JP2765578B2 - Waveform signal controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention]   The present invention relates to a wave used in an electronic guitar, for example.
Shape signal control device
About what to do. [Background of the Invention]   Conventionally, waveforms generated by playing natural musical instruments
Extracts the pitch (fundamental frequency) from the signal and configures it with an electronic circuit.
Control the generated sound source device to artificially produce sounds such as musical sounds.
Various types have been developed.   In this type of electronic musical instrument, the pitch of the input waveform signal is extracted
The tone scale corresponding to the pitch to the tone generator.
It is common to tell them to happen.   For such electronic musical instruments, performances such as tremolo
When performing a playing operation, that is,
If you perform a picking operation during sounding, the sound is re-sounded
In order to perform the
Detected a sudden increase in the level of the input waveform signal
In that case, give a sound generation start command again and
(Relative on) processing to change volume, tone, etc.
To be able to follow the actual playing operation
An application was filed (Japanese Patent Application No. 61-285985).   However, instead of using the tremolo technique described above,
When the input waveform signal rises at
And the level of the input waveform signal increases rapidly several times
May be done.   For example, the rising of the input waveform is as shown in FIG.
Although it may be. In this case, a₀Start at the rising edge of the waveform
Despite the start of the sound, a_ThreeWith the waveform
Also before a_TwoWaveform level suddenly becomes larger than
Here, the above-described relative on (reproducing) processing is performed.
Despite the single picking operation
It is said that the sound start process is performed multiple times in succession
There was a problem. [Object of the invention]   The present invention has been made in view of the above circumstances,
Repronunciation according to the playing style and normal pronunciation
Can make the onset of musical tones more accurate
It is an object of the present invention to provide a waveform signal control device. [The gist of the invention]   To this end, the present invention provides a
Detects that the signal amplitude level has exceeded the specified value,
Immediately after extracting the frequency of the waveform signal,
Instructs the start of the sounding of the musical tone corresponding to the frequency.
During the specified period after the sound starts, the vibration exceeding the specified value
Indicating means even when a width-level waveform signal is detected
Do not output a command to start re-sounding
The point is that. [Example]   Hereinafter, an embodiment in which the present invention is applied to an electronic guitar will be described.
This will be described in detail with reference to the drawings.   In the present embodiment, as described below, step S₇~
S₁₁, S₁₆~ S_{twenty one}Is executed by the CPU 100 as an instruction means.
2, low-pass filter 3, maximum peak detection circuit 4, minimum
Peak detection circuit 5, Zero cross point detection circuit 6, A / D converter
Barter 11, latch 12, OR gate 13, flip-flop
14 and 15 are detection means, step S_{twenty three}, S₈~ S₁₁, S₁₆~ S_{twenty one}
Is executed by the CPU 100 as control means.₃₃, S₃₄Real
CPU 100 to execute₃₂CPU to execute
100 correspond to the prohibition means, respectively. Overall circuit configuration   FIG. 1 shows the overall circuit configuration of the embodiment, and FIG.
The signals of the two input terminals 1 are extended on the body of the electronic guitar.
The vibration of the strings provided for each of the six strings
This is a signal from the pickup that is converted into a signal.   The tone signal from the input terminal 1 is a pitch extraction circuit P1
~ P6 (In the figure, only the first string P1 shows its internal structure.
doing. ) Amplified by each internal amplifier 2 ...
-High-frequency component is cut by the low-pass filter (LPF) 3 ...
The basic waveform is extracted and the maximum peak detection circuit (MAX)
4…, minimum peak detection circuit (MIN) 5… and zero mark
Loss point detection circuit (Zero) 6... Low pass
Filter 3 is the frequency 4 of the vibrating sound frequency f of the open string of each string.
The cutoff frequency is set to 4f times. this is,
The frequency of the output sound of each string is within 2 octaves
It is based on Maximum peak detection circuit 4
The maximum peak point of the sound signal is detected, and the detected pulse signal
The flip-flop 14 connected to the subsequent stage at the rise of ...
… The Q output becomes High level and this flip-flop
14 ... and zero-cross point detection circuit 6 ...
The AND output with the inverted output of the data 30 is an AND gate 24.
Interrupt command signal INT via…_a1~ INT_a6As CPU100
Similarly, the minimum peak detection circuit 5
The minimum peak point of the sound signal is detected, and the detected pulse signal
The flip-flop 15 connected to the subsequent stage at the rise of ...
… The Q output becomes High level and this flip-flop
15 and the output of the zero-cross point detection circuit 6
AND output is interrupted via AND gate 25 ...
