JP2001159893A - Effect device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reproduce a musical sound waveform with a rich musical sound even after various processing with respect to an effect device for receiving a waveform showing a musical sound and obtaining a reproduced waveform based on the input waveform. SOLUTION: This effect device is provided with a waveform reproduction means 34 for reproducing the waveform based on a prescribed partial waveform among the waveforms stored in a waveform memory 33 and based on a present pitch of the input waveform detected by a pitch detection means 37 after the reproduction of the waveform of a prescribed section from attack stored in the waveform memory means 33.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an effect device for inputting a waveform representing a musical tone and obtaining a reproduced waveform based on the input waveform in real time.

[0002]

2. Description of the Related Art Conventionally, there has been known an apparatus such as an effector which inputs a waveform representing a musical tone and generates a reproduced waveform in real time based on the input waveform.

[0003] Here, a musical tone such as a guitar or a percussion instrument which rises rapidly and attenuates (hereinafter referred to as "decay tone") has a richest harmonic component at the rising portion and the harmonic component decreases as it attenuates. And the waveform changes to a simple attenuated sinusoidal waveform. Such attenuation music
For example, when controlling or processing according to the performance using an effector or the like, for a part where the attenuation has progressed and the harmonic component has been reduced, a good playing effect and a good processing waveform can be obtained even if the part is controlled or processed using that part. There is a problem that it may not be possible to obtain. To solve this problem, it is conceivable to electronically synthesize a tone signal containing a large amount of harmonic components even in the attenuated part, which is different from the original attenuated tone, Since a tone signal different from the tone is synthesized, the tone tends to be uncomfortable in the sense of hearing.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide an effect device for reproducing a musical sound waveform in real time, which can provide a rich musical tone even after various processings. .

[0005]

To achieve the above object, the effect device of the present invention comprises: a waveform storage means for storing at least a portion within a predetermined section from the rise of an input waveform representing a musical tone; Pitch detection means for detecting a pitch, a first waveform is reproduced based on the waveform in the predetermined section stored in the waveform storage means, and subsequent to reproduction of the first waveform, the waveform storage means Waveform reproducing means for reproducing a second waveform based on the current pitch of the input waveform detected by the pitch detecting means, based on a waveform of a predetermined part of the stored waveforms. .

[0006] The effect device of the present invention is a
Instead of the waveform of the attenuated portion, the waveform is reproduced using the waveform of the portion of the same attenuated tone that is rich in harmonic components, so that a natural tone can be generated. The high-frequency component can be abundantly included in the portion where the attenuation is advanced, and a musical tone with a rich expression can be obtained even if various processes are performed.

Here, in the effect device of the present invention,
It is preferable that the apparatus further comprises a waveform control means for controlling the change of the predetermined part to another part during reproduction of the second waveform by the waveform reproduction means.

When the predetermined one portion is sequentially changed into separate portions by the waveform control means, the musical tone can be changed in accordance therewith, and a more natural feeling or a more positive effect is provided. Music can be generated.

Here, the waveform control signal may be such that the predetermined portion is sequentially changed into separate portions in accordance with a sequence created in advance or at random, or an operator is provided, and the operator is provided with the operator. It may be changed according to the operation.

The range in which the predetermined portion is changed by the waveform control means may exceed the range of the waveform stored in the waveform storage means.

For example, the range of the change of the predetermined part is extended to a part (the musical tone waveform does not exist here) before the attack part around the sharp rising part (attack part) of the musical tone. The attack part may be reproduced a plurality of times to reproduce a musical sound waveform having a strong impression.

In the effect device of the present invention, it is preferable that the waveform reproducing means includes a waveform expanding means for performing an expanding process for extending a waveform reproducing time while preserving a pitch in reproducing the first waveform. .

[0013] By doing so, it is possible to generate a tone which is less unnatural in audibility than a tone obtained by synthesizing a tone signal different from the original waveform tone, and to obtain a waveform which can be favorably processed by an effector or the like. Can be.

[0014]

Embodiments of the present invention will be described below.

FIG. 1 is a block diagram showing the overall configuration of a guitar amplifier incorporating the effect device of the present invention.

