JPH0934465A - Method and device for generating musical sound signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and device for generation of musical sound signals, which can reproduce natural instrument sounds containing noise components to a certain appropriate extent. SOLUTION: A method and device for generation of musical sounds comprises a second-order sonic waveform memory 10 to store the waveform data in second-order sonic waveform prepared by removing noise components from the waveform of natural instrument sounds, a noise waveform memory 11 to store the waveform data in the noise waveform consisting of the difference between the waveform of natural instrument sound and second-order sonic waveform, the first weighting means 12a, 13 to make the first weighting according to the first characteristics in compliance with the attack intensity on the waveform data read from the second-order sonic waveform memory, and the second weighting means 12b, 14 to make the second weighting according to the second characteristics in compliance with the attack intensity on the waveform data read from the noise waveform memory 11, and a mixing means 17 which is to mix the waveform data given by the first weighting means 12a, 13 with the waveform data given by the second weighting means 12b, 14, and therewith musical sound signals are produced on the basis of the output of the mixing means 17.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone signal generator and a musical tone signal generating method suitable for, for example, an electronic musical instrument or a sound source board, and more particularly to a technique for faithfully reproducing a natural musical instrument sound containing a noise component. .

[0002]

2. Description of the Related Art Conventionally, there is known a tone signal generator which adopts a pulse code modulation (PCM) system. This PC
In the M-system tone signal generator, for example, waveform data obtained by pulse code modulating a natural musical instrument sound is stored in advance in a waveform memory. Then, when a tone generation command is issued, waveform data is sequentially read from the waveform memory, and a tone signal is generated based on the read waveform data. In such a PCM tone signal generator, various techniques are employed to faithfully simulate the natural musical instrument sound. In the following, an acoustic piano (hereinafter simply referred to as "piano") will be described as an example of one of the natural musical instruments.

In general, when a keyboard of a piano is tapped, a striking sound is generated along with the original sound of the piano due to the vibration of the strings. This striking sound is generated by a shock applied to a shelf board by a keystroke, a shock when a hammer interlocking with the key strikes a string, or the like. One of the major characteristics of the piano sound is that it includes the above-mentioned striking sound in addition to the sound due to the vibration of the strings.

Therefore, even in an electronic musical instrument to which a PCM type musical tone signal generator is applied, a technique for generating a sound including a striking sound has been developed in order to faithfully simulate the characteristics of the piano sound. .

In the first technique, first, a piano sound generated by keystrokes and including a percussion sound component (hereinafter referred to as "original sound") is recorded and pulse code modulated. Waveform data is created by. The waveform data thus created is stored in the waveform memory. Then, when a tone generation command is issued, the waveform data is sequentially read from this waveform memory. The read waveform data is filtered by the filter. Then, a musical tone signal is generated based on the filtered waveform data. Filtering is performed in order to change the characteristics (sound quality) of sound.

In the second technique, the waveform of a piano sound that does not include a percussion sound component, that is, a sound generated only by vibration of a string (hereinafter referred to as "secondary sound") is pulse code modulated. To create waveform data and
Stored in the waveform memory of. The waveform data of the secondary sound is
For example, it is produced by removing the striking sound component from the original sound by using, for example, a comb filter, and subjecting this to pulse code modulation. On the other hand, for example, a hitting sound generated by hitting a shelf of a piano is recorded, and the waveform of the hitting sound is pulse code modulated to create waveform data. The waveform data thus created is stored in the second waveform memory. Then, when a sound generation command is issued, the waveform data of the secondary sound is read from the first waveform memory and the waveform data of the impact sound is read from the second waveform memory, and mixed. Then, a musical tone signal is generated based on the mixed waveform data.

[0007]

According to the tone signal generator to which the above-mentioned first technique is applied, waveform data is created based on the piano sound that originally contained the striking sound,
Since a tone signal is generated based on this waveform data,
A natural piano sound can be obtained even during playback. By the way, it is known that the ratio of the volume of the hitting sound to the volume of the entire piano sound (or the ratio of the volume of the secondary sound) changes depending on the strength of keystroke, that is, the touch strength. That is,
When the touch strength is low, the ratio of the volume of the batting sound to the volume of the entire piano sound is high, and conversely, when the touch strength is high, the ratio of the volume of the batting sound to the volume of the entire piano sound is low.