Signal INT_b1~ INT_b6To the CPU 100.   That is, the maximum peak point is detected and the flip-flop 14
Is high level, the waveform changes from positive to negative.
Interrupt command signal INT when turned off_a1~ INT_a6Gives to CPU100
Conversely, the minimum peak point is detected and the flip-flop is
The waveform changes from negative to positive when
Interrupt command signal INT when changed to_b1~ INT_b6Is CPU100
To enter.   Then, the CPU 100 receives these interrupt command signals.
Immediately after the digit, the corresponding flip-flops 14 ……, 15 ……
Clear signal CL_a1~ CL_a6, CL_b1~ CL_b6Occurs
set. Therefore, the next maximum peak or minimum peak
No matter how many times it passes through the zero-cross point until the
The corresponding flip-flops 14 ……, 15 …… are in reset state
Therefore, the CPU 100 is not interrupted.
You.   Then, the CPU 100 interrupts with the vibration output of the string.
Command signal INT_a1~ INT_a6Or INT_b1~ INT_b6Given
And at least one of the time intervals
It produces an unusual scale sound. Open at the start of sound production
Starts generating the scale of the string and corrects the frequency after pitch extraction.
It may be modified to a number. About the operation at the start of this sound
It will be described later.   The time interval is determined by a counter 7 as described later.
Using the work memory 101. That is,
Memory 101 immediately after the maximum peak point or
Counter 7 count at zero-cross point immediately after peak point
Various data such as default values are stored.   After the sound starts, the time interval data
Therefore, the frequency of the tone being generated is variably controlled.
That is, CPU 100 sends data specifying the scale to frequency ROM 8.
As a result, frequency data indicating the corresponding frequency is read.
The tone signal is sent to the tone generator 9 to generate a tone signal.
Sound is output from the round system 10.   Also, the tone signal from the low-pass filter 3...
Is given to the A / D converter 11 ..., and its waveform level
Is converted into digital data corresponding to   The output of the A / D converter 11 is a latch 12
…… is latched. Latch for this latch 12 ...
The signal is the output of the flip-flops 14 ..., 15 ...
Generated through OR gate 13 ...
Each time a point or minimum peak point is passed, latch 12
Stores a signal indicating the level of the waveform at that time. Ma
The latch signal L from this OR gate 13 ...₁~ L₆Is CP
Also given to the U100.   Then, the latch 12 outputs the output to the CPU 100 and generates a sound.
Control of start, stop, and output sound level (volume)
You will be done according to this data. Note that this latch 12
The peak value, which is the peak value stored in
It is sequentially written to the file 101.   That is, in the CPU 100, the A / D converter 11
The absolute value of the data indicating the waveform level
When a certain value is exceeded, the sound is started
At the same time, pitch (fundamental frequency) extraction is started,
When the data falls below a certain value, sound muting is issued and sound is emitted.
To end. The details of this operation will be described later.
You.   Note that FIG. 1 shows that the A / D converter 11
Circuits P1 to P6 are provided independently of each other, but one A / D
It is of course possible to use the barter in a time-division manner.   The frequency ROM 8 and the tone generator 9 are time-divisionally processed.
A tone generation system of at least six channels is formed.
You.   FIG. 2 shows the signal wave of each part in the pitch extraction circuit P1.
It shows a time chart of the shape.
The output of the filter 3 is the output of the maximum peak detection circuit 4.
Is the output of the minimum peak detection circuit 5, and is the zero cross
The output of the point detection circuit 6 is an interrupt command signal INT_a1~ INT
_a6, Is the interrupt command signal INT_b1~ INT_b6It is. motion   Next, the operation of this embodiment will be described. Figure 3 shows CPU1
FIG. 4 shows the flow of the interrupt routine of FIG.
It is a flow. FIGS. 3 and 4 show a single unit.
Only the processing for strings is shown, but the processing for all strings is
Since they are exactly the same, CPU 100
It may be considered to be executed in a divided manner. Registers in work memory 101   Before explaining the specific operation of the CPU 100,
Main registers in the memory 101 will be described.   The STEP register takes four levels, 0, 1, 2, and 3,
Vibration is performed (see FIG. 5 (a) or FIG. 6 (a)).
As shown in FIG. 5 (b) or FIG. 6 (b)
Its content changes as it does. This STEP register is 0
Time indicates a note-off (silence) state.   The SIGN register has the zero cross point for period measurement.
