JPH05216468A - Electronic musical instrument - Google Patents

Abstract

PURPOSE:To specify a pitch which accurately corresponds to an actual interval and to start generating a musical sound having the accurate pitch by correcting a pitch to the original pitch period by a pitch information correcting means only at the time of the rising of an input waveform signal when pitch informa tion is extracted from the input waveform signal. CONSTITUTION:Pitch extracting means 1 and 2 convert the input waveform signal which is previously obtained, into a digital signal, detect an effective peak value of the digital waveform and the zero-cross time right after it in order, and decide a pair of the effective peak value and zero-cross time right before or behind it to extract the pitch period. Then the pitch information correcting means 3 corrects only the pitch period which is detected by the pitch extracting circuit 1 and 2 for the 1st time by several percents. Then, a musical sound generating means 5 generates the musical sound having the pitch corresponding to the corrected pitch information at the time of the rising of the input waveform, but generates the musical sound with the pitch corresponding to the pitch information extracted by the pitch extracting means after the rising.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a synthesizer type electronic musical instrument such as an electronic stringed instrument such as an electronic guitar, and more particularly to an electronic instrument capable of extracting pitch information from an input waveform signal and producing a musical tone at a pitch corresponding thereto. Regarding musical instruments.

[0002]

2. Description of the Related Art When a guitar or the like is played, string vibrations or the like are detected as an electric signal, and a tone generating circuit composed of a digital circuit or the like is controlled according to the input waveform signal to synthesize a tone and emit the tone. Electronic musical instruments have been developed.

Particularly, in the type in which the pitch extracting circuit extracts the pitch period from the input waveform signal and the musical tone generating circuit generates a musical tone having a pitch corresponding to the pitch period, the pitch of the musical tone is the input waveform signal. Since it responds directly, it is possible to perform tone synthesis that is faithful to the performance of the performer.

In such a conventional pitch extraction type electronic musical instrument, pitch information is generated in direct proportion to the frequency corresponding to the pitch period extracted from the pitch extraction circuit, and a musical tone is generated in accordance with the pitch information. ..

[0005]

However, for example, in an acoustic guitar or the like, when a string vibration waveform when picking a string is picked up by a pickup or the like and a pitch period is extracted from the waveform signal, the picking start time, that is, the string vibration At the beginning (at the rising edge of the input waveform signal), there is a tendency that a pitch period shorter than the period corresponding to the pitch when the fret of the actual guitar is pressed is extracted. That is, the frequency corresponding to the extracted pitch period becomes higher than the actual pitch.
This is due to, for example, non-linearity of string vibration.
It should be noted that, after the start, it matches the actual pitch.

Therefore, if the pitch information that is directly proportional to the frequency corresponding to the extracted pitch period is generated, a pitch higher than the actual pitch is specified when the input waveform signal rises. In an electronic musical instrument which specifies a pitch with, there is a problem that the pronunciation of a musical sound is started with an incorrect pitch.

The above problem is not limited to electronic stringed instruments, but can generally occur in electronic musical instruments that can generate a non-linear input waveform signal at the time of rising. An object of the present invention is to make it possible to specify a pitch that accurately corresponds to an actual pitch even when the input waveform signal rises.

[0008]

SUMMARY OF THE INVENTION According to the present invention, for example, an electronic string instrument of the type in which metal string vibration is detected as an input waveform signal by a pickup, pitch information is extracted from this, and the pitch of a musical tone is controlled by the pitch information ( It is an electronic musical instrument realized as an electronic guitar).

And, firstly, it has a pitch extracting means for extracting pitch information from the input waveform signal. For example, the means digitizes an input waveform signal obtained in advance, sequentially detects an effective peak value and a zero-cross time immediately after that from the digital waveform, and sets the effective peak value and a zero-cross time immediately after or immediately before it. Is a means for extracting the pitch information, that is, the pitch period, as the interval of each zero-cross time. Alternatively, it is a means for extracting the pitch period as the interval between the peak values.

Next, the pitch information correction means is a means for correcting the pitch information at the rising edge of the input waveform extracted by the pitch extraction means. The means is a means for correcting only the pitch period first detected by the pitch extracting means by a few percent in a state where no input waveform is detected, for example.