However, in the first technique, since the volume of the secondary sound and the volume of the hitting sound cannot be controlled independently, the hitting sound (or 2) with respect to the entire volume of the piano sound according to the touch strength. It was not possible to change the volume ratio of the next tone). Therefore, in the tone signal generator to which the first technique is applied, the volume of the entire piano sound including the secondary sound and the percussion sound is changed according to the touch strength. For this reason, there is no problem when reproducing with the same touch strength as the touch strength at the time of recording, but there is a problem that when reproducing with a different touch strength, it becomes unnatural as a piano sound. For example, when smashing,
The volume of the striking sound was too loud relative to the volume of the entire piano sound, resulting in an unnatural sound as a piano sound.

Further, according to the second technique, the ratio of the volume of the hitting sound (or the secondary sound) to the volume of the entire piano sound can be arbitrarily changed. Therefore, the ratio can be changed according to the touch strength. Can bring the sound closer to a natural piano sound. However, since the striking sound waveform data is created based on the striking sound generated by striking the piano shelf, unlike the actual striking sound generated when the piano key is tapped, the piano sound is There was a problem of becoming unnatural.

The above problems have occurred not only in the case of a piano but also in the case of electrically producing sounds of keyboard instruments such as organs and harpsichords, wind instruments such as flutes and trumpets, and other various musical instruments. Generally, the impact sound included in the piano sound can be considered as a noise component having no periodicity. Sound caused by such noise components is also generated by breath leaking from the lips when, for example, playing a flute. The amount of such noise components changes depending on the attack strength (for example, the touch strength in the case of a piano, the magnitude of the breathing speed in the case of a flute, etc.). Therefore, with the above-described first and second techniques, it is difficult to faithfully simulate not only the piano sound but also the sound of other natural musical instruments containing noise components.

The present invention has been made in order to solve such a problem, and is a musical tone signal generating apparatus and a musical tone signal generating method for generating a musical tone signal capable of reproducing a natural musical instrument sound appropriately containing a noise component. The purpose is to provide.

[0012]

In order to achieve the above-mentioned object, a musical tone signal generator of the present invention stores a secondary sound wave which stores waveform data of a secondary sound waveform obtained by removing noise components from the waveform of a natural musical instrument sound. Shape memory, a noise waveform memory for storing waveform data of a noise waveform composed of a difference between the waveform of the natural musical instrument sound and the secondary sound waveform, and an attack on the waveform data read from the secondary sound waveform memory. First weighting means for performing a first weighting according to the first characteristic in accordance with the intensity, and waveform data read from the noise waveform memory, the first weighting means according to the second characteristic in accordance with the attack intensity. A second weighting means for performing a weighting of 2; and a mixing means for mixing the waveform data from the first weighting means with the waveform data from the second weighting means, based on the output of the mixing means. Generates a tone signal And wherein the door.

The secondary sound wave memory and the noise waveform memory are, for example, a read only memory (hereinafter referred to as "ROM"), a random access memory (hereinafter referred to as "RAM").
Say. ), And other memory.

The waveform data of the secondary sound waveform can be created by the following procedure, for example, by computer processing. Hereinafter, a case where waveform data of a piano sound is created will be described. First, the waveform of the piano sound (original sound) generated by keystrokes is recorded. An example of such an original sound spectrum is shown in FIG. The high-level spire-shaped component indicates overtones, and the low-level component existing between them indicates a frequency-independent noise component, that is, a percussion sound component. It is desirable that the keystroke at the time of recording is the strongest stroke, that is, the strongest touch. This is because the sound generated by the strongest hit includes overtones up to high order, and the control range of overtones by the first filter means described later is expanded. Then, the waveform of the recorded piano sound is digitized and loaded into a computer. This digitization can be done by pulse code modulation.

Next, a noise component is removed from the original sound by passing the recorded original sound through, for example, a comb filter. The waveform from which this noise component is removed is called "secondary sound waveform". An example of this secondary sound wave spectrum is shown in FIG. The secondary sound waveform is then digitized by pulse code modulation and captured by a computer. The one waveform data of the secondary sound waveform thus formed is composed of a plurality of sampling data and is stored in the secondary sound waveform memory.