Whether it is the zero crossing point next to the large peak (MAX) point or the minimum
Indicates whether it is the next zero crossing point after the MIN point
Where 1 is the former and 2 is the latter.   The REVERSE register is represented by the SIGN register above.
Zero cross point after the peak point on the opposite side of the zero cross point
Checks whether interrupt processing has been performed due to the arrival of
This register stores the data to be synchronized.
Switch (fundamental frequency) extraction control.   T register is used to measure the period of the input waveform.
The value of the point counter 7 is stored. Note that the counter 7 is
Perform free running operation that counts with a fixed clock.
ing.   The AMP (i) register is connected from the A / D converter 11 to the latch 12
The maximum or minimum peak value latched at
AMP (1) is the maximum peak
AMP (2) is a register for the minimum peak.   The PERIOD register is a pitch data indicating the measured cycle.
Data is input and the CPU 100
Performs frequency control on the frequency ROM 8 and the tone generator 9
It is.   The W register is incremented by one every period from the start of sound generation.
The input waveform signal is changed until it reaches `` 10 ''.
The current peak level against the previous peak level value of the signal
Even if the value suddenly increases, processing for re-sounding is performed.
It is controlled not to be.   Further, as will be described later, this embodiment is used for various determinations.
Three constants (threshold level) in CPU100
Is set.   The first one is ONLEV I, Fig. 5 (a),
6 As shown in FIG.
And a peak value larger than this ONLEV I value is detected.
When the string is picked, etc.
The CPU 100 starts executing the operation for this.   ONLEV II is in the note-on (sounding) state.
The difference between the current detection level and the current detection level is more than this value
If there is any
Performs the sound generation (relative on) process
It is for the purpose.   OFFLEV is a note as shown in Fig. 8 (a).
In the on (sounding) state, the peak value below this value
When detected, a note-off (silence) process is performed.   From the above description, the interrupt routine and menu
It will be easier to understand the operation of the in-routine. Interrupt processing at zero crossing point   Now, the output of AND gate 24 or AND gate 25
Is the interrupt command signal INT_a, INT_bThe arrival of CPU100
Therefore, the interrupt processing of FIG. 3 is performed.   That is, the interrupt command signal INT_aWhen inputting
Top P₁And set the a register in CPU 100 to 1
Command signal INT_bWhen inputting_TwoPlace
Is set to 2 in the a register.   And then step P_ThreeIn, t register in CPU100
The value of the counter 7 is preset in the data. Then run
Step P_FourNow, the peak level data of A / D converter 11
Data from latch 12 and set to b register in CPU100
I do.   And step P_FiveIn the flip-flop 14 also
Alternatively, the flip-flop 15 is cleared.   Next step P₆In the above, the contents of the a, b, and t registers
Is transferred to the work memory 101 and the interrupt processing ends.
You. Main processing   In the main routine (FIG. 4), step S₁At
The interrupt processing described above causes the work memory 101
Contents of a ', b', t '(same as a, b, t above,
They are indicated as a ', b', and t 'for being recorded. )But
Judge whether or not it has been written, no interrupt processing
If not, a NO determination is made and this step S₁
Is repeatedly executed.   Then, the above step S₁If you make a YES decision,
Step S_TwoAnd read the contents a ', b', t '
You. Then step S_ThreeIn the above AMP (a ') Regis
Of the same type (ie, maximum or minimum)
The peak value of the load point is read out to the c register in CPU100, and this time
The extracted peak value b 'is set in the AMP (a') register.
Set.   Well, next step S_Four~ S₆In the STEP register,
Judge whether the contents are 3, 2, 1 respectively. I
Also, if it is the first state, the STEP register is 0
Then, step S_Four, S_Five, S₆Both are judged NO. Soshi
And then step S₇The peak value b 'detected this time is ON
Judge if greater than LEV I.   If the peak value b 'is smaller than ONLEV I,
Step S since the process of starting pronunciation has not yet been performed₁Return to
Suppose ONLEV I is as shown in Fig. 5 (a) and Fig. 6 (a).
If a larger input is obtained, step S₇Judgment
YES, step S₈Proceed to.   And step S₈Set 1 in the STEP register with
Step S₉Set REVERSE register to 0 and continue
Step S_TenAnd a '(that is, the z immediately after the maximum peak point)
1 at low cross point, zero cross point immediately after minimum peak point
In the case of, input the value of 2) into the SIGN register.   And step S₁₁In the above, the value of t 'is stored in the T register.
set. As a result, the contents of a 'are stored in the SIGN register.