The musical tone generating means generates a musical tone at a pitch according to the pitch information corrected by the pitch information correcting means at the rising of the input waveform, and at the pitch extracting means after the rising. It is a means for generating a musical tone at a pitch according to the extracted pitch information.

[0012]

The operation of the present invention is as follows. For example, in an electronic stringed instrument such as a guitar synthesizer, when a player picks a string, the pitch extracting means extracts a pitch period (pitch information) from a string vibration waveform (input waveform signal) detected by a pickup or the like. In this case, only at the start of picking (at the rising edge of the input waveform signal), the pitch period is shortened by a few percent, for example. That is, in terms of frequency, frequencies higher than the pitch of the implementation are extracted.

Then, the pitch period is corrected by the pitch information correcting means so as to become the original pitch period. As a result, the musical tone generating means generates a musical tone at a pitch according to the corrected pitch period at the rising of the input waveform signal, so that the actual pitch can be accurately adjusted even at the rising of the input waveform signal. You can specify the corresponding pitch. It should be noted that after the rising time, a musical tone may be generated with a pitch corresponding to the pitch cycle extracted by the pitch extracting means.

[0014]

EXAMPLES Examples of the present invention will be described in detail below. In the following description, the symbols {},
Items will be grouped in order of the underlined headings that are surrounded by () and <<. {Structure of Electronic Musical Instrument According to the Present Invention} In this embodiment, six metal strings are stretched on the body, and a desired string is pressed while pressing the fret (fingerboard) provided at the bottom of the metal strings with a finger. It is realized as an electronic guitar that plays by picking. The appearance is omitted.

FIG. 1 is an overall configuration diagram of this embodiment.
First, the pitch extraction analog circuit 1 is provided for each of the six strings (not shown), and each output from the hex pickup for converting the vibration of each string into an electric signal is passed through a low-pass filter (not shown). By removing the harmonic components by using the waveform signals Wi (i = 1
~ 6) is obtained. Further, each time the sign of the amplitude of each waveform signal Wi changes to positive or negative, a pulse-shaped zero-cross signal Zi (i = 1 to 6) that becomes high level or low level is generated. Then, these six types of waveform signals Wi and zero-cross signals Zi are converted into a time-division serial zero-cross signal ZCR and a digital output (time-division waveform signal) D1 by an A / D converter or the like not particularly shown, and output. To do.

The pitch extraction digital circuit 2 includes a peak detection circuit 201 and a time constant conversion control circuit 2 as shown in FIG.
02, a peak value acquisition circuit 203, and a zero-cross time acquisition circuit 204. Each of these circuits shown in FIG. 2 includes a serial zero-cross signal ZCR and a digital output D1 from the pitch extraction analog circuit 1 (FIG. 1) obtained by time division of 6 strings.
Based on, the 6-strings are time-divisionally processed. In the following description, the processing for one string will be described for ease of explanation, and the serial zero-cross signal ZCR and the digital output D1 will be described as an image of the signal for one string. It is assumed that division processing is being performed.

In FIG. 2, first, the peak detection circuit 20.
1 detects the maximum peak point and the minimum peak point of the digital output D1 based on the serial zero-cross signal ZCR and the digital output D1. Therefore, although not shown in the figure, a peak hold circuit that holds the absolute value of the past peak value while subtracting (attenuating) the absolute value is provided inside the circuit 201. The peak detection circuit 201 detects the absolute value of the digital output D1 after the next serial zero-cross signal ZCR is generated using the peak hold signal output from the peak hold circuit as a threshold after the previous peak value detection. The timing of the peak value is detected when the threshold value is exceeded. The peak value timing detection is performed for each of the case where the digital output D1 has a positive sign and the case where the digital output D1 has a negative sign. Then, at the peak value detection timing, the maximum peak value detection signal MAX is output in the case of a positive sign and the minimum peak value detection signal MIN is output in the case of a negative sign. It should be noted that each of these signals is actually, of course, a time division signal for six strings.

Next, the time constant conversion control circuit 202 is a circuit for changing the attenuation rate of the peak hold circuit in the peak detection circuit 201, and the maximum / minimum peak value detection signals MAX, MIN from the peak detection circuit 201. , And the control from the central control unit (MCP, the same hereinafter) 3 in FIG. This will be described later.