Next, the secondary sound waveform data is subtracted from the original sound waveform data. The waveform formed by the waveform data obtained by this subtraction represents a noise waveform composed of only noise components. An example of this noise waveform is shown in FIG. The waveform data of the noise waveform thus produced is composed of a plurality of sampling data and is stored in the noise waveform memory.

Since the main component of the striking sound is often lower than the fundamental frequency (fundamental sound) of the piano sound (see FIG. 5),
By passing the original sound through a low-pass filter, pseudo waveform data of a noise waveform can be created.

Although an example of creating waveform data of piano sounds has been described above, waveform data of sounds of other keyboard musical instruments such as organs and harpsichords, wind instruments such as flutes and trumpets, and other various musical instruments are also described. ,
It can be manufactured by the same procedure as described above.

The first weighting means can be composed of, for example, a multiplier and a first weighting coefficient generator. This multiplier is a central processing unit (hereinafter referred to as "CPU") or a digital signal processor (hereinafter referred to as "CPU").
It is called "DSP". ), Or hardware.

Further, the first weighting factor generator can be constituted by a table in which data representing the attack strength and the weighting factor C1 are stored in association with each other. As the data representing the attack strength, for example, touch data representing touch strength in the case of a piano, and data obtained by converting the breathing speed into the touch strength in the piano can be used in the case of a flute. The same applies to the following second weighting factor generator. In the table, for example, as shown in FIG. 2, for each value of the data indicating the attack strength (touch data is used in FIG. 2), the weighting coefficient C1 according to the first characteristic for the secondary sound. Can be configured to output. Further, the first weighting factor generator can also be configured by a function generator. In this case, it is possible to use a function generator that inputs the touch data and generates the weighting coefficient C1 according to the first characteristic for the secondary sound of FIG. 2, for example. This function generator can be configured by arithmetic processing of CPU or DSP, or hardware. The output of this first weighting means is supplied to the mixing means.

The second weighting means can be composed of, for example, a multiplier and a second weighting coefficient generator. This multiplier can be configured by arithmetic processing by a CPU or DSP, or by hardware.

The second weighting factor generator can be constructed by a table in which data representing the attack strength and the weighting factor C2 are stored in association with each other. In this table, for example, as shown in FIG. 2, for each value of the data indicating the attack strength (touch data is used in FIG. 2), the weight according to the first characteristic for noise (striking sound). It can be configured to output the coefficient C2. The second weighting factor generator can also be configured by a function generator. In this case, a function generator which inputs touch data and generates a weighting coefficient C2 according to the second characteristic for noise (striking sound) of FIG. 2 can be used. This function generator is C
It can be configured by arithmetic processing of PU or DSP, or hardware. The output of this second weighting means is supplied to the mixing means.

The mixing means can be composed of, for example, an adder. This adder can be configured by arithmetic processing of CPU or DSP, or hardware. The mixing means includes the waveform data from the first weighting means and the second data.
The waveform data from the weighting means are mixed by addition. A tone signal is generated based on the waveform data from this mixing means.

Further, the tone signal generating apparatus of the present invention can further comprise first filter means for filtering the output of the first weighting means and supplying it to the mixing means. This first filter means is, for example, the first
And a first parameter generator that generates a parameter to be given to the first filter. The first parameter generator may be configured to generate a parameter for increasing the cutoff frequency as the data representing the attack strength increases and to supply the parameter to the first filter. A low-pass filter, for example, can be used as the first filter forming the first filter means.

The tone signal generating apparatus of the present invention may further comprise second filter means for filtering the output of the second weighting means and supplying it to the mixing means. This second filter means is, for example, the second
And a second parameter generator that generates a parameter to be given to this second filter. The second parameter generator may be configured to generate a parameter for changing the sound quality of noise according to the data representing the attack strength and supply the parameter to the second filter. A low pass filter, for example, can be used as the second filter forming the second filter means.

The musical tone signal generator of the present invention has a structure having either one of the first filter means or the second filter means, and has both the first filter means and the second filter means. The configuration may be such that neither the first filter means nor the second filter means is provided.