(SIGN is 1 in the case of FIG. 5 (a) and FIG. 6 (a).
The contents of b 'are stored in the AMP register and the contents of t' are stored in the T register.
It will be set in the Vista. And step again
S₁Go back.   By the way, in the above description, FIGS. 5 (a) and 6 (a)
Complete the main routine immediately after the Locross point Zero1
Will do.   By the way, next, the main loop just after the zero crossing point Zero2
The processing in the chin will be described. In that case, step S above₁→
S_Two→ S_Three→ S_Four→ S_FiveData processing and pronunciation stage control
And execute the following step S₆Is judged YES at
Step S₁₂Go to   Now, as shown in FIGS. 5 (a) and 6 (a), waveforms are inputted.
When the signal changes in the positive direction during force, the SIGN register is set to 1.
Since the negative peak has passed this time,
Since the register is 2, a judgment of NO is made. In addition, if the same polarity
When the zero-cross point arrives immediately after the peak value of
Step S₁₂And make a judgment of YES and continue without any steps
S₁Return to   Well, now this step S₁₂So no judge
Step S₁₃Then, set the STEP register to 2 (Fig. 5
(B), see FIG. 6 (b)).   And step S₁₃Followed by step S₁₄Run and before
Peak value (AMP (SIGN)) and current peak value
(B ') is compared. Now, as shown in Fig. 5 (a),
The value x of₀Is smaller than the current value (x₁> X₀) Then, YES
In order to use the current time t 'as the measurement start point of the cycle (fifth
(See Fig. (C)) Step S₁₄From step S_Ten, S₁₁Real
And set the SIGN register to 2 and the t 'register
Transfer the contents to the T register.   Conversely, if the previous peak value is larger than the current peak
In other words, as shown in FIG.₁<X₀Then,
Step S₁₄Judge NO and step S_FifteenREVERSE cash register at
The star is 1. The SIGN register stores the previous value 1
Will keep. Therefore, in this case, the previous zero-cross point
(Zero1) is the starting point of the cycle measurement (Fig. 6)
(C)).   After passing the next zero crossing point (Zero3),
First, when executing the main flow, step S_FiveAnd YES
Judge being Step S₁₆Proceed to. This time a 'is 1
In FIG. 5, SIGN is 2, and in FIG. 6, SI is
Since GN is 1, in the case of FIG. 5, step S₁₆
NO Judge at Step S_FifteenHeyuki Step S
₁Return to In other words, after starting the cycle measurement,
The second peak (amplitude x_TwoCPU100 recognizes that
I do.   In the case of FIG. 6, step S₁₆So yes
Judgment is made, step S₁₇Heyuki REVERSE register is 1
Judge whether or not. If not 1, make a NO decision
Tep S₁Return to step S as described above._FifteenFruit
This register is set to 1 depending on the row, and step S
₁₇From step S₁₈Set the STEP register to 3 (6th
(Refer to Fig. (B).)₁₉At, t 'Regis
The value of counter 7 accepted by the current interrupt
From the T register, that is, the time of the zero-cross point Zero1
Is subtracted and stored in the PERIOD register.   That is, the size shown in FIG. 6 (c) is the length of one cycle.
Followed by step S₂₀Transfers the contents of t 'to the T register
To start a new cycle measurement.   And step S_{twenty one}In the above PERIOD register
CPU100 sounds to frequency ROM 8 and tone generator 9 with the contents of
Issue a command. Therefore, the musical tone is generated from this point.
You. Subsequently, the CPU 100 proceeds to step S₃₃With the value of the W register
Let it be "1".   By the way, in the case of FIG.
In the process of the main flow after the crossing point (Zero4),
Top S_FiveFrom step S₁₆Jump. Now, SIGN
Step 2₁₆Then, make a judgment of YES and continue
Step S as above₁₇→ S₁₈→ S₁₉→ S₂₀→ S_{twenty one}→ S₃₃
Is executed, and this time, the sound generation process shown in FIG.
The CPU 100 sets the low-cross point Zero2 to Zero4 as one cycle.
Recognize and start producing a tone with a frequency based on this length
(See FIG. 5 (d)).   In this way, the next zero
Start the cycle measurement process from the
Measurement ends at the zero-cross point next to the peak point on the side
In this way, one cycle of the waveform of the low-pass filter 3 output is
Has been extracted.   Then, after the sound generation start processing, the main routine is started.
In step S_FourFrom step S_{twenty two}Proceed to
As shown in FIG.