Subsequently, the peak value acquisition circuit 20 in FIG.
3 demultiplexes the digital output D1 sent from the peak extraction analog circuit 1 in a time division manner into peak values for each string, and outputs a peak value detection signal from the peak detection circuit 201. The peak value is held according to MAX and MIN. Then, the MCP 3 (FIG. 1) sequentially outputs the maximum peak value or the minimum peak value of the string accessed through the address decoder 4 (FIG. 1) to the MCP 3 via the bus BUS. In addition to the peak value, the peak value capturing circuit 203 can also output the instantaneous value of vibration for each string.

The zero-cross time acquisition circuit 204 outputs the output of the time-base counter 2041 common to each string according to the serial zero-cross signal ZCR from the pitch extraction analog circuit 1 (FIG. 1), strictly speaking, at the zero-cross time of each string. , The maximum / minimum peak value detection signals MAX and MIN output from the peak detection circuit 201 are latched at the zero cross point immediately after the passage timing of the maximum peak point and the minimum peak point determined by MIN. When this latch operation is performed, the zero-cross time acquisition circuit 204 subsequently outputs the interrupt signal INT to the MCP3 in FIG. This allows MCP3
In accordance with a control signal (described later) output from the address decoder 4 (FIG. 1) via the address decoder 4, a string number at which a zero-cross occurs, a zero-cross time corresponding to the latched string and positive / negative information (described later) are stored in the bus BUS. Through MCP3
Sequentially output to.

Further, the timing generator 20 shown in FIG.
Timing signals for processing operations of the circuits shown in FIGS. 1 and 2 are output from 5. Next, returning to FIG.
The MCP 3 is a memory such as a ROM 301 and a RAM 30.
2 and a timer 303. ROM3
Reference numeral 01 is a non-volatile memory that stores programs for controlling various tones described later, and RAM 302 is a rewritable memory used as a work area for various variables and data during the control. Further, the timer 303 is used for note-off (silence) processing described later.

The tone generating section 5 includes a tone generating circuit 501 and D
The A / A converter 502, the amplifier 503, and the speaker 504 emit a musical tone corresponding to the musical tone control information from the MCP 3. An interface (Musical Instrument Digital Interface) is provided on the input side of the tone generation circuit 501.
MIDI is provided and is connected to the MCP 3 via a dedicated bus MIDI-BUS for transmitting tone control information. When the musical tone generating section 5 is provided in the guitar body, another internal interface may be used.

The address decoder 4 receives the interrupt signal IN from the zero-cross time acquisition circuit 204 (FIG. 2) described above.
After the generation of T, the zero-cross time acquisition circuit 2 is generated in accordance with the address read signal AR generated from MCP3 (FIG. 1).
To 04, a string number read signal outside 1 is supplied, and then a time read signal outside 2 (i = 1 to 6) is supplied. Also, the same

[0024]

[Outer 1]

[0025]

[Outside 2]

In this way, the waveform read signal RDAj (j = 1 to 18) is output to the peak value fetch circuit 203 (FIG. 2). Details of these operations will be described later. {Schematic operation of the present embodiment} The operation of the embodiment having the above configuration will be described below.

First, the general operation of this embodiment until the generation of a musical sound will be described. D1 of FIG. 10 is an analog representation of one string of the digital output D1 output from the pitch extraction analog circuit 1 of FIG. This waveform is obtained by picking one of the six strings of a guitar (not shown) and outputting an electric signal detected from the corresponding pickup as a digital signal. The fret (not shown) of the string is output. A waveform having a pitch period as shown in FIG. 10T _{0 to} T ₅ is generated according to the position to be pressed on the fingerboard.

In the present embodiment, this pitch period T _{0 to} T ₅
, Etc. in real time to extract the MCP of FIG.
3 generates pitch information corresponding to the pitch information, and the musical tone generating circuit 501 of FIG. 1 causes the musical tone of that pitch to be generated. Therefore, when the performer changes the tension of the strings during the performance by a tremolo arm (not shown), the pitch period of the digital output D1 changes accordingly, and the pitch information also changes in real time accordingly. , It is possible to add rich expressions to musical sounds.

Further, in this embodiment, the peak values a _{0 to} a ₃ or b _{0 to} b ₃ of the digital output D1 of FIG. 10 are detected, and especially when the MCP _{3 of} FIG. 1 rises (when picking a string). By creating volume information based on the maximum peak value a ₀ of) and transferring it to the tone generation circuit 501, it is possible to generate a tone having a volume corresponding to the picking strength of the string.