The musical tone signal generating method of the present invention comprises: (A) creating secondary sound waveform data in which noise components are removed from the waveform of the natural musical instrument sound; and (B) the waveform of the natural musical instrument sound and the secondary sound waveform. Waveform data of a noise waveform composed of a difference from the sound waveform is prepared in advance, and (C) the second sound waveform waveform data according to the sounding command, and the first characteristic according to the attack strength according to the first characteristic. 1 is applied, (D) the waveform data of the noise waveform is subjected to a second weighting according to a second characteristic according to the attack strength, and (E) the second weighted secondary sound wave is applied. Shape waveform data and the waveform data of the second weighted noise waveform are mixed, and a tone signal is generated based on the mixed waveform data.

The waveform data of the secondary sound waveform in the step (A) and the waveform data of the noise waveform in the step (B) can be produced by the same method as in the case of the tone signal generator described above.

The first weighting in the step (C) can be performed, for example, by multiplying the waveform data of the secondary sound waveform by the weighting coefficient C1. Weighting coefficient C1
As is the case with the tone signal generator described above,
It is possible to use the weighting coefficient C1 that changes according to the first characteristic for the secondary sound according to the data representing the attack strength.

Similarly, the second weighting in the step (D) is performed by adding a weighting factor C to the waveform data of the noise waveform.
This can be done by multiplying by 2. As the weighting coefficient C2, as in the case of the tone signal generator described above, noise (striking sound) is generated according to the data representing the attack intensity.
A weighting factor C2 that varies according to the second characteristic for can be used.

The mixing in the step (E) can be performed by adding the first weighted waveform data and the second weighted waveform data. A tone signal is generated based on the mixed waveform data.

Further, the tone signal generating apparatus of the present invention further comprises a step of filtering the waveform data of the first weighted secondary sound waveform in the step (E), and the filtered secondary sound wave. The waveform data of the shape can be configured to be mixed with waveform data of a noise waveform before being filtered or waveform data of a noise waveform after being filtered. In this filtering, it is possible to perform filtering with a characteristic that the cutoff frequency becomes higher as the data representing the attack strength becomes larger.

Further, the tone signal generator of the present invention further comprises a step of filtering the waveform data of the second weighted noise waveform in the step (E).
The waveform data of the filtered noise waveform can be configured to be mixed with the waveform data of the secondary sound waveform before filtering or the waveform data of the secondary sound waveform after filtering. In this filtering, it is possible to perform filtering with a characteristic that changes the sound quality of noise according to the data representing the attack strength.

[Action]

In the tone signal generator and tone signal generating method of the present invention, the secondary sound waveform data obtained by removing the noise component from the waveform of the natural musical instrument sound is stored in the secondary sound waveform memory,
Waveform data of a noise waveform formed by the difference between the waveform of the natural musical instrument sound and the secondary sound waveform is stored in the noise waveform memory. Then, for example, when a sounding command is issued, the waveform data is read from the secondary sound waveform memory, and the first weighting according to the first characteristic for the secondary sound is performed on the waveform data. Similarly, the waveform data is read from the noise waveform memory and the second waveform for noise is read according to the attack intensity.
The second weighting is performed according to the characteristic of. Then, the first weighted waveform data and the second weighted waveform data are mixed, and a tone signal is generated based on the mixed waveform data.

As described above, according to the musical tone signal generating apparatus and the musical tone signal generating method of the present invention, the waveform data of the secondary sound waveform and the waveform data of the noise waveform are stored separately, and they are stored according to the attack strength. Since each waveform data is individually weighted to adjust the mixture ratio of the secondary sound and noise, it is possible to realize the same mixture ratio as the mixture ratio of the secondary sound and noise in the natural musical instrument. As a result, it is possible to generate a musical tone signal having a quality close to that of a natural musical instrument. Further, since the waveform data of the noise waveform is created by subtracting the secondary sound from the natural musical instrument sound, the noise waveform of the noise waveform is generated based on the striking sound generated by striking the shelf of the piano as in the conventional technique. The unnaturalness that occurs when waveform data is created is eliminated.

[0036]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of a musical tone signal generating apparatus and a musical tone signal generating method of the present invention will be described below in detail with reference to the drawings. In the following embodiments, as an example of a natural musical instrument sound, a case where a piano sound is electrically created will be mainly described.