An analog musical tone waveform generated by playing the electric guitar is input from an input terminal 1 of the guitar amplifier 10 and converted into digital waveform data by an A / D converter 2 to be converted into a DSP (Digital Si).
(Gnal Processor) 3. An effect device as one embodiment of the present invention is realized inside the DSP 3, and further, the DSP 3
Then, an effect such as distortion is applied in real time to the reproduced waveform obtained by the effect device.

The digital musical sound waveform output from the DSP 3 to which the effect has been applied is converted into an analog musical sound waveform by the D / A converter 4, further amplified by the amplifier 5, and emitted to the space as a musical sound by the speaker 6. Is done.

The amplifier 5 has a power supply voltage +
When A and -A are given and the level of the input signal to the amplifier 5 is large, the output signal from the amplifier 5 is almost a signal clipped by the power supply voltages + A and -A.

FIG. 2 is a functional block diagram showing a configuration of the effect device as one embodiment of the present invention, which is realized in the DSP 3 shown in FIG.

The digital input waveform data obtained by the A / D converter 2 shown in FIG. 1 is inputted to the effect device 30 shown in FIG. 2, and the limiter 31, the attack detecting means 35, the envelope detecting means 36, and the pitch It is input to the detecting means 37.

The limiter 31 is arranged such that when the waveform signal based on the waveform data output from the DSP 3 (see FIG. 1) including the effect device 30 passes through the amplifier 5, the amplitude becomes smaller than the maximum output level + A, -A. In addition, the amplitude of the input waveform is limited.

The input waveform data passing through the limiter 31 is input to the writing means 32. This writing means 32
Stores the input waveform data input via the limiter 31 in a time-series manner in the waveform memory means 33 in which the waveform data is stored in response to the write start instruction from the control means 39.

The input waveform is, for example, as shown in FIG.
The attack detecting means 35 detects the attack (start of inputting the waveform) of the input waveform, and the envelope detecting means 36 detects the instantaneous envelope of the input waveform. The pitch detecting means 37 detects the instantaneous pitch of the input waveform. The detection results of the attack detecting means 35, the envelope detecting means 36, and the pitch detecting means 37 are input to the control means 39.

Also provided here are controls 38, such as pedal controls, which are controlled by the player.
Information representing the operation amount of the operation element 38 is also input to the control means 39.

The control means 39 comprises an attack detecting means 35, an envelope detecting means 36, a pitch detecting means 3
7, and controls the writing unit 32 and the waveform generation unit 34 based on information input from the operation unit 38 and the like.

The waveform generating means 34 includes a waveform memory means 33
In this embodiment, the first generation unit 3 generates the reproduced waveform data based on the waveform data stored in the first generation unit 3.
41, a second generating means 342, and an adder 343 for mutually adding the waveforms generated by the first and second generating means 341 and 342.

FIG. 3 is a functional block diagram showing the structure of the first generating means 341 constituting the waveform generating means 34 of the effect device 30 shown in FIG.

The reading means 3411 constituting the first generating means 341 is a waveform memory means 33 (see FIG. 2).
The control means 39 (see FIG. 2)
Is read out at the address sidewalk amount corresponding to the expansion coefficient Ts input from. Here, the reading speed is equal to or lower than the writing speed of the waveform data to the waveform memory unit 33 by the writing unit 32 shown in FIG. 2, and when the reading speed is lower than the writing speed, only the lower speed is used. , The input waveform data is read out with time expansion.

The pitch converting means 3412 of the first generating means 341 shown in FIG. 3 performs the reading operation when the reading means 3411 reads from the waveform memory means 33 at a speed different from the writing speed by the writing means 32. Since the pitch of the obtained waveform data is different from the pitch of the original input waveform data, it is a means for correcting the pitch. Here, the read speed is equal to or lower than the write speed, and when the read speed is lower than the write speed, the pitch of the read waveform data is smaller than the pitch of the original input waveform data. Therefore, the pitch conversion means 3412 performs pitch conversion such that the pitch is the same as the pitch of the original input waveform data.

The conversion means 3413 is connected to the control means 39 (FIG. 2).
This is means for converting the expansion coefficient Ts input from the reference means into a pitch conversion amount.
In advance, a correspondence table between the expansion coefficient Ts and the pitch conversion amount is constructed. When the expansion coefficient Ts is input, the conversion means 3413 refers to the correspondence table and converts the expansion coefficient Ts into the pitch conversion amount. The pitch conversion amount is passed to the pitch conversion means 3412. The pitch converter 3412 converts the waveform data read by the reader 3411 into waveform data whose pitch has been converted by the received pitch conversion amount.