Next, in the present embodiment, when the pitch period T ₀ of FIG. 10 is extracted, the tone generation is not started immediately, but the first several pitch periods are logically judged,
When a stable pitch period can now extract, it begins pronunciation in FIG 10t _3, for example. This is because when the string is picked, the digital output D1 is easily disturbed at the start of the string, a stable pitch period cannot be extracted, and if a musical tone is produced with a pitch corresponding to the pitch period as it is, a strange musical tone is produced. This is because Therefore,
The first pitch period at the start of musical tone generation is, for example, as shown in FIG.
It is T ₁ .

Here, in a general live guitar (acoustic guitar) or the like, when a string vibration waveform when a string is picked is picked up by a pickup or the like and a pitch period is extracted from the waveform signal, picking starts, that is, string vibration. There is a tendency that a pitch period shorter than the period corresponding to the pitch when the fret of the actual guitar is pressed is extracted at the start of (the rising edge of the input waveform signal). That is, the frequency corresponding to the extracted pitch period becomes higher than the actual pitch. Figure 1
In the example of 0, the pitch period T _{1 and the} like are extracted longer than the actual period.

Therefore, in this embodiment, only the pitch period at the start of sounding, for example, T ₁ is corrected so that its value is shortened by several percent, and pitch information is generated by the correction value to generate the musical tone of FIG. Output to the generation circuit 501. Each pitch period since at the start of sounding example FIG 10T ₂ ~
For T _{5 and the} like, the corresponding pitch information is generated as it is without correction and is output to the tone generation circuit 501. As a result, even if the frequency corresponding to the pitch period is detected to be higher at the start of sound generation, it is possible to start the sound generation of the musical tone with the corrected original pitch. This is a major feature of this embodiment.

Since the above operation is time-division-processed with respect to the time-division digital output D1 for six strings of the guitar, the musical-tone generating circuit 501 can produce six strings of musical sounds aurally at the same time. Since these musical tones can be set to any volume and tone color and various effects can be added electronically, an extremely large performance effect can be obtained. {Operation of Pitch Extraction Digital Circuit} The operation of the present embodiment for realizing the above operation will be described in detail below. (Schematic Operation) First, the operation of the pitch extraction digital circuit 2 of FIG. 1 or 2 will be described. In the following description, only one string will be described, and the serial zero-cross signal ZCR, the digital output D1, and the maximum / minimum peak value detection signals MAX and MIN will be described with the image of one string. Minutes are time-shared.

In the circuit 2, for each string, the digitizer shown in FIG.
Peak value a from the output D1₀~ A ₃Or b₀~ B₃etc
And at the same time, the zero-cross time t immediately after each peak value
₁~ T₇Etc., and further immediately before each zero-cross time
Information that indicates 1 or 0 is extracted depending on whether the peak value is positive or negative.
And supplies it to MCP3 in FIG. Based on this, MC
P3 is the pitch of FIG. 10 from the interval of the zero cross time.
Cycle T₀~ T_FiveEtc., and other various music
Generates sound information, and if necessary, as described later,
Error handling, note-off (silence) processing, relative on
・ Turn off processing.(Detailed operation) Therefore, the peak detection circuit 201 of FIG.
Then, the digital output D1 input as shown in FIG.
In contrast, first, in the part where the value becomes negative, its absolute value
When x exceeds 0 x₀Then, as shown in FIG.
The small peak value detection signal MIN becomes high level.

As a result, the peak value acquisition circuit 203 of FIG.
Is the timing x ₁ immediately after the minimum peak value detection signal MIN becomes high level, and detects the minimum peak value (peak value on the negative side) b ₀ (absolute value) from the separately input digital output D 1 and Hold in a latch not shown,
At the same time, the minimum peak value detection signal MIN is returned to low level.

On the other hand, a serial zero-cross signal ZCR as shown in FIG. 10 is input from the pitch extraction analog circuit 1 of FIG. 1 to the zero-cross time acquisition circuit 204 of FIG. This signal is a signal output as a high-level or low-level binary digital signal from a comparator (not shown) in the pitch extraction analog circuit 1 which determines whether the digital output D1 is positive or negative in accordance with the determination.