FIG. 1 is a block diagram showing the configuration of an embodiment of the musical tone signal generating apparatus of the present invention. This tone signal generator includes a secondary sound waveform memory 10, a noise waveform memory 11, weighting coefficient generators 12a and 12b, multipliers 13 and 14,
It is composed of filters 15 and 16, an adder 17, a multiplier 18, parameter generators 19 and 20, and an envelope generator 21.

Tone data, tone range data, and touch data are externally supplied to the tone signal generator. Among these data, the "tone color data" is created based on the setting state of the tone color selection switch provided on the operation panel of the electronic musical instrument to which the musical tone signal generator is applied, for example. In addition, "range data" and "touch data",
It is produced based on the operation state of the keyboard device of the electronic musical instrument to which the musical tone signal generator is applied. In this case, as the keyboard device, a two-contact type keyboard device in which each key has two key switches that are turned on at different pressing depths is used. Then, the data representing the key range to which the key number of the pressed key is applied is used as the range data. Also, the time from when one key switch is turned on to when the other key switch is turned on is measured according to the keystroke, and the data corresponding to the measured time is used as touch data representing the keystroke strength. . For example, a CPU (not shown) fetches each of these data from the operation panel or keyboard device and sends them to the tone signal generator.

Further, in the electronic musical instrument having the MIDI interface, the tone color number included in the received program change message can be used as tone color data. Further, the data representing the key range to which the note number included in the note-on message belongs can be used as the range data, and the velocity can be used as the touch data. For each of these data, for example, a CPU (not shown)
It is configured to be extracted from the IDI message and sent to the tone signal generator.

The secondary sound wave type memory 10 in this embodiment is
It is assumed to be composed of ROM. The secondary sound wave type memory 10 may be composed of a RAM. In this case, it is necessary to load the waveform data of the secondary sound waveform into the secondary sound waveform memory 10 before generating the tone signal. For example, immediately after the power is turned on, the load loads a plurality of secondary sound waveform waveform data stored in a storage medium such as a magnetic disk, an optical disk, or an IC card in advance.
This can be realized by transferring to the next sound wave memory 10.

The secondary sound waveform memory 10 stores waveform data of a plurality of secondary sound waveforms prepared for a plurality of tone ranges of a plurality of tone colors. As described above, the waveform data of each secondary sound waveform is created based on the sound generated by the keystroke with the strongest touch. The "plurality of tones" include, for example, tones such as piano tone, harpsichord tone, flute tone, and trumpet tone. The tone color data is used to select one tone color from the plurality of tone colors. As the "tone range" of each tone color, for example, a tone range separated by one note, two to several notes, one octave, or the like is used. Range data is used to select one range from a plurality of ranges. As described above, the tone color data and the tone range data are supplied from the CPU (not shown) to the tone signal generator.

The tone color data and tone range data are used as address data for the secondary sound waveform memory 10. Then, the sampling data included in the waveform data of one secondary sound waveform that is addressed by the tone color data and the tone range data is sequentially read from the secondary sound waveform memory 10 and supplied to the multiplier 13.

The noise waveform memory 11 is RO in this embodiment.
It is assumed to be composed of M. The noise waveform memory 11 may be composed of RAM. In this case, it is necessary to load the waveform data of the noise waveform into the noise waveform memory 11 before generating the tone signal. The loading can be realized, for example, by transferring waveform data of a plurality of noise waveforms stored in advance in a storage medium such as a magnetic disk, an optical disk, an IC card to the noise waveform memory 11 immediately after the power is turned on.

In the noise waveform memory 11, waveform data of a plurality of noise waveforms corresponding to the waveform data of a plurality of secondary sound waveforms stored in the secondary sound waveform memory 10, that is, a plurality of tone colors. Waveform data of a plurality of noise waveforms created for each tone range is stored. The waveform data of each noise waveform is created by the method described above.
The tone color data and range data are stored in the noise waveform memory 11
Is used as address data for. Then, the sampling data included in the waveform data of one noise waveform addressed by the tone color data and the range data is sequentially read from the noise waveform memory 11,
It is supplied to the multiplier 14.

The first weighting means of the present invention comprises a weighting coefficient generator 12a and a multiplier 13. The weighting factor generator 12a generates a weighting factor C1 for the waveform data of the secondary sound waveform.

The weighting coefficient C1 generated by the weighting coefficient generator 12a is data that changes according to the touch data, as shown by the first characteristic for the secondary sound in FIG. 2, for example. Although FIG. 2 shows only the weighting coefficient C1 for one secondary sound waveform waveform data, in reality, the weighting coefficient C1 corresponds to each of a plurality of secondary sound waveform data. Is generated. Which secondary sound waveform data is used to generate the weighting coefficient C1 is specified by the tone color data and the tone range data.

The relationship between the touch data input to the weighting coefficient generator 12a and the weighting coefficient C1 output from the weighting coefficient generator 12a is shown in, for example, the first characteristic for the secondary sound in FIG. As described above, the increase rate of the touch data gradually increases from the initial value α. Further, it may be configured so as to have a characteristic of linearly increasing from the initial value α and other various characteristics.

The weighting coefficient generator 12a is, for example, RO
M. The weighting factor C1 corresponding to each value of the touch data is stored in advance in the ROM, for example, in a table format. Further, the weighting coefficient generator 12a may be composed of a RAM. In this case, as in the case where the secondary sound waveform memory 10 is composed of a RAM, the weighting coefficient C1 is stored in the RAM before the tone signal is generated.
Need to be loaded on. For loading, for example, immediately after the power is turned on, the weighting coefficient C1 stored in advance in a storage medium such as a magnetic disk, an optical disk, or an IC card is read as R.
It can be realized by transferring to AM.

The weighting factor generator 12a can also be constituted by a function generator, as described above. In this case, a function that inputs touch data and outputs a weighting coefficient C1 according to the first characteristic for secondary sound as shown in FIG. 2 is used. This weight coefficient generator 12a
The weighting coefficient C1 generated in 1 is supplied to the multiplier 13.

The multiplier 13 receives the waveform data of the secondary sound waveform from the secondary sound waveform memory 10 and the weighting coefficient generator 12 described above.
Multiply with the weighting coefficient C1 from a. By this multiplication,
The ratio of the volume of the secondary sound to the volume of the entire piano sound to be generated can be controlled. The output of the multiplier 13 is supplied to the filter 15.

The second weighting means of the present invention comprises a weighting coefficient generator 12b and a multiplier 14. The weighting factor generator 12b generates a weighting factor C2 for the waveform data of the noise waveform.

The weighting coefficient C2 generated by the weighting coefficient generator 12b is data that changes according to the touch data, as shown in the second characteristic for noise (striking sound) of FIG. 2, for example. Although only the weighting coefficient C2 for the waveform data of one noise waveform is shown in FIG. 2, the weighting coefficient C2 is actually generated corresponding to each of the waveform data of a plurality of noise waveforms. Which noise waveform is used to generate the weighting coefficient C2 is determined by the tone color data and the tone range data.

The relationship between the touch data input to the weight coefficient generator 12b and the weight coefficient C2 output from the weight coefficient generator 12b is, for example, the second characteristic for noise (striking sound) in FIG. As shown in (3), the increase rate of the touch data is gradually increased from the initial value β. Further, it may be configured to have a characteristic that it linearly increases from the initial value β and other various characteristics.

The weighting coefficient generator 12b is, for example, RO
M. The weighting coefficient C2 corresponding to each value of the touch data is stored in advance in the ROM, for example, in a table format. Further, the weighting coefficient generator 12b may be composed of a RAM. In this case, as in the case where the secondary sound wave type memory 10 is composed of a RAM, the weighting coefficient C2 is stored in the RAM before the tone signal is generated.
Need to be loaded on. For loading, for example, immediately after the power is turned on, the weighting coefficient C2 stored in advance in a storage medium such as a magnetic disk, an optical disk or an IC card is read as R.
It can be realized by transferring to AM.

The weighting coefficient generator 12b can also be constituted by a function generator, as described above. In this case, the touch data is input and, for example, the weighting coefficient C2 according to the second characteristic for noise (striking sound) as shown in FIG.
A function that outputs is used. The weighting coefficient C2 generated by the weighting coefficient generator 12b is supplied to the multiplier 14.

The multiplier 14 multiplies the waveform data of the noise waveform from the noise waveform memory 11 and the weighting coefficient C2 from the weighting coefficient generator 12b. By this multiplication, the ratio of the volume of noise to the volume of the entire piano sound to be generated can be controlled. The output of the multiplier 14 is supplied to the filter 16.

The weighting coefficient generator 12a and the weighting coefficient generator 12b can be constructed by sharing the same table. In this case, the weighting coefficient C1 and the weighting coefficient C2 for each touch data value may be stored in one table.

The first filter means of the present invention comprises the filter 15 and the parameter generator 19.
As the filter 15, for example, a low pass filter is used. The filter 15 filters the waveform data of the secondary sound waveform from the multiplier 13 according to the filter parameter from the parameter generator 19 and outputs it.

The parameter generator 19 generates a filter parameter for the secondary sound waveform data. The waveform data of the secondary sound waveform to be filtered is specified by the tone color data and the range data. The parameter generator 19 generates a filter parameter for changing the cutoff frequency according to the touch data, for example. That is, the parameter generator 19 generates a parameter that increases the cutoff frequency when the touch data becomes large, and conversely decreases the cutoff frequency when the touch data becomes small. With this configuration, the number of overtones included in the musical tone is increased in the case of a hard hit, and conversely, the number of overtones included in the musical tone is decreased in the case of a weak hit, so that a tone signal having a sound quality closer to that of a natural musical instrument can be generated. .

The filter 15 and the parameter generator 19 are not always essential in the present invention. The filter 15 and the parameter generator 19 may be removed in the tone signal generator that does not need to control the number of overtones according to the touch strength. The output of the filter 15 is supplied to the adder 17.

The second filter means of the present invention comprises the filter 16 and the parameter generator 20.
As the filter 16, for example, a low pass filter is used. The filter 16 filters the waveform data of the noise waveform from the multiplier 14 according to the filter parameter from the parameter generator 20 and outputs it.

The parameter generator 20 generates a filter parameter for the waveform data of the noise waveform. Waveform data of a noise waveform to be filtered is specified by tone color data and range data. The parameter generator 20 generates, for example, a parameter that variously changes the sound quality of noise according to touch data. As the value of this filter parameter, an appropriate value can be selected by, for example, cut and try. As a result, a musical tone signal having a sound quality close to that of a natural musical instrument can be generated.

The filter 16 and the parameter generator 20 are not always essential in the present invention. For example, in a tone signal generator that does not need to change the sound quality of noise, these filter 16 and parameter generator 2 are used.
It is also possible to remove 0. The output of the filter 16 is supplied to the adder 17.

The mixing means of the present invention comprises an adder 17. The adder 17 adds the waveform data from the filter 15 and the waveform data from the filter 16. By this addition, the waveform data of the waveform in which the secondary sound waveform and the noise waveform are mixed at a ratio according to the touch strength is created. The output of the adder 17 is supplied to the multiplier 18.

The envelope generator 21 generates envelope data for generating an envelope having a shape peculiar to a tone color and a tone range and having an amplitude corresponding to the touch strength. The output of the envelope generator 21 is supplied to the multiplier 18.

The multiplier 18 multiplies the waveform data from the adder 17 and the envelope data from the envelope generator 21. By this multiplication, a musical tone signal is generated in which waveform data peculiar to the timbre and range is also added with an envelope peculiar to the timbre and range. This tone signal is output to the outside as the output of the tone signal generator.

Next, the operation when the weighting factors generators 12a and 12b generate the weighting factors C1 and C2 according to the first and second characteristics as shown in FIG. 2 will be described. For example, when tone color data and tone range data designating a piano sound in a predetermined key range is supplied from the CPU (not shown) and touch data representing the strongest touch is supplied, maximum values are respectively generated as weighting factors C1 and C2. It The ratio between the weighting factors C1 and C2 is "1: 1". When a musical sound is generated based on the musical sound signal generated in this case, the piano sound (original sound) tapped with the strongest touch is reproduced, and the volume of the entire piano sound becomes maximum.

On the other hand, when tone color data and tone range data designating a piano sound in a predetermined key range is supplied and touch data representing the weakest touch is supplied, minimum values are respectively generated as weighting factors C1 and C2. It The ratio between the weighting factors C1 and C2 in this case is, for example, “1: 2”. That is, in the case of the weakest touch, the ratio of the volume of the striking sound to the volume of the entire piano sound that is produced is the maximum, and
The volume ratio of the next note is minimized, and the volume of the entire piano sound is minimized. Therefore, when a musical tone is generated based on the musical tone signal generated in this case, a piano tone having a large percussion tone component can be obtained although the volume of the entire piano tone is small, as compared with the case of the strongest touch.

Further, as the touch data gradually increases from the weakest touch, the volume of the entire piano sound gradually increases, and the ratio of the impact sound to the volume of the entire piano sound gradually decreases. By the control as described above, it is possible to generate a sound close to an actual piano sound that appropriately includes a hitting sound component for each touch strength.

In the above embodiments, the case where the piano sound is created electrically has been mainly described, but the present invention is not limited to the piano sound, and keyboard instruments such as organs and harpsichords, flutes, trumpets, etc. Wind instrument,
The same applies to the case of electrically creating sounds of various other natural musical instruments.

[0071]

As described in detail above, according to the present invention,
It is possible to provide a musical tone signal generating device and a musical tone signal generating method for generating a musical tone signal capable of reproducing a natural musical instrument sound including an appropriate noise component.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a musical tone signal generating apparatus of the present invention.

FIG. 2 is a diagram for explaining weighting factors used in the embodiment of the present invention.

FIG. 3 is a diagram showing an example of an original sound waveform used in an embodiment of the present invention.

FIG. 4 is a diagram showing an example of a secondary sound waveform used in an embodiment of the present invention.

FIG. 5 is a diagram showing an example of a noise waveform used in the embodiment of the present invention.

[Explanation of symbols]

 10 Second-order sound waveform memory 11 Noise waveform memory 12 Weighting coefficient generator 13, 14, 18 Multiplier 15, 16 Filter 17 Adder 19, 20 Parameter generator 21 Envelope generator

Claims (8)

[Claims] 1. A secondary sound waveform memory for storing waveform data of a secondary sound waveform obtained by removing noise components from a waveform of a natural musical instrument sound, and a difference between the waveform of the natural musical instrument sound and the secondary sound waveform. A noise waveform memory that stores waveform data of a noise waveform, and a first weighting that performs first weighting on the waveform data read from the secondary sound waveform memory according to the first characteristic according to the attack intensity. Means, second weighting means for performing a second weighting on the waveform data read from the noise waveform memory according to a second characteristic in accordance with the attack strength, and a waveform from the first weighting means. A musical tone signal generating apparatus comprising: mixing means for mixing the data and the waveform data from the second weighting means, and generating a musical tone signal based on the output of the mixing means. 2. The musical tone signal generator according to claim 1, further comprising first filter means for filtering the output of said first weighting means and supplying it to mixing means. 3. The musical tone signal generator according to claim 1, further comprising second filter means for filtering the output of said second weighting means and supplying it to mixing means. . 4. The natural instrument sound waveform is a piano sound waveform, the noise component is a percussion sound component included in the piano sound, and the attack intensity is a touch intensity. The musical sound signal generator according to any one of claims 1 to 3. 5. (A) Waveform data of a secondary sound waveform is prepared by removing noise components from the waveform of the natural musical instrument sound, and (B) is composed of the difference between the waveform of the natural musical instrument sound and the secondary sound waveform. Waveform data of a noise waveform is created in advance, and according to (C) pronunciation command,
The secondary sound waveform waveform data is first weighted according to the first characteristic according to the attack strength, and (D) the waveform data of the noise waveform is subjected to the second characteristic according to the attack strength. Second weighting according to (E) mixing the waveform data of the second weighted secondary sound waveform with the waveform data of the second weighted noise waveform, and mixing A tone signal generating method characterized in that a tone signal is generated based on the waveform data. 6. The musical tone signal generating method according to claim 5, wherein said step (E) further comprises a step of filtering the waveform data of the first weighted secondary sound waveform. 7. The tone signal generating method according to claim 5, wherein said step (E) further comprises a step of filtering waveform data of the second weighted noise waveform. . 8. The waveform of the natural musical instrument sound is a waveform of a piano sound, the noise component is a percussion sound component included in the piano sound, and the attack intensity is a touch intensity. The musical tone signal generating method according to any one of claims 5 to 7.