JPH067337B2 - Music signal generator - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone signal generator used in electronic musical instruments, game machines, etc., and particularly to a musical tone signal generating a musical tone signal whose waveform changes with time. Regarding the generator.

(Prior Art) Conventionally, an apparatus of this type is disclosed in, for example, Japanese Patent Laid-Open No. 61-10729.
As disclosed in Japanese Patent Publication No. 8, a set of musical tone waveform data generating means for sequentially switching and outputting a plurality of sets of musical tone waveform data each consisting of a plurality of sampling data and representing different musical tone waveforms, and one set of musical tone waveform data. Storage means capable of storing a plurality of sampling data
Each sampling data stored in the storage means is corrected based on each sampling data constituting one set of musical tone waveform data output from the musical tone waveform data generating means, and each sampling data stored in the storage means is generated. The musical tone signal to be reproduced is sequentially and repeatedly output at a rate proportional to the frequency of the musical tone, and a musical tone signal whose waveform changes with time is generated at a desired frequency.

(Problems to be Solved by the Invention) However, in the above-mentioned conventional apparatus, the effect of simultaneous generation of a plurality of musical tone signals whose frequencies or phases are slightly shifted,
For example, the ensemble effect could not be obtained. That is, in the same apparatus, if the reading rate of each sampling data stored in the storage means is changed, it is possible to obtain an effect, for example, a vibrato effect by one series of tone signals whose frequency or phase is slightly changed. Even if it was possible, it was not possible to obtain the ensemble effect in which multiple musical tones were pronounced at the same time. Further, in order to obtain the ensemble effect by using the conventional device, it is possible to prepare a plurality of series of storage means and the like, but such a method has a problem that the cost of the entire device becomes high.

The present invention has been devised in view of the above problems, and an object of the present invention is to provide a musical tone signal generating device for generating a musical tone signal whose waveform changes with the passage of time, and a musical tone signal for realizing the ensemble effect at low cost. To provide a device.

(Means for Solving Problems) In order to solve the above problems and achieve the object of the present invention,
The structural feature of the present invention is to configure one set of musical tone waveform data and a musical tone waveform data generating means for sequentially switching and outputting a plurality of sets of musical tone waveform data each consisting of a plurality of sampling data and representing different musical tone waveforms. Storage means for storing a plurality of sampling data to be stored in different positions, and each sampling data stored in the storage means is output from the tone waveform generating means 1
Correcting means for correcting a set of musical tone waveform data based on the plurality of sampling data, and reading rate for controlling the reading rate of each sampling data stored in the storage means in proportion to the frequency of the musical tone to be generated. A plurality of sampling data stored in the storage means are sequentially and repeatedly output at a rate controlled by the read rate control means,
In a musical tone signal generating device for generating a musical tone signal corresponding to musical tone waveform data composed of the same sampling data, a pitch modulation signal generating means for independently generating a plurality of pitch modulation signals and the memory means are stored. There is provided a read position changing means for changing the read position of each sampling data independently according to the plurality of pitch modulation signals and outputting a plurality of sets of musical tone waveform data corresponding to the pitch modulation signals. .

(Operation of the Invention) In the present invention configured as described above, the musical tone waveform data generating means is composed of a plurality of sampling data, and a plurality of sets of musical tone waveform data representing different musical tone waveforms are sequentially switched and output according to the passage of time to be corrected. The means corrects each sampling data stored in the storage means on the basis of the plurality of sampling data constituting the musical tone waveform data, so that the musical tone waveform data composed of the plurality of sampling data stored in the storage means is elapsed with time. Change according to. The read rate of each sampling data forming the changing tone waveform data is controlled by the read rate control means, and the read position is changed to a plurality of pitch modulation signals from the pitch modulation signal generating means by the read position changing means. Based on each independently controlled,
It is independently output as a plurality of musical tone waveform data. Since this change of the read position means changing the output timing of each sampling data, a plurality of tone signals whose waveforms change with the passage of time have different phases or frequencies depending on the pitch modulation signal. , Will be output at the same time.

(Effects of the Invention) As can be understood from the above description of the operation, according to the present invention, even if the storage unit or the like is not provided for a plurality of series, only the read position of each sampling data in the storage unit is changed, and each time the time passes. Since a plurality of tone signals that change and have different frequencies or phases are output, an ensemble effect that a plurality of musical instruments are simultaneously producing sound can be obtained at low cost.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing an electronic musical instrument to which the musical tone signal generating apparatus according to the present invention is applied.