In the zero-cross time acquisition circuit 204, immediately after the minimum peak value detection signal MIN output from the peak detection circuit 201 becomes high level at the timing x ₀ , the serial zero-cross signal ZCR changes the edge timing. , That is, at the time of zero crossing of the digital output D1, the time base counter 2041 of FIG.
The time t ₀ (FIG. 10) clocked at is latched. A positive or negative flag of 1 or 0 (which indicates ₀ for the minimum peak value b ₀ ) indicating whether the immediately preceding peak value is positive or negative is added to the most significant bit of this latch data. ..

Further, the zero-cross time acquisition circuit 204 continues to the above operation and interrupts the MCP3 shown in FIG.
Is output. Thus, at the time when the interrupt signal INT is generated, the minimum peak value b ₀ (absolute value) is held in the peak value acquisition circuit 203 of FIG. 2, the minimum peak value b ₀ is the zero-crossing time receiving circuits 204 The zero-cross time including the positive / negative flag immediately after the occurrence is latched.

Then, after outputting the interrupt signal INT,
By the access (described later) performed from the MCP 3 of FIG. 1 via the address decoder 4, the zero-cross time and the minimum peak value b ₀ including the positive / negative flag are transferred to the MCP 3 via the bus BUS. Since the above processing is time-divisionally processed for 6 strings, as will be described later, before output of each information described above, information indicating which string number the interrupt has occurred is acquired at the zero cross time point. Output from the built-in circuit 204 to MCP3.

Next, in the peak detection circuit 201 of FIG.
The internal peak hold circuit (not shown) is shown in FIG.
The minimum peak value b ₀ (absolute value) is held at the peak, and the peak hold signal q ₀ of FIG. 10 is output. As a result, the peak detection circuit 201 uses the peak hold signal (absolute value) as a threshold value and detects the minimum peak value again at the timing x _{2 when the} absolute value of the negative side of the digital output D1 exceeds the threshold value. The signal MIN is set to high level.

As a result, as shown in FIG.
Of the minimum peak value detection signal MI
N The following minimum peak value b ₁ (absolute value) at the timing x ₃ immediately after the high level is held at the zero-crossing time acquisition circuit 204 of FIG. 2, the positive and negative immediately after occurrence of the minimum peak value b ₁ The zero-cross time t ₂ including the flag (also 0 in this case) is latched, transferred to the MCP 3 after sending the interrupt signal INT.

Based on the above, the digital output D of FIG.
Minimum peak value b _{0 to} b ₃ (absolute value) for the negative side of 1,
The maximum peak value a _{0 with} respect to the positive side of the digital output D1 is performed in exactly the same manner as the detection of the zero-cross times t ₀ , t ₂ , t ₄ , t _6, etc., and the output operation of the peak hold signals q ₀ -q _3, etc. ~a ₃ detection, such as, the zero-crossing time t _1, t _3, t
₅ , t ₇ etc. detection and peak hold signals p _{0 to} p ₃
Output operations such as are performed in parallel. In this case,
Maximum peak value detection signal MAX from the peak detection circuit 201
Is output as shown in FIG. 10, and in the peak value acquisition circuit 203 and the zero-cross time acquisition circuit 204 of FIG.
Based on this signal MAX, maximum peak values a _{0 to} a _3, etc.,
And zero-cross times t ₁ , t ₃ , t ₅ , t _7, etc. including the positive and negative flags (in this case, 1 because it is a positive peak) are latched.

By the above-described operation, the interrupt signal INT is output from the zero-cross time acquisition circuit 204 of FIG. 2 at each time of the zero-cross times t _{0 to} t ₇ of FIG.
It is output to CP3, and b ₀ and t ₀ , a ₀ and t ₁ , b ₁ and t ₂ , a ₁ are set as a set of the minimum or maximum peak value (absolute value) and zero-cross time for each time based on this. , T ₃ , ... Are sequentially output to the MCP ₃ . Where M
In CP3, it is possible to determine whether the peak value is the minimum peak value (negative peak value) or the maximum peak value (positive peak value) by the positive / negative flag added to the most significant bit at the zero-cross time. is there.