Here, the expansion coefficient Ts is generated by the control means 39 shown in FIG. 2, and its initial value is
The value is 1.0 which indicates reading at the same speed as the writing speed (that is, not expanding), and the envelope detection means 36 sets the attack level (peak level) of the input waveform data to the maximum of the amplifier 5 (see FIG. 1). When a predetermined value Amax (see FIG. 11A) corresponding to the amplitude is exceeded, the value is set to a value smaller than 1.0 in accordance with the exceeded level. As described above, the expansion coefficient Ts represents the increment of the waveform reading by the reading unit 3411. The smaller the expansion coefficient Ts is, the more slowly the waveform data is read and the longer the data is expanded. . This is not directly related to the present invention.
According to the method described in 348474, when the attack level of the input waveform exceeds the maximum output of the amplifier, the effect of obtaining a sense of volume equal to or higher than the maximum output of the amplifier is provided by time-expanding the attack portion of the input waveform. This is the process to perform.

The waveform data output from the pitch conversion means 3412 is multiplied by a cross-fade coefficient C11 (see FIG. 7 (e)) input from a cross-fade generation means 3415 described below by a multiplier 3414 and output. Is done.

Crossfade coefficient C2 generating means 3415
Is the crossfade coefficient C according to the instruction of the control means 39.
By generating 2 and inputting it to the multiplier 3414,
It controls the waveform data input to the multiplier 3414, and in accordance with the start-up instruction of the control means 39,
The crossfade coefficient C2 is changed from a value 0 to a value 1 at a predetermined slope.
The cross-fade coefficient C2 falls from a value of 1 to a value of 0 at a predetermined gradient in response to a fall instruction from the control means 39.

FIG. 4 is a functional block diagram showing the configuration of the second generating means 342 constituting the waveform generating means 34 of the effect device 30 shown in FIG.

FIGS. 5 to 8 show the second generation means 34.
FIG. 5 is an operation explanatory diagram of FIG.
FIG. 6 shows the envelope of the waveform data stored in the waveform data stored in the waveform memory means 33 (see FIG. 2). FIG. 6 shows the waveform of the waveform data stored in the waveform memory means 33 in the vicinity of the reference address shown in FIG. FIG. 7 is a view showing data, and FIG. 7 is a graph showing a window function having the waveform data shown in FIG. 8 is a diagram (a) in which the waveforms in FIG. 7 are superimposed while being shifted by the reproduction pitch P, and a diagram (b) in which the superimposed waveforms are combined. The reproduction pitch here corresponds to the instantaneous pitch of the input waveform obtained by the pitch detection means 37.

The phoneme segmentation type waveform generating means 3421 constituting the second generating means 342 shown in FIG.
9 (see FIG. 2), the information on the reference address and the reproduction pitch is received, and as shown in FIGS. 5 to 8, two periods (2 × P) of the reproduction pitch period P starting from the reference address. 7 is cut out in a state where a window function as shown by a broken line in FIG. 7 is multiplied, and the waveform data for the two cycles is repeatedly used and shifted by a period P as shown in FIG. In addition, the waveform data is synthesized as shown in FIG. The phoneme extraction method waveform generation means 3421 of the second generation means 342 shown in FIG. 4 cuts out the phonemes for two pitches and generates the waveform in this manner. Here, by setting the reference address to a portion where the harmonic components still remain in the waveform data, the phoneme segmentation type waveform generating means 3421 reproduces the waveform data rich in the harmonic components. You.

The envelope multiplier 3422 constituting the second generating means 3421 shown in FIG. 4 receives the envelope data of the input waveform data from the control means 39. The envelope multiplier 3422 includes a phoneme extraction type waveform generating means. An envelope is added to the reproduced waveform data generated in 3421. The reproduced waveform data to which the envelope has been added is input to another multiplier 3423, and the multiplier 3423 adds the envelope-added reproduced waveform data to the cross-fade coefficient C3 as shown in FIG. Is multiplied. The reproduced waveform data multiplied by the crossfade coefficient C3 is added to an adder 342.
7 is input.

Crossfade coefficient C3 generating means 3424
Is for controlling the waveform data input to the multiplier 3423 by generating a cross-fade coefficient C3 in accordance with an instruction from the control means 39 and inputting the same to the multiplier 3423. 3423
Then, according to the start-up instruction from the control means 39, the value 0
The cross-fade coefficient C3 in the state of (1) rises to a value of 1 with a predetermined slope, and the cross-fade coefficient C3 of the state of 1 falls to a value of 0 with a predetermined slope in accordance with a fall instruction from the control means 39. .

In still another multiplier 3425, the input waveform data to the effect device 30 shown in FIG. 2 (the waveform data output from the A / D converter 2 shown in FIG. 1) is added to FIG.
A crossfade coefficient C1 as shown in FIG. The input waveform data multiplied by the crossfade coefficient C1 is input to the adder 3427. In the adder 3427, it is input to two multipliers 3243 and 3425. In the adder 3427, the two waveform data output from the two multipliers 3243 and 3425 are added to each other and output.

Crossfade coefficient C1 generating means 3426
Generates a cross-fade coefficient C1 in accordance with an instruction from the control means 39 and inputs the same to the multiplier 3425. The cross-fade coefficient C1 generation means 3426 outputs a value in accordance with a start-up instruction from the control means 39. The value of the crossfade coefficient C1 in the state of 0 is set to 1 with a predetermined slope.
The cross-fade coefficient C1 in the state of the value 1 is lowered to the value 0 with a predetermined slope in accordance with the falling instruction from the control means 39.

The waveform data generated by the first generation means 341 shown in FIG. 3 and the waveform data generated by the second generation means 342 shown in FIG.
3 are added to each other and output.

FIG. 9 is a flowchart showing the processing (main routine) of the control means 39 constituting the effect device 30 shown in FIG.

This main routine is started when the power is turned on. Initially, an initial setting is performed in step S01. Thereafter, in step S02, it is monitored whether or not an attack is detected by the attack detecting means 35 (see FIG. 2). Are repeated, and when an attack is detected, the waveform generation processing of step S03 is executed. This waveform generation processing will be described in detail below. When this waveform generation processing ends, the process returns to the step S02, and the detection of the next attack is monitored.

FIG. 10 is a flowchart of a waveform generation process executed in step S03 of the main routine of FIG.
FIG. 11 is a signal waveform diagram showing a signal waveform of each unit.

In FIG. 11, the horizontal axis is a time axis.
(A) shows the envelope of the input waveform input to the effect device 30 shown in FIG. 2, and (b) shows the second waveform shown in FIG.
Generation means 342 of the cross-fade coefficient C1 generation means 3
426 represents a cross-fade coefficient C1 generated,
(C) shows the envelope of the input waveform data written into the waveform memory means 33 by the writing means 32 of the effect device 30 of FIG. 2, and (d) shows the first envelope shown in FIGS.
(E) represents the cross-fade coefficient C2 generated by the cross-fade coefficient C2 generation means 3415 of the first generation means 341 shown in FIG. f) represents the envelope of the reproduced waveform data generated by the second generation means 342 shown in FIGS. 2 and 4, and (g) shows the cross-fade coefficient generated by the cross-fade coefficient C3 generation means 3424 shown in FIG. C3 is shown.

Hereinafter, the waveform generation processing shown in FIG. 10 will be described. In the following, description will be made with reference to FIG. 11, FIG. 2, FIG. 3, and FIG.

First, in step S101, the first generation means 341 is initialized. This initialization includes initialization of the readout unit 3411 of the first generation unit 341, initialization of setting the expansion coefficient Ts to a value of 1, and generation of the crossfade coefficient C3 generation unit 3424 of the second generation unit 342. And the initialization for lowering the crossfade coefficient C3 to a value 0.

Next, in step S102, the writing means 32 is instructed to start writing the input waveform data. The writing means 32 receives the instruction, and thereafter,
The input waveform data is stored in the waveform memory means 33.

In step S103, the cross-fade coefficient C1 generating means 3426 is instructed to lower the cross-fade coefficient C1, and in step S104, the cross-fade coefficient C2 generating means 3415 is instructed by the cross-fade coefficient C1.
2 is instructed, and in step S105,
The first generation unit 341 is instructed to start waveform generation.
After receiving this instruction, the first generation unit 341 performs the above-described waveform reproduction.

Next, in step S106, the envelope detected by the envelope detecting means 36 is monitored, and the attack level (peak level) is set to a predetermined value Amax.
It is monitored whether or not this is the case. This predetermined value Amax
Corresponds to the level at which the amplifier 5 shown in FIG. 1 is clipped by the power supply voltages + A, -A.

If the attack level is equal to or higher than Amax in step S106, the process proceeds to step S107, where the expansion coefficient Ts is changed. Here, according to the extent that the attack level has exceeded the predetermined value Amax, the greater the extent of exceeding the predetermined value, the smaller the expansion coefficient Ts is changed to a smaller value.

If the attack level is less than Amax in step S106, the process proceeds to step S108 while keeping the expansion coefficient Ts at the initial value 1.

In step S108, the first generating means 3
Until the amplitude value of the waveform generated in 41 falls below the predetermined value V (step S109), the waveform amplitude value is monitored.

If it is determined in step S109 that the amplitude value of the waveform generated by the first generation means 341 has become equal to or smaller than the predetermined value V, the flow advances to step S110, and a write end instruction is issued to the writing means 32. Then, in step S111, the cross-fade coefficient C2 generating means is instructed to fall to the value 0 of the cross-fade coefficient C2, and in step S112, the first generating means 341 is generated.
At step S113, the cross-fade coefficient C3 generating means 3424 is instructed to raise the value of the cross-fade coefficient C3 to 1, and at step S114, the second generating means 342 is instructed to output the reference address and The reproduction pitch is set, and the start of waveform generation is instructed. The reference address set in step S114 is a read address when the first generation unit 341 completes waveform generation in step S112. As a result, a transition is made from waveform generation by the first generation unit 341 to waveform generation by the second generation unit 342.

Next, in step S115, the envelope data of the input waveform obtained from the envelope detection means 36 is set in the envelope multiplier 3422 of the second generation means 342, and is changed according to the operation state of the operation element 38. The reference address thus set is set in the second generation unit 342.

In the present embodiment, the reference address can be changed up to the attack portion (the portion where the writing of input waveform data to the waveform memory means 33 is started).

In step S117, the pitch detecting means 3
7 is monitored to determine whether or not the current pitch of the input waveform can be detected. If a pitch has been detected, the process proceeds to step S118, and the detected pitch is determined by the second generation unit 342.
Is set as the reproduction pitch, and the process returns to step S115.

If the input waveform is attenuated such that the pitch of the input waveform cannot be detected while repeating the processing of steps S115 to S118, it is determined in step S117 that the pitch could not be detected.
Move to 9.

In step S119, the second generation unit 3
In step S42, the cross-fade coefficient C3 generating means 3424 is instructed to lower the cross-fade coefficient C3.
At 120, the second generation means 342 is instructed to start waveform generation, and at step S121, the cross-fade coefficient C1 generation means 3426 of the second generation means 342 is instructed to lower the cross-fade coefficient C1.

As described above, from the effect device 30, the waveform generated by the first generating means 341 (FIG. 11)
(D)) is output, and subsequently, the falling time (rising time of the cross-fade coefficient C3) of the cross-fade coefficient C2 (FIG. 11 (f)) is applied to the second generation means 342. The process proceeds to the output of the waveform (FIG. 11F), after which the waveform generated by the second generation means 324 is output, and thereafter, the cross-fade coefficient C3
Multiplying the rise time of FIG. 11 (g) (cross-fade coefficient C1 (fall time of FIG. 11 (b))), it shifts to the output of the attenuated portion of the input waveform (FIG. 11 (a)).
As a result, a tone signal in which the effect of the attack emphasis and the effect of abundantly containing higher harmonic components in the attenuator are added to the input waveform signal is reproduced.

Here, the reproduced waveform generated by the first generating means 341 uses a portion of the input waveform rich in harmonic components, and is generated by the second generating means 342. Even if the reproduced waveform is not operated by the operator 38, the waveform is reproduced on the basis of the portion of the input waveform portion used by the first generating means, which has a sufficient remaining harmonic component. Therefore, the second generation unit 342
The waveform generated by the above also sufficiently contains harmonic components.

In the present embodiment, each of the above-described processes is executed in real time from the viewpoint of a human.

The present invention is not limited to the above embodiment, and may be, for example, as follows.

(1) A reference address may be set according to a change in the pitch of an input waveform such as a guitar choking performance, instead of the operation amount of the operator.

(2) Rather than directly associating a change in the operation amount of the operator or a change in the pitch of the input waveform with a change in the reference address, the reference address is changed according to the operation amount or the differential value of the change in the pitch. Alternatively, the reference address may be changed in accordance with the result of the calculation using the operation amount or the pitch change amount as a variable.

(3) In the above-described embodiment, the second generation means 342 cuts out an address of a predetermined section (in the above-described embodiment, two cycles of the reproduction pitch) starting from the reference address, and forms a waveform for the two cycles. Although the reproduced waveform is generated from the above, the section where the waveform is cut may be changed around the reference address over time.

(4) In the above-described embodiment, the change range of the reference address is up to the attack portion. However, the change range of the reference address is included before the attack portion, and the attack portion is reproduced a plurality of times. You may make it.

(5) In the above-described embodiment, when the amplitude decreases to a level at which the pitch of the input waveform cannot be detected, the second generation means 342 performs loop reproduction of the waveform cut out from the reference address portion, Switching to the waveform, but without switching to the input waveform, continues loop playback with the last pitch that could be detected, or the inability to detect the pitch means that the amplitude has become sufficiently small. Therefore, the reproduction of the waveform may be terminated at that time.

[0069]

As described above, according to the present invention, it is possible to reproduce a tone waveform that is natural and rich in harmonic components.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the overall configuration of a guitar amplifier incorporating an effect device of the present invention.

FIG. 2 is a functional block diagram illustrating a configuration of an effect device according to an embodiment of the present invention implemented in the DSP illustrated in FIG. 1;

FIG. 3 is a functional block diagram illustrating a configuration of a first generation unit that forms a waveform generation unit of the effect device illustrated in FIG. 2;

FIG. 4 is a functional block diagram illustrating a configuration of a second generation unit included in the waveform generation unit of the effect device illustrated in FIG. 2;

FIG. 5 is an explanatory diagram of an operation of a second generation unit.

FIG. 6 is an explanatory diagram of an operation of a second generation unit.

FIG. 7 is an explanatory diagram of an operation of a second generation unit.

FIG. 8 is a diagram illustrating the operation of a second generation unit.

FIG. 9 is a flowchart showing a process (main routine) of a control unit constituting the effect device shown in FIG. 2;

FIG. 10 is a flowchart of a waveform generation process executed in step S03 of the main routine in FIG. 9;

FIG. 11 is a signal waveform diagram showing a signal waveform of each unit.

[Explanation of symbols]

 Reference Signs List 1 input terminal 2 A / D converter 3 DSP 4 D / A converter 5 amplifier 6 loudspeaker 10 guitar amplifier 30 effect device 31 limiter 32 writing means 33 waveform memory means 34 waveform generation means 35 attack detection means 36 envelope detection means 37 pitch Detecting means 38 Operator 39 Control means 341 First generating means 342 Second generating means 343 Adder 3411 Reading means 3412 Pitch converting means 3413 Converting means 3414 Multiplier 3415 Crossfade coefficient C2 generating means 3421 Phoneme segmentation type waveform generating means 3422 Envelope multiplier 3423 Multiplier 3424 Crossfade coefficient C3 generating means 3425 Multiplier 3426 Crossfade coefficient C1 generating means

Claims (3)

[Claims] 1. An input waveform representing a musical tone, at least
Waveform storage means for storing a portion within a predetermined section from a rise; pitch detection means for detecting a pitch of the input waveform; first waveform based on a waveform within the predetermined section stored in the waveform storage means After the reproduction of the first waveform, and based on the waveform of a predetermined part of the waveforms stored in the waveform storage means and the current pitch of the input waveform detected by the pitch detection means. An effect device comprising: a waveform reproducing unit that reproduces a second waveform based on the waveform. 2. The effect device according to claim 1, further comprising a waveform control unit for controlling the change of the predetermined part to another part during reproduction of the second waveform by the waveform reproduction unit. . 3. The waveform reproducing means according to claim 1, wherein said waveform reproducing means includes a waveform decompressing means for performing decompression processing for extending a waveform reproduction time while preserving a pitch when reproducing said first waveform. The effect device as described.