This electronic musical instrument has a key switch circuit 11 and a tone color selection switch circuit 12. The key switch circuit 11 is composed of a plurality of key switches corresponding to the respective keys of the keyboard, and the switches are opened and closed according to the pressing and releasing of the keys. A key press detection circuit 13 is connected to the key switch circuit 11, and the detection circuit 13 detects the opening and closing of each key switch in the key switch circuit 11 to detect the pressing and releasing of each key on the keyboard. The key code KC representing the key pressed on the keyboard, and a high level "1" (hereinafter simply "1") when the key is pressed, and a low level "0" (hereinafter simply "1") when the key is released. A key-on signal KON that is "0" is output. The tone color selection switch circuit 12 is composed of a plurality of tone color selection switches corresponding to the respective operators of the tone color selection operator group provided on the front panel of the electronic musical instrument, and the switch circuit 12 includes the tone color selection operator group. A tone color selection signal TSW representing the tone color selected by is output.

These key press detection circuit 13 and tone color selection switch circuit 1
An address generator 20 is connected to 2. The address generator 20 controls the first and second address signals AD for controlling the reading of the waveform data stored in the waveform memory 31.
1, AD2, and has a note clock generator 21, as shown in detail in FIG. The note clock generator 21 generates a pitch frequency of 3 of the pressed key based on the key code KC supplied from the key pressing detection circuit 13.
A note clock signal φ _3n having a frequency m times (3m is a value equal to 3 times the number of sampling data of a tone waveform for one cycle) is output. This note clock signal φ _3n is output from the address generator 20 and the same generator 2
It is supplied to the 1/3 frequency divider 22 within 0. The 1/3 frequency divider 22 _{divides the} note clock signal φ _3n by 1/3 and multiplies the pitch frequency of the key pressed by m (m is a value equal to the number of sampling data of one tone waveform). A divided note clock signal φ _n having a frequency is formed, and the clock signal φ _n is output from the address generator 20 and supplied to the counter 23 in the generator 20. The counter 23 counts the frequency-divided note clock signal φ _n to obtain “0” to “m−
The count value that repeatedly changes over "1" is output as the first address signal AD1. The reset terminal R of the counter 23 is supplied with a key-on pulse signal KONP obtained by differentiating the key-on signal KON by a differentiating circuit 24, and the counter 23 outputs the key-on pulse signal KON.
It is designed to be reset when a key is pressed in response to the arrival of ONP.

In addition, the count value of the counter 23 from "m-1" to "0"
The carry signal CA generated from the counter 23 every time it changes to the counter 25 is input to the counter 25, and the counter 25 counts the carry signal CA for each cycle of the first address signal AD1 (generated tone The count value is incremented by "1" every 1 cycle).
The reset terminal R of the counter 25 is supplied with a key-on pulse signal KONP and a coincidence signal EQ from a comparator 26 described later via an OR circuit OR, and the counter 25 receives the key-on pulse signal KONP or the coincidence signal EQ. Depending on the key press or the coincidence signal EQ from the comparator 26
When it occurs, it will be reset. The count value output from the counter 25 is supplied to one input of the comparator 26, and the repeat count value is supplied from the repeat count memory 27 to the other input of the comparator 26. The repeat count memory 27 stores a count value for repeatedly outputting the same waveform stored in the waveform memory 31, for each waveform, and the tone color selection signal T from the tone color selection switch circuit 12 is stored.
The repeat count value is output according to the count value (second address signal AD2) from the SW and the counter 28. Accordingly, the comparator 26 outputs the coincidence signal EQ and resets the counter 25 via the OR circuit OR at the time when the count value from the counter 25 matches the repeat count value from the repeat count memory 27, and at the same time, Same match signal E
Q is a counter 2 via one input of an AND circuit AND
It is designed to supply to 8. The counter 28 counts the supplied coincidence signal EQ and outputs a count value that is incremented by "1" as the second address signal AD2. The other input of the AND circuit AND is supplied with the output of the NAND circuit NAND which detects that all the bits of the second address signal AD2 have become "1", and all the bits of the second address signal AD2 are " When the count value of 1 "(in this embodiment, the count value of the counter 28 becomes" 7 "), the coincidence signal EQ is not supplied to the counter 28 via the AND circuit AND. This makes the counter 2
Reference numeral 8 increases the count value by "1" in response to the arrival of the coincidence signal EQ, and outputs the second address signal AD2 which changes from "0" to "7", for example. Also counter 2
The key-on pulse signal KONP is supplied to the reset terminal R of the counter 8, and the counter 28 is reset when a key is pressed.

The waveform memory 31 corresponds to n timbres and is designated by the timbre selection signal TSW from the timbre selection switch circuit 12 as waveform data memories 31-1, 31-2 ... 31-n.
Have. Each waveform data memory 31-1, 31-2 ...
31-n are a plurality of areas (8 areas E ₀ and E _{1 in} this embodiment, respectively) designated by the second address signal AD2.
·· E ₇₎ is divided into tape each area E _0, E ₁ ·
..E ₇ stores m pieces of sampling data each representing a waveform of one cycle of a musical sound. In addition, each area E ₀ ,
The series of sampling data stored in E ₁ ... E ₇ show different waveforms, but the phases of the waveforms match each other.

An interpolation circuit 40 is connected to the waveform memory 31, and the circuit 40 includes a subtractor 41, a multiplier 42, an adder 43, and a shift register 44. The subtractor 41 subtracts the sampling data supplied from the final stage of the shift register 44 from the sampling data supplied from the waveform memory 31, and outputs difference data resulting from the subtraction to the multiplier 42. The multiplier 42 multiplies the difference data by the gain coefficient g and supplies the result to one input of the adder 43. The gain coefficient g is supplied from the gain coefficient memory 51. The memory 51 corresponds to n timbres and is designated by the timbre selection signal TSW from the timbre selection switch circuit 12 as a gain coefficient data memory 51-1. , 51-2 ・
.. has 51-n. Each gain coefficient data memory 51-
1, 51-2 ... 51-n are second address signals A, respectively.
A plurality of gain coefficient data (8 data g ₀ , g ₁ ... G _{7 in} this embodiment) addressed by D2 are stored. The other input of the adder 43 is supplied with the sampling data from the final stage of the shift register 44, and the adder 43 adds the sampling data and the input data from the multiplier 42 to obtain the shift register 4
4 to the first stage. The shift register 44 has m stages for storing m pieces of sampling data corresponding to one period of the tone waveform, and the sampling data stored in each stage is a divided note clock signal φ _n.
Are sequentially shifted by and are reset by the key-on pulse signal KONP.

The shift register 44 can take out each sampling data stored in each stage in parallel, and the output of each stage is connected to the selector 52. The selector 52 selects and outputs sampling data from each stage of the shift register 44 according to the selection signal SEL from the multiplexer 53. Macchiplexa 5
3 is the first to third pitch modulation signal generators 54 ₁ , 54 ₂ ,
54 _{3 is} a pitch modulation signal from the shift register 4
The signal instructing the output of each stage of No. 4 is multiplexed with the note clock signal φ _3n supplied from the note clock generator 21 and output. Each pitch modulation signal generator 5
4 ₁ , 54 ₂ and 54 ₃ are digital type low frequency oscillators for generating a periodic waveform signal LFO as shown in FIG. 3A,
And a key-on signal KO as shown in FIGS. 3B and 3C.
A digital form waveform forming circuit for generating a pitch bend waveform signal PBD and a glide waveform signal GLD obtained by differentiating and integrating N are respectively provided, and these waveform signals LFO and PFO are provided.
One or more of BD and GLO are output at the same time according to the selection of performance sound. The waveform signal LFO is set so that the frequency or the phase is different for each of the first to third pitch modulation signal generators 54 ₁ , 54 ₂ and 54 ₃ (LFO ₁ ,
LFO ₂ , LFO ₃ ) and the waveform signals PBD, GLD are set so that the waveform shapes are different for each of the generators 54 ₁ , 54 ₂ , 54 ₃ (PBD ₁ , PBD ₂ , PBD ₃ and GL).
D ₁ , GLD ₂ , GLD ₃ ).

A demultiplexer 55 is connected to the selector 52. The demultiplexer 55 is the note clock generator 21.
Controlled by the note clock signal φ _3n supplied from the selector 52, the sampling data output from the selector 52 in synchronization with the signal φ _3n is demultiplexed into three series of sampling data, and each sampling data is frequency-divided note clock signal φ Each of the first to third digital filters 5 is synchronized with _n.
It outputs to 6 ₁ , 56 ₂ , and 56 ₃ . The first to third digital filter 56 _1, 56 _2, 56 ₃ adds adds the desired frequency characteristic in each sampling data supplied 57
Output to. The adder 57 outputs the sum of the respective sampling data from the first through third digital filter 56 _1, 56 _2, 56 _3.

A multiplier 58 is connected to the adder 57, and the multiplier 58 multiplies the sampling data from the adder 57 and the envelope waveform data and outputs the result. This envelope waveform data is supplied from an envelope generator 59, and the generator 59 forms and outputs envelope waveform data representing an envelope waveform of a musical tone in response to a key-on signal KON from the key depression detection circuit 13. The envelope waveform is controlled by the tone color selection signal TSW from the tone color selection switch circuit 12 and is formed into a shape suitable for the tone color of a musical tone.

A digital-analog converter 61 is connected to the multiplier 58, and the converter 61 converts the digital signal from the multiplier 58 into an analog signal and outputs it to the sound system 62. The sound system 62 is composed of an amplifier, a speaker, etc., and has a digital analog converter 61.
Generates a musical sound according to the analog signal supplied from.

The operation of the embodiment configured as described above will be described. When any key is pressed on the keyboard and the key switch corresponding to the pressed key in the key switch circuit 11 is closed,
The key-depression detection circuit 13 detects this key-depression and supplies the key code KC and the key-on signal KON representing the depressed key to the address generator 20. In the address generator 20, the differentiation circuit 24 raises and differentiates the key-on signal KON to generate the key-on pulse signal KONP,
With this key-on pulse signal KONP, the counter 23,
25 and 28 are reset respectively. The key-on pulse signal KONP is also supplied to the shift register 44 of the interpolation circuit 40, and the shift register 44 is reset.

After these resets, the counter 23 in the address generator 20 counts the divided note clock signal φ _n that determines the sampling time, and outputs the first address signal AD1 that repeatedly changes from “0” to “m−1”. Waveform memory 3
Supply to 1. A tone color selection signal TSW representing a tone color selected by the player is supplied from the tone color selection switch circuit 12 to the waveform memory 31, and the second address signal AD2 set to "0" by the reset.
Is supplied from the counter 28, and the memory 31 stores the waveform data memory 31-i (i is 1 to n) corresponding to the selected tone color.
M (one cycle) of the sampling data stored in the first area E ₀ in any of the integers) is sequentially read to one input of the subtracter 41 of the interpolation circuit 40 according to the first address signal AD1. Output repeatedly. The subtractor 41 subtracts the sampling data from the final stage of the shift register 44 from the sampling data, and outputs the subtracted difference data to the multiplier 42. The multiplier 42 includes a gain coefficient data memory 51-i (i is 1 to n) in the gain coefficient memory 51 specified by the tone color selection signal TSW.
2) which is set to "0"
Gain coefficient data g read by the address signal AD2
₀ is supplied, and the multiplier 42 multiplies the supplied data by the gain coefficient data g ₀ and supplies it to one input of the adder 43. Since the adder 43 adds the supplied data and the sampling data from the final stage of the shift register 44 and supplies it to the first stage of the shift register 44, the shift register 44 stores the sampling data stored in the register 44. The data is circularly stored while being corrected according to the difference data from the subtracter 41 via the multiplier 42. As a result, the sampling data in the shift register 44 which is initially reset gradually approaches the sampling data stored in the first area E ₀ in the waveform data memory 31-i.

In this way, the sampling data cyclically stored in the shift register 44 is supplied to the selector 52 for each stage of the register 44, and the selector 52 selectively outputs the supplied sampling data according to the selection signal SEL. The selection signal SEL is the waveform signal LF from the _{first to} third pitch modulation signal generators 54 ₁ , 54 ₂ , 54 _3.
O ₁ , LFO ₂ , LFO ₃ (PBD ₁ , PBD ₂ , PBD ₃ or GLO ₁ , GLO ₂ , GLO ₃ ) multiplexer 53
Since it is multiplexed with the note clock signal φ _{3n, the} signal output from the selector 52 is a pitch modulation signal generator during one sampling time corresponding to the cycle of the divided note clock signal φ _n . 54 ₁ , 54 ₂ , 5
It becomes a signal obtained by multiplexing three kinds of sampling data whose output positions are defined by the respective waveform signals from 4 ₃ . The multiplexed signal is Demaruchi by Demaruchi circuit 55, the three types of sampling data is supplied to the divider note clock signal φ first to third digital Fi motor 56 in synchronism with the _{n 1,} 56 _2, 56 _3, the Filter 56 ₁ ,
Desired frequency characteristics are given at 56 ₂ and 56 ₃ and supplied to the adder 57 as sampling data strings of three systems. In this case, the three series of sampling data strings correspond to the sampling data in which the read positions of the shift register 44 are changed by the waveform signals from the pitch modulation signal generators 54 ₁ , 54 ₂ , 54 _3. The tone waveforms represented by the three series of sampling data sequences have different phases or frequencies.

In this way, the three series of sampling data strings supplied to the adder 57 are summed by the adder 57 and supplied to the multiplier 58, and the multiplier 58 in the multiplier 58.
It is multiplied by the envelope waveform data supplied from the digital signal converter 9 and supplied to the digital analog converter 61. The sampling data multiplied by the envelope waveform data is converted into an analog signal by the digital analog converter 61, and this analog signal is supplied to the sound system 62, and the system 62 generates a musical sound corresponding to this analog signal. . As a result, the generated tone becomes a combination of tone signals having different phases or frequencies.

In addition, after the lapse of time from the key depression, the counter value of the counter 25 becomes the tone color selection signal TSW and the second address signal A.
When equal to the repeat count value read from the repeat count memory 27 addressed by D2, the comparator 26
Outputs the coincidence signal EQ. The coincidence signal EQ resets the counter 25 to start counting from "0" again, and the counter 28 increases the count value by "1" to change the second address signal AD2 to "1".
By changing the second address signal AD2, the waveform data memory 31-i repeatedly outputs the sampling data stored in the second area E ₁ . At the same time, the gain coefficient data memory 51-i outputs the gain coefficient data g ₁ . Also in this case, the interpolation circuit 40 outputs the sampling data stored in the shift register 44 from the second area E ₁ of the waveform data memory 31-i, as in the case where the second address signal AD2 is “0”. It gradually approaches the read sampling data.

Similar to the above case, such sampling data also produces three series of tone signals having different phases or frequencies by the actions of the selector 52, the multiplexer 53, the pitch modulation signal generators 54 ₁ , 54 ₂ , 54 ₃ and the demultiplexer 55. Since the three series of sampling data strings respectively represented are read out and output, the sound system 62 obtains musical tones in which the three series of musical tone signals are synthesized at the same time as the waveform gradually changes.

Further, when time passes and the count value (second address signal AD2) of the counter 28 increases to “2”, “3”, ... And reaches “7”, the NAND circuit NAND outputs “0”. Therefore, the second address signal A by the counter 28
The update of D2 stops. As a result, the waveform of the generated musical tone does not change thereafter.

As can be understood from the above description of the operation, according to the above-described embodiment, it is not necessary to provide the interpolation circuit 40 for a plurality of series, and the musical tone signal for generating the musical tone signal whose waveform gradually changes with the lapse of time is formed with a simple configuration. In the generator, it is possible to obtain a musical tone in which a plurality of musical tone signals having different phases or frequencies are combined, that is, a musical tone with an ensemble effect. As a result, rich musical tones can be produced at low cost.

The present invention can be implemented even if the above embodiment is modified as follows.

(1) In the above embodiment, the waveform memory 31 is used as a means for generating sampling data,
As means for generating this waveform data, a circuit for generating sampling data representing a desired tone waveform by a method such as calculation or oscillation may be used.

(2) In the above embodiment, the sampling data from all the stages of the shift register 44 is guided to the selector 52. However, the pitch modulation signal generators 54 ₁ , 5
If the depth of the phase or frequency modulation based on the waveform signals from 4 ₂ and 54 ₃ may be shallow, the sampling data from some stages of the register 44 are guided to the selector 52. Good.

(3) In the above embodiment, the pitch modulation signal generator 5
The waveform signals from 4 ₁ , 54 ₂ and 54 ₃ are time-division multiplexed by the multiplexer 53, but the same generators 54 ₁ , 54 ₂ ,
If a single pitch modulation signal generator that forms and outputs waveform signals for each series in time division and serially is used instead of 54 ₃ , the multiplexer 53 can be omitted. In the above embodiment has been to use a digital filter 56 _1, 56 _2, 56 ₃ spatially separated for each series, these digital filters 5
If 6 _1, 56 _2, 56 to use a digital filter for processing a time division ₃ of the function, the demultiplexer 5
5 may be omitted. In that case, instead of the adder 57, it is preferable to use an accumulator that accumulates and outputs the sampling data of three series supplied from the digital filter in time division for each three series. Further, in the above embodiment, the three series of sampling data are once time-division multiplexed, and then the time-division multiplexing of the same data is released. However, three selectors are provided for each selector, and each selector is pitch-modulated. The signals may be independently controlled by the signal generators 54 ₁ , 54 ₂ , and 54 ₃ , and the outputs of the selectors may be independently guided to the digital filters 56 ₁ , 56 ₂ , and 56 ₃ . In this case, the multiplexer 53 and the demultiplexer 55 are unnecessary.

(4) Further, according to the above-mentioned embodiment, the shift register 44 is used as a means for storing the sampling data which changes with the passage of time. However, as a means for storing the sampling data, the storage position of the data is changed by the address signal. A writable memory (RAM) in which writing and reading of data are specified and controlled may be used. In this case, in the operation of reading out and outputting each sampling data from the RAM, at the same time as the address counter for generating the add-sustain signal is advanced by the divided note clock signal φ _n , the pitch modulation signal generator
The waveform signal values from 4 ₁ , 54 ₂ , and 54 ₃ are added to and subtracted from the address signal value in synchronization with the note clock signal φ _n to change and control the address signal value, and the RAM read position is changed for each series. It is good to do so.

(5) In the above embodiment, the digital filters 56 ₁ ,
56 _2, 56 after summing up the 3 series of sampled data from ₃ by an adder 57, a multiplier 58 and the Enburopu generator 59 has been adapted to impart each series common envelope, the envelope in each series each independently May be given. In this case, it provided the multiplier 58 and the envelope generator 59 is independently for each series, each digital filter 56 _1, 56 _2, 56 respectively may be to multiply the envelope waveform data sampling data from _three. In addition, the sampling data to which the envelopes are independently added for each series is added again and guided to the digital-to-analog converter 61 so that the sound system 62 can generate a musical sound by synthesizing the musical sound signals of three series. Alternatively, the digital-analog converter 61 and the sound system 62 may be provided in each of three series so that a musical tone can be independently generated for each series.

(6) In the above embodiment, three series of musical tones are generated at the same time, but two or four or more series of musical tones may be generated simultaneously by selecting the number of separations for each series.

[Brief description of drawings]

FIG. 1 is a block diagram of an electronic musical instrument to which a tone signal generator according to an embodiment of the present invention is applied, FIG. 2 is a diagram showing a detailed example of the address generator of FIG. 1, and FIGS. Three
FIG. 6C is a diagram for explaining the waveform signal output from the pitch modulation signal generator of FIG. Explanation of reference numerals 20 ... Address generator, 21 ... Note clock generator, 22 ... 1/3 frequency divider, 23, 25, 28 ... Counter, 31 ... Waveform memory, 40 ... Interpolation circuit, 41 ......
Subtractor, 42 ... Multiplier, 43 ... Adder, 44 ... Shift register, 51 ... Gain coefficient memory, 52 ... Selector, 54 ₁ , 54 ₂ , 54 ₃ ... Pitch modulation signal generator.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI technical display location G10K 15/04 302 G 7227-5H

Claims (1)

[Claims] 1. A musical tone waveform data generation means for sequentially switching and outputting a plurality of musical tone waveform data each of which is composed of a plurality of sampling data and represents different musical tone waveforms, and a plurality of sampling data constituting one set of musical tone waveform data. Are stored in different positions, and each sampling data stored in the storage means is corrected based on the plurality of sampling data constituting one set of musical tone waveform data output from the musical tone waveform data generating means. Correction means for controlling the read rate of each sampling data stored in the storage means in proportion to the frequency of the musical sound to be generated, and a plurality of read rate control means stored in the storage means. Sampling data is sequentially repeated at the rate controlled by the reading rate control means. In a tone signal generator that generates a tone signal corresponding to tone waveform data composed of the same sampling data by outputting, a pitch modulation signal generating means for independently generating a plurality of pitch modulation signals and the storage means. Read position changing means for independently changing the read position of each of the stored sampling data in accordance with the plurality of pitch modulation signals and outputting a plurality of sets of tone waveform data corresponding to the pitch modulation signals. A musical tone signal generator characterized in that