Judge if the vehicle exceeds OFFLEV.   If you exceed this level now, step S₃₂Proceed to
Then, it judges whether the value of the W register is "10" or more.
Is the value of the W register just now "1"?
Step S_{twenty four}To the contents of a 'and the SIGN register
Performs a match comparison of the contents, and if they do not match, S_FifteenGo to the next
In preparation for interrupt processing at the crossing point, if they match,
Passes peaks with opposite characteristics (positive / negative peaks)
Step S_{twenty five}Go to REVERSE register
Judge whether it is 1 or not, and do nothing if NO
Without step S₁Return to step S_{twenty five}And YES
If the judgment is made, step S_{twenty five}From step S₂₆
Go to t 'register to find new cycle (pitch)
Subtract the contents of the T register from the contents of the
Set to.   And step S₂₇The contents of the t 'register
Transfer to T register and follow step S₂₈PER determined at
CPU (frequency) control based on the value of the IOD register
For the frequency ROM 8 and the tone generator 9.   In other words, in this embodiment, the change in the vibration frequency of the string
And control the frequency accordingly.
You will do it in time.   And step S₂₈From step S₂₉Go to REVERS
The next cycle measurement is performed with the contents of the E register set to 0, and the
Top S₃₄Then, the value of the W register is set to +1 to “2”.
After that, step S₂₆~ S₂₉, S₃₄No
Switch extraction control processing, and sequentially increments the value of the W register by +1.
To go.   When the value of the W register becomes “10” or more, the CPU 1
00 is Step S₃₂From step S_{twenty three}Proceed to Relative
Judge whether to perform on (relative on) processing.
So that That is, specifically, the current peak value (b ')
Is ONLEV II greater than the previous peak value (c),
Whether the extracted peak value suddenly increased during
Judge.   Normally, if a string vibrates, natural attenuation will occur.
Top S_{twenty three}Is NO, but if you use tremolo
Therefore, before the previous string vibration has attenuated,
Is operated, this step S_{twenty three}Judgment of YES
There is.   In that case, step S_{twenty three}And judge YES.
Step S₈Jump to step S₉~ Step S₁₁Pronunciation pronunciation
Perform initial preparation processing. As a result, the STEP register becomes 1
This is exactly the same operation as the above-mentioned operation at the start of sound generation.
And then execute it. That is, step S again₁₆~ S_{twenty one}, S₃₃
To start re-sounding.
Will be.   Now, by one picking operation, as shown in FIG.
Assuming that such a waveform signal is input, the waveform a₀of
At the next zero cross point, step S₁~ S₁₁Data set
Tone processing, sounding stage discrimination processing, and sounding start preparation processing are performed.
And the next waveform a₁At the zero crossing point immediately after₁₆
~ S_{twenty one}, S₃₃Start processing is performed.
Step S_{twenty four}~ S₂₉, S₃₄Pitch extraction control processing is performed
Will go on. At this time, the waveform a_ThreeAs shown in the wave
The shape is not stable and the register of the fourth waveform from the beginning suddenly
Waveform a becomes larger_TwoWaveform a_ThreeLevel is ON
When it becomes larger than LEV II, the value of W register becomes
It is still "2", and the CPU 100₃₂W register
Is not more than “10”, so step S_{twenty three}, S
₈~ S₁₁Relative-on processing is not performed.
Tep S_{twenty four}~ S₂₉, S₃₄The pitch extraction control process is performed
Will go.   In this way, the picking operation is performed and the waveform rises
Sometimes the waveform is not stable and the level of the waveform suddenly
Even if it gets bigger, re-sounding processing is not performed and picking
Produce multiple times in succession even though the work is once
Will not be performed.   Then, as described above, the string vibration is attenuated,
So that the peak value falls below OFFLEV as shown in Fig. 8.
Then, step S_{twenty two}From step S₃₀Heyuki STEP Regis
Data to 0 and the following step S₃₁Note-off processing at
Sound processing) and mute the musical tones
Then, the CPU 100 instructs the tone generator 9.   During the period in which re-sounding is not performed, the period of the input waveform signal
Instead of a time period based on numbers, a time counter is stored in CPU100.
The time counter is used to count
It may be based on. In this case, the steps shown in FIG.
Step S₃₄Is omitted, and step S₃₃In the time counter
Performs reset processing, and then automatically counts
Step S₃₂And the time count value is constant time T_SWhether or not
Judge processing.   Step S in the above embodiment₃₂Was determined to be NO
Then, after making the values of b 'and c the same, step S_{twenty three}Proceed to
And as a result, do not perform relative on processing.
You may.   In the above embodiment, step S_{twenty three}Then, the above-mentioned
The waveform between the peak value b 'and the current peak value c
I tried to judge whether it changed or not, for example,
b ′ / c> S (S is a predetermined value of 1 or more)
So, judge whether the waveform changed suddenly
May be.   In addition, the waveform is compared with the previous peak value and the current peak value.
I judge whether it changed suddenly, but before
The comparison may be performed by comparing the crest value between the current time and the current time.   Due to other condition detection, the peak value may be
A sudden increase may be detected.   In the above embodiment, the zero crossing immediately after each peak point
CPU100 performs interrupt processing at the point, starts sound generation, period meter
Calculation, relative on, mute start, etc.
However, even if these processes are performed directly when each peak point is detected,
Good. In that case, the same result can be obtained. So
In addition, for example, by detecting the zero-cross point immediately before the peak point
Thus, the same processing as described above may be performed. Other criteria
The points can be variously changed.   Further, in the above embodiment, each processing is performed in the main flow.
Was executed, but the same
The processing may be executed.   Further, in the above embodiment, the present invention is applied to an electronic guitar.
Applied, but not necessarily limited to
There is an audio signal input from a microphone etc.
Or by extracting pitch from the electrical vibration signal
Signal and the corresponding pitch or scale circumference
What kind of form does the system generate at the wave number?
It may be something. Specifically, those with a keyboard
For example, electronic pianos, electronic versions of wind instruments, string instruments, examples
For example, it is equally suitable for electronic violins and kotos.
Can be used. [The invention's effect]   The present invention, as described in detail above, provides a
Signal amplitude level exceeds a predetermined value,
Immediately after extracting the frequency of the waveform signal, the
Instructs the user to start producing the musical tone corresponding to the wave number, and this musical tone is emitted.
Within the predetermined period when the sound starts, the amplitude again exceeds the predetermined value.
Even when a level waveform signal is detected,
Don't output re-pronounce start command
When a sound is generated and the waveform rises,
The shape is not stable, and the waveform level suddenly increases in the middle
However, the re-sounding process is not performed, and the picking operation is performed once.
Despite the fact that the sound processing was performed multiple times in succession
And as a result, depending on the tremolo playing style
The pronunciation of musical tones is clearly distinguished from
The effect is that the start can be made more accurate.
I do.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing the overall circuit configuration of an electronic musical instrument input control device according to an embodiment of the present invention, and FIG.
Time chart diagram showing waveforms and the like appearing in each part in the figure,
FIG. 3 is a diagram showing a flowchart of an interrupt routine of the CPU, FIG. 4 is a diagram showing a flowchart of a main routine of the CPU, FIG. 5 and FIG. FIG. 7 is a time chart showing an example in which the input waveform at the start of sound generation is not stable. FIG. 1 ... input terminal, 4 ... maximum peak detection circuit, 5 ... minimum peak detection circuit, 6 ... zero cross point detection circuit, 7 ...
... Counter, 9 ... Sound source circuit, 12 ... Latch, 14, 15 ...
... Flip-flop, 100 ... CPU, 101 ... Work memory, P1 to P6 ... Pitch extraction circuit.

   ────────────────────────────────────────────────── ─── Continuation of front page       (56) References JP-A-51-123133 (JP, A)                 JP-A-55-159495 (JP, A)                 JP-A-54-138477 (JP, A)                 Tokiko Sho 61-2954 (JP, B2)

Claims (1)

(57) [Claims] Frequency extracting means for detecting that the amplitude level of the supplied waveform signal has exceeded a predetermined value and extracting the frequency of the waveform signal; and immediately after the frequency extracting means has extracted the frequency of the waveform signal, Instruction means for instructing the start of sounding of a musical tone corresponding to the frequency of the waveform signal; and within a predetermined period after the instruction means gives an instruction for starting sounding of the musical tone, the frequency extracting means again exceeds the predetermined value. A waveform signal control device comprising: a prohibition unit that controls the instruction unit so as not to output a command to start re-sounding even when a waveform signal of an amplitude level is detected. 2. 2. The method according to claim 1, wherein the prohibition unit recognizes the predetermined period by counting whether or not the number of cycles corresponding to the frequency of the waveform signal has reached a predetermined value. Waveform signal control device. 3. 2. The method according to claim 1, wherein the prohibiting unit recognizes the predetermined period by measuring time.
Item 6. The waveform signal control device according to Item 1.