In addition to the above operation, the peak value fetch circuit 203 of FIG. 2 can arbitrarily output the instantaneous value of the digital output D1 by accessing from the MCP3. This will be described later.

Further, the respective decay rates (time constants) of the peak hold signals p _{0 to} p ₃ , q _{0 to} q ₃ etc. of FIG. 10 generated by the peak hold circuit in the peak detection circuit 201 of FIG.
Is a time constant conversion control circuit 20 of FIG. 2 under the control of MCP3.
It is changed at any time by 2.

Basically, the time constant is changed so that the peak hold signal is rapidly attenuated after, for example, one pitch cycle time of the digital output D1. Then, the setting of the pitch period information at this time is performed by the MC of FIG.
The time constant conversion control circuit 202 via the bus BUS after each pitch period is extracted by the operation of P3 described later.
Pitch cycle information is set in a time constant conversion register CHTRR (described later) in the table. As a result, the time constant conversion control circuit 202 performs coincidence comparison between the internal counters of the respective strings (not shown) provided therein, which are not shown in particular, and the pitch cycle information set in the time constant conversion register CHTRR from the MCP 3 to compare the pitch cycle. When a coincidence output is generated as time passes, the time constant change signal is sent to the peak detection circuit 20.
Send to 1. By this operation, the peak hold signal is rapidly attenuated when the time of one pitch period elapses, whereby the peak of the next pitch period is properly detected.

If the next maximum or minimum peak value detection signal MAX or MIN is detected in the peak detection circuit 201 before one pitch period elapses, the counter is set at the falling timing of these signals. It is reset and the next peak hold signal is generated.

Further, since the vibration cycle of each string varies widely depending on the position on the fret where the string is held by the player, when the waveform corresponding to each string of the digital output D1 rises, the vibration of that waveform is promptly changed. In order to detect, the peak hold signal is rapidly attenuated at the time of the highest tone cycle of each string, and immediately after that, the time of the open string period (lowest tone period) of each string elapses so as not to pick up the overtone of each pitch period. The setting is made so that it rapidly decays. Then, after the pitch cycle is effectively extracted, the setting is made so that the pitch cycle time is rapidly attenuated, and changes in the pitch cycle of each string of the digital output D1 due to the performance operation are followed.

Further, in the pitch detection circuit 201, whether the peak hold control is performed for positive or negative peak value is determined by whether the serial zero-cross signal ZCR is high level or low level. (See FIG. 10). {Operation of central controller (MCP)} By the above operation,
Based on the maximum or minimum peak value, the zero-cross time, and the positive / negative flag indicating whether the peak value is positive or negative, which is supplied from the pitch extraction circuit 2 of FIG. 1, the MCP 3 of FIG. 1 performs pitch extraction and extraction of parameters relating to volume and tone color. By performing this, musical tone control information for controlling the musical tone generating circuit 501 is generated. The MCP 3 executes the operation flowcharts shown in FIGS. 3 to 9 according to the program stored in the ROM 301, as described in detail below. (Explanation of Variables) First, each variable used in the control program shown in the operation flowcharts of FIGS. 3 to 9 described later will be listed below.

AD: Input crest value (instantaneous value) obtained by directly reading the input waveform D1 to the pitch extraction digital circuit 2 of FIG. 1 AMP (0, 1) ... Positive or negative previous (old) wave High value (peak value) AMRL1 ... Previous amplitude value (peak value) for checking relative off (off) stored in the amplitude register. Here, the relative off corresponds to a muffling process when the fret operation is stopped and the string moves to an open string by silencing the cue value when the crest value is rapidly attenuated.

AMRL2 ... The amplitude value (peak value) of the previous two times for the relative off stored in the amplitude register, and the value of AMRL1 is input to this.

CHTIM ... Cycle corresponding to highest tone fret (22nd fret) CHTIO ... Cycle corresponding to open string fret CHTRR ... Time constant conversion register, and time constant conversion control circuit 202 (see FIG. It is provided inside 2).

DUB ... Flag indicating that the waveforms have successively come in the same direction, FOFR ... Relative off counter, HNC ... Waveform number counter MT ... Flag on the side from which pitch extraction is to be performed (positive =
1, Negative = 0) NCHLV ... No change level (constant) OFTIM ... Off time (equivalent to the open string period of the string in question) OFPT ... Normal off check start flag ONF ... