ago is subtracted from , and a level change rate signal ΔED representing the difference is output to the memory control unit 100. The memory control unit 100 uses this level change rate signal ΔED as vertical axis data for selecting the waveform group shown in FIG. 2, and outputs a waveform group selection signal SL. however,
Since the waveform group stored in the musical sound waveform memory 70 corresponds to discrete values even in terms of the degree of change in the same level of volume, the envelope signal ED and the level change rate signal ΔED are A situation occurs in which the volume level corresponding to the stored waveform is an intermediate value that does not exactly match the degree of change of the same level. Therefore, in order to interpolate the waveform in such a case with the stored waveform group, the memory control unit 100 selects four waveforms including the same intermediate value from the waveform group arranged in a matrix and creates a waveform group. Outputs selection signal SL. As a result, four waveform sample data are read out from the musical waveform memory 70, and the memory controller 10 reads out four waveform sample data.
The inter-waveform group interpolator 8o performs interpolation based on the interpolated data frac output from the waveform group interpolator 8o. For example, based on the correspondence state with each waveform group, the envelope signal ED and the level change rate signal ΔED are set to the integer part 1 of the upper bits. J and the fractional part i of the lower bit. '', and if each waveform group arranged two-dimensionally according to the volume level (X) and the degree of change in the volume level (y) is specified by W (x, y), the selected 4
The waveform group is W(I, J). W(1,J+1), W(T+1.J), W(1+1,J
+1), and the interpolation calculation between the waveform groups is as follows. w (1, i, J, j) = (1-i)X (1-j)XW (1, J) + (1-1
)XjXW(1,J lacquer 1)+jX (1-j)X
W (1+I, J)+1XjXW(++1.J
+1) Specifically, as shown in FIG. 5, the memory control unit 100 has an interpolation counter 101 and a read memory selection circuit 102, and the interpolation counter 101 repeatedly repeats interpolation timing signals "0" to "5". As the output is output, the read memory selection circuit 1゜2 and the inter-waveform group interpolator 80 select "reset", "select waveform group W'(1, J)", and "select waveform group W(1, J11)". , "Waveform group W(1+1.'
J) selection", "Waveform group W(1+1.J+1) selection"
, repeats the process of "interpolation calculation and calculation result output". The waveform sample data output from the inter-waveform group interpolator 80 is inputted to the inter-waveform interpolator 9o as described above, and the same waveform amount interpolator 90 receives successive address signals (ADR3, A
Interpolation between waveforms is performed based on the two waveforms read by DR3+1) and interpolation data FRAC. The waveform data output from the inter-waveform interpolator 90 is sent to the multiplier 1.
30, and the multiplier 130 multiplies the same waveform data by the envelope signal ED. The signal output from the multiplier 130 is given an envelope, passes through the mixer section 140, is converted from digital to analog signal by the D/A converter 150, and is converted into a musical tone by the sound system 160, which includes an amplifier, a speaker, etc. and output. Next, the operation of the musical tone signal generator having the above configuration will be explained. The player can turn on the electronic musical instrument's power switch (not shown). Let's say you turn on the switch and start playing. It is assumed that the performer initially depresses the Expression A pedal to obtain a normal volume, and the volume level is a medium value. On the other hand, the performance on the keyboard is output from the performance operator group 1o as keyboard performance data KD, and the operator control section 30 outputs a key food signal KC to the frequency number conversion section 50 in response to the keyboard performance. The frequency number converter 50 converts the frequency number signal FN corresponding to the same key code signal MKC
is output to the address generator 60, and the address generator generates the same frequency number signal FN every time a clock signal is input.
It accumulates the phase increment value represented by , and outputs the integer part to the tone waveform memory 70 as the address signal ADR3. The operator control unit 30 outputs a key-on signal KON and a key touch signal KT to an envelope generator 110 in response to keyboard operations, and the envelope generator 110 generates an envelope signal ED based on each signal. When the volume level is ed (t) [t: time ko, the level change rate signal △ calculated by the level change rate calculation unit 120 is
ED, however, tl>to, tl-tO=1
Although expressed in msec, the volume level during a performance is determined roughly according to the pedal signal EP representing the amount of depression of the x-brain pedal, so ed (t)#f
(EP) holds true. Therefore, while the depression amount of the brake pedal is constant, the level change rate signal ΔED detected by the level change rate calculation unit 120 is approximately “0”, and the memory control unit 1
00 is a musical tone that selects a waveform group selection signal SL that selects a waveform group near A in FIG. Output to waveform memory 7o. In the musical waveform memory 70, four waveform groups are sequentially selected based on the waveform group selection signal SL from the memory control unit 100, and waveform sample data of a phase value corresponding to the address signal ADR3 is output for each waveform group. . The same waveform sample data is input to the waveform group interpolator 80, and one interpolated waveform sample data is derived from the four wave t sample data based on the interpolation data frac output by the memory control unit 100. Note that the interpolation itself is a normal linear interpolation in the vertical axis direction and the horizontal axis direction. After outputting the address signal ADR3, the address generator 60 outputs an address signal (ADR3+) that is one address different from the same address.
1), and the inter-waveform interpolator 90 outputs the waveform interpolator 90 based on the waveform sample data output from the inter-waveform interpolator 80 for each address and the fractional part FRAC of the phase value accumulated by the address generator 60. Performs linear interpolation between two waveforms. The waveform data output by the inter-waveform interpolator 90 is sent to the multiplier 130.
The envelope signal ED is input to the multiplier 130.
is multiplied by , a predetermined amplitude is given, the signal is converted to an analog value by the D/A converter 150 via the mixer section 140, and then generated as a musical tone by the sound system 160. While the expression pedal is not operated, different key code signals qKc are output based on the keyboard performance, and almost the same waveform group is selected in the musical waveform memory 70, and waveform data is obtained through the above two types of interpolation. and musical tones are generated. On the other hand, let us assume that the expression pedal is gradually depressed in order to cause the force to rise. Then, the volume level represented by the envelope signal ED increases, and the level change rate calculation unit 120
The level change rate signal ΔED calculated in is a positive value. Therefore, the memory control unit 100 outputs a waveform group selection signal SL that selects the waveform near B in FIG. 2 based on the envelope signal ED and the level change rate signal ΔED. The address generator 60 outputs an address signal ADR8 to the musical waveform memory 7° based on the frequency number signal FN corresponding to the keyboard operation, but in this case, the memory controller 100
Since the waveform group selection signal SL output by selects the waveform group near B, even if the waveform sample data is read for the same code, it is different from the waveform sample data read from the waveform group near A. is slightly different, with more harmonic components. The waveform sample data read out from the musical waveform memory 70 is output as a musical tone from the sound/stem as described above, and the subtle waveform changes at the time of creso/end are reproduced, giving a more realistic expressiveness. It is a musical sound. On the other hand, when you want to play at a normal volume and then return to a clear end, gradually return the express and nozzle pedals. When the pedal is released, the pedal signal EP corresponding to the amount of depression of the pedal gradually decreases, and the volume level represented by the envelope signal ED decreases.
The level change rate signal ΔED calculated by the level change rate calculation unit 120 has a negative value. Therefore, the memory control unit 10
0 is the same envelope signal ED and level change rate signal ΔED
A waveform group selection signal SL for selecting a waveform near C in FIG. 2 is output based on the waveform group selection signal SL. The waveform group near C has fewer harmonic components than the waveform group near A, and the musical tone output from the sound/stem 160 has a smoother waveform. In this way, according to this embodiment, the waveform memory to be read is changed depending on the volume level and the degree of change in the volume level, and in particular, the waveform is changed in response to the degree of change in the volume level. Therefore, it is possible to reproduce subtle changes in waveform when creating a frenon end or a deklenon end. FIG. 6 shows another example of a circuit for calculating the volume level signal ecl. The performance input device [170 outputs an initial touch signal IT, an aftertouch signal AT, and an external pedal (iqEP), which are data related to volume among performance data corresponding to keyboard operations, etc., and includes an envelope generation circuit. 111 forms an envelope signal ED based on the initial touch signal IT.The envelope signal ED output from the envelope generating circuit 111 is multiplied by the waveform data by the multiplier 130 in the above example, but at the same time, the envelope signal ED is output by the envelope signal ED based on the initial touch signal IT. The aftertouch signal AT and pedal signal EP output by the device 170 are input to the adder 112, and the respective data are added by the adder 112.The added value may exceed the permissible value for the volume level. The added value is input to the limiter 113, and when the added value is less than the allowable value, the limiter 113 outputs the added value as a volume level signal, and when the added value exceeds the allowed value, the upper limit value is set as the volume level signal. The memory control unit 100 outputs the volume level signal ed as a signal ed.
Level change rate calculation unit 1 based on the same volume level signal ed
A waveform group in the tone waveform memory 70 is selected based on the level change rate signal Δec3 calculated by the tone waveform memory 70. In such a configuration, the amount of depression of the aftertouch or express pedal, which directly specifies the volume, can be changed.
Since this signal is added to the envelope signal, which constantly changes with each key press, and used to determine the volume level, it is possible to more accurately determine talessi end and decresi end. Next, a second embodiment of the present invention will be described. FIG. 7 schematically shows the entirety of another electronic musical instrument to which the present invention is applied, which synthesizes harmonic components based on spectrum data, reproduces desired waveform data, and produces musical sounds. Generate a signal. Components that are the same as or correspond to those in the first embodiment are given the same reference numerals. The envelope signal ED output from the envelope generator 110 is input to the spectrum data control section 180 and the level change rate calculation section 120. Spectral data representing a predetermined waveform is read out from the spectrum data memory 190 based on the output level change rate signal ΔED, and the address generation section 60 is controlled based on the same data. That is, the address generating section 60 accumulates the phase increment value represented by the frequency number signal FN, but the spectral data control section 18
0 causes the phase value of each harmonic component represented by the spectrum data to be calculated based on the in-phase increment value, and provides an address to the sine wave memory 200 in order to read out the sine wave corresponding to the calculated phase value of each harmonic component. The signal ADRS is output. In addition, at this time, consecutive addresses (A D RS, A
DRS+1) is output, and linear interpolation is performed by the waveform interpolator 90 as in the first embodiment. The spectrum data control section 180 controls the address generation section 60 as described above, and also controls the synthesis of each harmonic component in the synthesizer 210. Synthesizer 2
The details of 10 are shown in FIG. 8, and include a multiplier that multiplies the sine wave data of each harmonic component output from the inter-waveform interpolator 90 and the level of each harmonic component output from the spectral data control unit 180. 211, an adder 212 that sequentially accumulates each harmonic component output from the multiplier 211, and a shift register 213. Note that when each harmonic of one waveform is synthesized by a selector (not shown), waveform data is output from the synthesizer 210, and at the same time, the accumulated value up to that point is reset. The spectrum data memory 190 from which the spectrum data is read out in this way stores spectrum data corresponding to a plurality of waveforms as shown in FIG. The degrees of change correspond to different states. Then, under the control of the spectrum data control section 180, spectrum data representing a predetermined waveform is read out. Next, the operation of this embodiment having the above configuration will be explained. Assuming that the performer depresses the expression pedal to obtain a normal volume, the volume level will be at a medium value. Further, when the performer operates the keyboard to start playing, the keyboard performance data KD is output from the performance operator group 10, and the operator controller 30 converts the key code signal KC to the frequency number converter in response to the keyboard performance. 5
Output to o. The frequency number conversion section 50 outputs the frequency number signal FN corresponding to the key code signal KC to the address generation section 60, and the address generation section 60 converts the frequency number signal FN corresponding to the same key code signal KC to the address generation section 60. Accumulate phase increment values. On the other hand, the operator control unit 30 outputs a key-on signal KON corresponding to the keyboard operation, a key touch signal KT, and a pedal signal EP corresponding to the amount of depression of the expression pedal.
is output to an envelope generator 110, which generates an envelope signal ED based on each signal.
occurs. As long as the amount of depression of the expression pedal is constant, the level change rate signal ΔED detected by the level change rate calculating section 120 becomes approximately "0" as described above, and the spectral data control section 180 Based on the volume level determined according to the signal EP and the level change rate signal ED having a value of approximately "0", the spectral data memory 19
From 0, spectrum data near D in FIG. 9 is read out. After reading out the spectrum data, the spectrum data control unit 180 causes the address generation unit 60 to sequentially calculate the phase value corresponding to each harmonic component, and sends the integer part of the phase value to the address signal q(ADR3°ADRS+1). and output it to the sine wave memory 200 as FR
AC is output to the inter-waveform interpolator 90. As a result, sine wave data linearly interpolated based on the sine wave data at two consecutive addresses stored in the sine wave memory 200 and the decimal part FRAC is output to the synthesizer 210, and the sine wave data is In the synthesizer 210, the signal is multiplied by a multiplier 211 together with the level of each harmonic component output from the spectrum-data control section 180. The product output from the multiplier 211 is waveform data of each harmonic component, and the product is output from the adder 212 and the shift register 21.
3, each harmonic component is accumulated sequentially. After all harmonic components have been accumulated, the output of the shift register 213 is transferred to the multiplier 1.
30 and starts accumulating waveform data for the next phase value. The waveform data output from the synthesizer 210 is sent to the multiplier 13
It is multiplied by the envelope signal El) output from the envelope generator 110 by 0 to give a predetermined envelope. After that, mixer section 140 and D/A converter 15
0 and is converted into an analog signal and generated as a musical tone by the sound system 160. As long as the depression amount of the expression pedal is not changed, even if different key food signals KC are output based on the keyboard performance, the spectral data read from the spectral data memory 190 is of almost the same waveform group, and such spectral data A musical tone is generated based on waveform data corresponding to the waveform data. On the other hand, as the performer gradually depresses the expression pedal to bring the watercress to an end, the volume level represented by the envelope signal ED increases, and the level calculated by the level change rate calculation unit 120 increases. The rate of change signal ED takes a positive value. Therefore, the spectrum data control unit 180 reads spectrum data corresponding to the waveform near E in FIG. 9 based on the envelope signal ED and the level change rate signal ΔED. As in the case of the above-described embodiment, the waveform during watercress ending is slightly different from the waveform during normal performance, and the spectrum data is also different. By synthesizing waveforms in accordance with such spectral data, the sound system outputs musical tones that are more realistic and expressive, reproducing the subtle changes in waveforms that occur when watercress ends. On the other hand, Toki, who had been playing at a normal volume until the end of the play, gradually returns the XBRENNE pedal. When the pedal is released, the pedal signal EP corresponding to the amount of depression of the pedal gradually decreases and becomes the envelope signal E.
As the volume level represented by D decreases, the level change rate signal Δ calculated by the level change rate calculation unit 120 increases.
ED becomes a negative value. Therefore, the spectrum data control unit 180 uses the envelope signal ED and the level change rate signal Δ
Based on ED, spectrum data near F in FIG. 9 is read out. The waveform based on the spectrum data near FN has fewer harmonic components than the waveform near D, and the synthesizer 21
The musical waveform signal synthesized by 0 and given an envelope by the multiplier 13o becomes a musical tone with a smoother waveform. In this way, in this embodiment, spectrum data of the waveform played in each case is stored corresponding to the volume level and the degree of change in the volume level, and the spectrum data of the waveform played in each case is stored. In addition to reading out predetermined spectrum data, harmonic components are waveform-synthesized based on the read spectrum data to reproduce the waveform data. Next, a third embodiment of the present invention will be described. FIG. 10 schematically shows the whole of another electronic musical instrument to which the present invention is applied, and the same reference numerals are given to the same or corresponding parts as in the first embodiment or the second embodiment. ing. In this embodiment, the performance control group 15 detects the speed at which the key is pressed as well as the suppression when the key is pressed, and detects the speed at which the key is pressed as an initial touch, and uses the suppression as an aftertouch in the keyboard performance data. It is sent to the operator control section 35 by the KD. The operator controller 35 detects initial touch and aftertouch from the keyboard performance data KD sent from the performance operator group 15, and outputs an initial touch signal IT and an aftertouch signal AT representing each to the envelope generator 114. Furthermore, tone selection setting data SET is output to the operator control unit 35 in accordance with the operation status of the operators in the tone selection setting operator group 4o, and the operator control unit 35 outputs the tone selection setting data SET based on the data SET. The signal TC is sent to the waveform signal generation circuit 3.
oo and is output to the envelope generator 114. The envelope generator 114 receives the key food signal KC, cow-on signal KON, and pedal signal EP as well as the initial touch signal IT, aftertouch signal AT, and tone signal Tc. As shown, there is a basic envelope generating circuit (EG B) 114a that generates a basic envelope, an initial touch envelope generating circuit (EG IT) 114b that generates an initial touch envelope, and an aftertouch envelope generating circuit that generates an aftertouch envelope. Envelope generation circuit (EG AT) 1j4c and pedal envelope generation circuit (EG EP) 1 that generates a pedal envelope.
14d. The envelope waveform and output timing of each envelope generation circuit are controlled based on the key-on signal KON, the key code signal KC, and the tone signal TC, and further includes an inital touch envelope generation circuit 114b and an aftertouch envelope generation circuit 114c. Regarding the pedal envelope generation circuit 114d, a multiplier 114e is used, respectively.
~114g, the output envelope is multiplied by an initial touch signal IT, an aftertouch signal AT, and a pedal signal EP. Note that each envelope generation circuit 18 has a memory in which the envelope is stored, and the envelope generated by each envelope generation circuit has an attack section, a loop section, and a loop section, as shown in FIGS. Based on the key-on signal KON, the attack part envelope is output immediately after the key is pressed, and after the attack part is output, the loop part is repeatedly output until the key is released. The release part is output when the key is released. Of course, the envelope may be determined by calculation. Note that when the release section is output, the subsequent envelope waveform is output from a portion at the same level as the previous volume level so that the previous volume level does not change. Further, a basic envelope waveform is selected based on the timbre signal TC, and key scaling is performed based on the key code signal KC. As is clear from each envelope waveform, the shape of the envelope changes in various ways depending on the combination of the initial touch, aftertouch, and pedal operation amount. In addition, it is possible to realize characteristics suitable for actual situations in which the initial touch is effective immediately after a key is pressed, and the aftertouch effect gradually increases. Furthermore, the loop section of the aftertouch envelope provides a vibrato effect. The basic envelope and the above-mentioned multiplied Inin Yalta, Chamber Loft After Tano Chamber Robe Pedal Envelope are input to the adder 114h, and the same adder]1
4h adds each envelope and outputs it to the interpolation circuit 1 41. Since the envelope is a discrete value over time, the interpolation circuit 1141 performs interpolation. The values indicated by the envelopes so far are logarithmic values, and since it is more realistic to correspond to logarithmic values in terms of hearing, the output of this interpolation circuit 1141 is used for timbre control. Furthermore, since it is better to use a constant when giving an envelope to waveform data, the constant is converted into a constant by the logarithmic constant conversion circuit 115 and then input to the multiplier 130 . In the above example, a total of four envelope generation circuits are used, but the basic envelope waveform can be selected using initials and keys, so that the configuration can be made up of three envelope generation circuits. Specifically, as shown in FIG. 16, the initial touch signal IT along with the key-on signal KON, key food signal KC, and tone signal TC is input to the basic envelope control circuit 114j, and the basic envelope control circuit 114J generates the basic envelope generating circuit. 11
The configuration is such that the generation of the basic envelope by 4K is controlled. Furthermore, one envelope generation circuit may be used in a time-division manner. The logarithmic envelope signal EDt output by the interpolation circuit 1141 in the envelope generator 114 is input as is to the waveform signal generation circuit 300, and is also input to the level change rate calculation unit 123 to generate the level change rate signal ΔEDt.
and is input to the same waveform signal generation circuit 300. The level change rate calculation unit 123 in this embodiment calculates the level change rate signal ΔEDt at even more minute timing than in the first and second embodiments, and corresponds to the timing at which one waveform data is calculated. It is calculated as follows. That is, as shown in FIG. 7, it has a 1 clock bit delay circuit 124 and a subtracter 125.
A subtracter 125 calculates the difference between the envelope signal EDtl of one clock bit before and the current envelope signal EDt delayed in step 24. Although the level change rate calculation unit 123 shown in FIG. 17 calculates only the slope of the volume level, it may be configured to take into account the acceleration of the volume level change. in particular,
As shown in FIG. 18, delay circuits 124a, 12
Connect 4 b in series! While holding the envelope signal ffi signal EDtl with a clock bit delay and the envelope signal EDt2 with a two clock bit delay, the coefficient α, (1-α) generated by the coefficient generation circuit 126 is applied to the multiplier 1.
27a and 127b multiply the envelope signal EDt by the envelope signal EDt2 delayed by one clock pit, and the adder 128 multiplies the envelope signal EDt2 after the multiplication.
t and envelope signal EDt2 are added together and 1
Subtract the clock pit delayed envelope signal EDtl. In such a configuration, the coefficient generation circuit 126
It is also possible to select one that takes acceleration into account and one that does not take acceleration into account, depending on the coefficient α generated by . For example, when α=0.5, EDt=ed t (t) EDtl=ec! t (t-Δt) EDt2=edt (t, -2Δt), then ΔEDt=O15edt(t) -edt (t-Δt) +0. 5eclt(t-2Δt) = 1/2 ((edt (t)-edt (t-
Δ1))-(edt(t-Δt)-ed t(t
-2Δ1)), so it becomes acceleration, and in the case of α-1, ΔEDt=edt (t)-edt (t-Δt)
Therefore, like the one shown in FIG. 17, only the inclination is considered. The waveform signal generation circuit 300 receives the phase information adrs output from the address generator i65 and the envelope signal EDt.
and level change rate signal ΔEDt, and tone signal TC
The waveform signal is generated in accordance with the tone color selected based on the key code signal KC and the pitch of the played tone, and similarly to the first embodiment, the musical tone is generated from the musical waveform memory shown in FIG. It consists of something that reads out waveforms. Note that the address generator 65 accumulates the frequency number signal fn output by the frequency number converter 55, but the frequency number signal fn output by the frequency number converter 55 is an integer value, and the address generator 65 accumulates the frequency number signal fn output by the frequency number converter 55. The cumulative value accumulated by 65 is also only integer digits. The reason for doing this is to avoid performing interpolation between waveforms, but it is also possible to perform interpolation calculations between waveforms using decimal places as in the first and second embodiments. . The waveform signal generation circuit 300 may have other configurations as long as it reproduces a musical sound waveform based on the input signal. For example, FIG. 19 shows a configuration for reproducing a musical sound waveform by synthesizing harmonic components as in the second embodiment. In this case, if the envelope signal EDt, the level change rate signal ΔEDt, the timbre signal TC, and the key code signal KC are input to the phase information aclrs, the coefficient generation circuit 301 is the mixing ratio Lk (EDt, ΔEDt, TC, KC) and the phase deviation φk (EDt, ΔEDt, TC, K) corresponding to the level of the harmonic component.
C), and the sine wave generator 302 outputs phase information adrs
and the phase deviation φk(EDt, ΔED
t, TC, KC) w 0(adrs, k, φk(EDt, ΔEDt, T
C, IC))=sin(k-adrs+φk(EDt,
By outputting ΔEDt, TC, KC)), multiplying it by the mixing ratio Lk in the multiplier 304, and combining it in the synthesizer 210, wl=Σ 1Lk(EDt, ΔEDt,
Ding C, KC) K=1 x 5in (k-adrs+φ k (EDI, △
Harmonic synthesis is performed according to the arithmetic expression EDt, TC, KC)>1. That is, when the overtone slot counter 303 outputs the slot signal k to control the coefficient generation circuit 301, sine wave generator 302, and synthesizer 210, rmJ
If is the number of harmonics for harmonic synthesis, the same slot signal repeatedly changes from "0" to "m+1", "0" represents a reset, and "1" to rmJ represents the calculation of each harmonic component. and represents the accumulation, and rm+1j represents the output of the composite value. Note that while the second embodiment outputs only the mixing ratio Lk as spectrum data, the present embodiment also outputs the phase deviation φ. In addition, as shown in Fig. 20, below the sharpness of the resonance, rQJ
That's what it means. ) and a digital filter 309 whose cutoff frequency can be changed to pass the tone waveform signal read from the waveform memory 306 to reproduce a desired tone waveform. In this example, the phase information a'drs and the key food signal KC
The readout control circuit 305 outputs the readout address of the waveform memory 306 in response to the tone signal TC and the tone signal TC, and inputs the musical waveform signal stored at the readout address to the digital filter 309. On the other hand, envelope signal E
Dt and the level change rate signal △EDt are connected to the Q parameter conversion circuit (FQEG) 307 and the cutoff frequency conversion circuit F.
CEG) 308, and the Q and cutoff frequency corresponding to each case are input to the digital filter 309.
It is output to. Therefore, the tone waveform signal read from the waveform memory 306 passes through the digital filter 309 having the characteristics of the specified Q and cutoff frequency, and becomes a desired tone waveform signal. Further, FIG. 21 shows an example of changing calculation parameters in a frequency modulation tone synthesis calculation circuit 310 that forms a waveform signal using a frequency modulation method. The frequency modulation musical tone synthesis calculation is to calculate and output a musical sound waveform based on the calculation formula v=Acos(ωct+5Hsin(ωs)+φn), and select the parameters in the calculation formula as appropriate according to the timbre and pitch. Generates a musical tone signal with a predetermined pitch and timbre.In the example shown in FIG. The readout parameters are appropriately selected according to the envelope signal EDt and the level change rate signal ΔEDt. In this way, the waveform signal generation circuit 300 selects the waveform based on the envelope signal EDt and the level change rate signal ΔEDt. Specifically, in the example shown in FIG. 5, waveform sample data in the musical waveform memory 70 is selected,
In the case shown in FIG. 19, spectral data is selected, and in the case shown in FIG. 20, characteristic parameters of a digital filter are selected. Each parameter is stored in a memory, and a two-dimensional parameter table as shown in FIG. 22 specified by the envelope signal EDt and the level change rate signal ΔEDt is prepared in the memory. Therefore, in the case of a musical waveform memory, each element is the sampling waveform data for one waveform, in the case of spectrum data, it is the mixing ratio and phase deviation of harmonic components in one waveform, and in the case of a digital filter, it is the sampling waveform data for one waveform. In order to reproduce this, the characteristic parameters Q and cutoff frequency are used. This parameter table also shows the level change rate signal Δ
When EDt is a large positive value, the envelope signal ED
Regardless of the value of t, a parameter representing the waveform of the attack part (when the AT area is memorized and the level change rate signal ΔEDt is a positive value close to "0", a parameter representing the waveform of the loop part with a slightly larger amplitude value) ■10 areas) are memorized,
When the level change rate i No. ΔEDt is a negative value near “0”, a parameter (V-region) representing the waveform of the loop portion whose amplitude value is slightly small is stored, and the level change rate signal ΔE
When Dt becomes small, the temperature is divided into the following sections according to the value of the envelope signal EDt. That is, when the envelope signal EDt has a thick %z value, the parameter (1stD area) representing the waveform of the first decay part following the attack part is stored, and when the envelope signal EDt has a small value, the level change rate signal ΔEDt is stored.
As becomes smaller, a parameter (2ndD area) representing the waveform of the second decay part immediately before the release part is stored,
Furthermore, when the level change rate signal ΔEDt becomes smaller, the parameter (RL region) representing the waveform of the release part is memorized. The harmonic components are gradually attenuated from the attack part to the release part. Fig. 23 shows the correspondence between the envelope waveform and the above region. Next, the operation of this embodiment having the above-mentioned configuration will be explained.Assuming that the performer depresses the adjustable pedal to obtain a normal volume, the volume level will be at a medium value. Furthermore, when the performer operates the keyboard to start playing, the performance operator group 15 outputs keyboard performance data KD, and the operator control unit 35 changes the key code signal KC to a frequency in response to the keyboard performance. Number converter 55
Output to. The frequency number converter 55 outputs the frequency number signal fn corresponding to the key code signal KC to the address generator 65, and the address generator 65 converts the phase increment represented by the frequency number signal fn every time a clock signal is input. Accumulate values. On the other hand, the operator control unit 35 outputs a key-on signal KON, a key code signal KC, an initial touch signal IT, an aftertouch signal AT corresponding to the keyboard operation, and a pedal signal EP corresponding to the depression amount of the express pedal. , a timbre signal TC corresponding to the operation of the timbre operator in the timbre selection setting operator group 40 is output to the envelope generator 114, and each envelope generating circuit 114a to 114d in the envelope generator 114 outputs a key-on signal KON and a key code. Each envelope is generated based on the signal KC and the tone signal TC. Among these envelopes, for the foot pedal envelope, an initial touch representing the absolute volume is applied to the output of each envelope generating circuit 114b to 114d at the power of 15114e to 114g. The signal IT, the aftertouch signal AT, and the pedal signal EP are multiplied, and the envelope signal after the multiplication and the basic envelope output from the basic envelope generation circuit 114a are added by an adder 114h. The envelope output from the adder 114h is sent to the interpolation circuit 11.
41, the waveform signal generator 300 and the level change rate calculator 1
2.3, the signal is converted into a constant via the logarithmic constant conversion circuit 115, and then input to the multiplier 130. Immediately after a musical note is sounded, the value indicated by the envelope is thick (as well as its rate of change is large) because it is the rising edge of the musical tone.Therefore, the envelope signal EDt changes from "0" to a large positive value, and the rate of change in level increases. The signal ΔEDt becomes a large positive value and is input to the waveform signal generation circuit 300. On the other hand, the key code signal KC corresponding to the pitch of the key pressed by the performance operator group 15 is sent to the frequency number converter. 55 and corresponds to the same key code signal KC.A predetermined frequency number signal fn is input to the address generation section 65.The address generation section 65 accumulates the same frequency number signal fn and generates phase information actrs into a waveform. It is output to the signal generation circuit 300, and the same waveform signal generation circuit 300 receives the same phase information a.
drS is input to the musical waveform memory 70, and the envelope signal EDt and level change rate signal ΔEDt are
, and the tone signal TC and key food signal KC are input to the read memory selection circuit 102. The read memory selection circuit 102 selects a waveform group related to the timbre specified by the timbre signal TC and a key code signal K.
Within the waveform group of the sound range specified by C, the integer part 1. of the envelope signal EDt and the level change rate signal ΔEDt. The readout waveform of the musical waveform memo IJ 70 is selected according to the interpolation timing signal k output from the interpolation counter 101. That is, in the two-dimensional table shown in FIG. 22, when the interpolation timing signal k changes from "0" to "5", waveform group W (+, J) is selected with "1", and "
2" selects waveform group W (+, J+1), "3" selects waveform group W (1+L J), and "4" selects waveform group W
Select (1+1.J+1'). Note that rOJ represents reset, and at "5", interpolation calculation and calculation result output are performed. However, at the rising edge of a musical tone, the envelope signal E
Since both Dt and the level change rate signal ΔEDt are large positive values, waveform data representing the waveform of the attack portion is selected as shown in the parameter table. The selected waveform data is sequentially read out from the musical waveform memory 70 in accordance with the phase information adrs outputted by the address generation section 65, and the interpolation counter 1 is input by the waveform group interpolator 80.
The decimal part of the envelope signal EDt and the level change rate signal △EDt according to the interpolation timing signal k output by 01.
+ Perform the interpolation described above from J. The waveform signal output from the inter-waveform group interpolator 80 is sent to the multiplier 1.
At step 30, the envelope output from the logarithmic constant conversion circuit 115 is added, and the envelope output from the logarithmic constant conversion circuit 115 is applied to the input signal output section 140 and the D/Ap conversion circuit 91.
It is output as a musical tone through 50 sounds/stem 160. Therefore, at the beginning of a musical note. This means that the waveform data in the lower part has been selected and reproduced as a musical tone. Following this, when the peak of the attack portion is exceeded, the envelope signal EDt gradually decreases, and the level change rate signal ΔEDt changes to a large negative value. Therefore, the parameter table outputs waveform data representing the waveform of the first decay section. By the time the attack part ends, the change in the envelope is small (and the level change rate signal ΔEDt it
r OJ, and the envelope signal qEDt shows an intermediate value in the same table, and changes only by the vertical fluctuation due to the loop portion of the aftertouch envelope. Then, in the same parameter table, waveforms with slightly larger amplitude values and waveforms with slightly smaller amplitude values are read out alternately, resulting in a musical tone with vibrato. Also, when the loop section ends, the envelope signal EDtO
The value becomes even smaller, and the level change rate signal △EDt
becomes a negative value as the envelope signal EDt decreases. Therefore, the parameter table outputs waveform data representing the waveform of the second decay section, and furthermore, the envelope signal EDt and the level change rate signal ΔEDt.
When becomes smaller, waveform data representing the waveform of the release portion in the same table will be output. In this manner, as each key is pressed, the waveforms in each case are reproduced from the attack section to the first decay section to the loop section, and from the second decay section to the release section, and are output as musical tones. Up until now, the normal pronunciation process has been assumed without any particular change in volume. However, if the performer gradually depresses the keyboard pedal to create a crenoscendo, or changes the suppression of the key to increase the volume level, a difference will occur in the selection of waveform data in the parameter table. For example, suppose a performer depresses the expression pedal while the attack section is being sounded normally and the loop section is being sounded. Since the pedal envelope is a constant envelope that slightly increases or decreases as shown in FIG.
The pedal envelope a- multiplied by the pedal signal EP, which is increasing at g, also increases. Therefore, the envelope signal EDt to which the pedal envelope 0 is added by the adder 114h also increases, and the level change rate signal ΔEDt takes a positive value corresponding to the degree of increase. Then, the waveform data of the attack section will be referenced in the parameter table. In an actual performance, when the sound pressure increases, the harmonic components of the waveform become larger and become closer to the waveform of the attack section. Therefore, as described above, by reproducing the waveform of the attack section while the loop section is being sounded, a waveform that matches the cresciendo performance will be reproduced. On the other hand, suppose that, on the other hand, while the attack section is being sounded normally and the loop section is being sounded, the performer gradually returns the amount by which the keyboard pedal is depressed. Based on the decreasing pedal signal EP, the pedal envelope temperature decreases, and the envelope signal EDt to which the pedal envelope is added also decreases. Therefore, the envelope No. 8 EDt becomes a slightly small value, and the level change rate signal ΔEDt becomes a negative value, so that the waveform data of the second decay section or release section is referred to in the parameter table. Referring to the actual performance in this case, when the sound pressure decreases, the harmonic components of the waveform become smaller.This is because the waveform from the second decay part to the release part becomes smaller when compared to the waveform in the loop part. will come close to. Therefore, as described above, by reproducing the waveform of the release section while the loop section is being sounded, a waveform suitable for the Decrease End performance will be reproduced. In the above example, we explained the case of reproducing waveform data from waveform memory, but it is also possible to reproduce a musical sound waveform by changing the harmonic synthesis mixing ratio, or by changing the characteristics of a digital filter, When a waveform signal is formed using the frequency modulation method, the same effect can be obtained if the parameters read from the parameter table are set to have the same characteristics as the waveform data described above. For example, in the case of harmonic synthesis, the mixing ratio and phase parameters are stored in the same table as the waveform data, and in the case of a digital filter, the Q parameter and cutoff frequency are stored in the same table as the waveform data, and the frequency modulation method In this case, the calculation parameters may be stored in the same table as the waveform data. In the above embodiment, an example of a keyboard instrument is shown, but the structure may be applied to other wind instruments, string instruments, or the like. Furthermore, although an instrument that changes pitch is described, it may be one that reproduces the sound of an instrument that does not change pitch, for example, it may be one that reproduces musical sounds such as drums or maracas. Furthermore, it is not limited to electronic musical instruments.
It may also be a device or a toy that produces sound effects. Similarly, the means to adjust the volume are not limited to keyboards and expression pedals, but also include breath sensors,
The operation of a drum pad, boltament bar, wheel, joystick, etc. may be detected by a volume sensor, a position sensor, an angle sensor, or the like. In this case, in the above embodiment, the timbre is changed depending on the volume, but it is not necessarily necessary to change the timbre depending on the volume. In addition, in the first embodiment, the parameters for each waveform group are also interpolated by an interpolator, but interpolation may be omitted between waveform groups.On the other hand, the interpolation calculation between waveforms is performed using a straight line. In addition to interpolation (first-order interpolation), various interpolation operations such as higher-order interpolation (polynomial interpolation, Lagrange interpolation, etc.) and digital filter interpolation are possible, depending on the system configuration and required interpolation accuracy. You can choose accordingly.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an electronic musical instrument to which a musical tone signal generating device according to an embodiment of the present invention is applied; FIG. 2 is a diagram showing waveforms stored in a musical sound waveform memory;
Fig. 3 is a diagram showing the change state of the envelope signal, Fig. 4 is a block diagram showing details of the level change rate calculation section, and Fig. 5 is a memory control section for performing interpolation between waveform groups, musical waveform memory, and waveforms. FIG. 6 is a block diagram showing the relationship between intergroup interpolators, FIG. 6 is a block diagram showing another configuration for detecting the volume level, and FIG. 7 is an electronic diagram to which a musical tone signal generating device according to another embodiment of the present invention is applied. Block diagram showing the schematic configuration of the musical instrument, No. 8
9 is a block diagram showing details of the synthesizer, FIG. 9 is a diagram showing spectral data stored in the spectral data memory, and FIG. 10 is a diagram to which a musical tone signal generation device according to another embodiment of the present invention is applied Figure 11 is a block diagram showing the configuration of the envelope generator, Figures 12 to 15 are diagrams showing envelope waveforms generated by the envelope generation circuit, and Figure 16 is a diagram showing the general configuration of the electronic musical instrument. FIG. 17 is a block diagram showing another example of the configuration of the envelope generator; FIG. 17 is a block diagram showing an example of the configuration of the level change rate calculating section; FIG.
The figure is a block diagram showing another example of the configuration of the level change rate calculation section, FIG. 19 is a block diagram of a waveform signal generation circuit using harmonic synthesis, and FIG. 20 is a block diagram of a waveform signal generation circuit using a digital filter. FIG. 21 is a block diagram of a waveform signal generation circuit that generates a musical sound waveform using a frequency modulation method, FIG. 22 is a diagram showing a parameter table, and FIG. 23 is a diagram showing the relationship between envelope waveforms and areas of the parameter table. . Explanation of the symbols 10, ' 15...Performance operator group, 20...Expression pedal, 30.35...Operator control unit,
50°55... Frequency number conversion section, 60.65...
・Address generation section, 70... Tone waveform memory, 80.
... Interpolator between waveform groups, 90 ... Interpolator between waveforms, 100
...Memory control unit, 110, 114...Envelope I:
l -F R generator, 120.123... Level change rate calculation unit, 130... Multiplier, 180... Spectrum data control unit, 190... Spectral data memory, 2
00...Sine wave memory, 210...Synthesizer, 211
... Multiplier, 212 ... Adder, 213 ... NFT register, 300 ... Waveform signal generation circuit, 306.
...Waveform memory, 307...Q parameter conversion circuit,
308...Cutoff frequency conversion circuit, 309...
digital filter.

Claims (5)

[Claims] (1) A musical tone signal forming means for forming and outputting a musical tone signal, and forming a control waveform signal that changes over time to control volume and timbre, and outputting the same control waveform signal to the musical tone signal forming means to produce the musical tone. A musical tone signal generating device comprising a control waveform signal generating means for controlling the volume and timbre of the signal, in addition to detecting the degree of change in the control waveform signal and controlling the timbre based on the control waveform signal. A musical tone signal generating device characterized in that the tone color in the musical tone signal is controlled based on the degree of change in the musical tone signal. (2) In the musical tone signal generating device according to claim 1, the musical tone signal forming means stores a plurality of sample data groups of musical waveforms corresponding to timbres, and the musical tone signal forming means stores a plurality of sample data groups of musical waveforms corresponding to timbres, and What is claimed is: 1. A musical tone signal generating device, characterized in that a musical tone signal is generated based on a group of sample data of a musical waveform corresponding to a degree of change. (3) In the musical tone signal generation device according to claim 1, the musical tone signal forming means includes harmonic synthesized musical tone signal forming means for synthesizing each harmonic component of a musical tone at a predetermined ratio to form a musical tone signal. A musical tone signal generating device, characterized in that the ratio is changed according to the control waveform signal and the degree of change in the detected tone color to form a musical tone signal corresponding to a desired tone color. (4) In the musical tone signal generating device according to claim 1, the musical tone signal forming means has a filter for controlling the tone of the musical tone signal, and the control waveform signal and the degree of change in the detected tone are A musical tone signal generating device, characterized in that a musical tone signal corresponding to a desired tone is formed by changing filter characteristics of the filter according to the desired tone. (5) In the musical tone signal generating device according to claim 1, the musical tone signal forming means has a modulation calculation musical tone synthesis means for forming a musical tone signal by a modulation calculation, and the musical tone signal generating means includes a modulation calculation musical tone synthesis means for forming a musical tone signal by a modulation calculation, and the musical tone signal generation device includes a modulation calculation musical tone synthesis device that forms a musical tone signal by a modulation calculation, and the musical tone signal generation device includes a 1. A musical tone signal generating device that generates a musical tone signal corresponding to a desired tone color using calculation parameters corresponding to the degree of change in tone color.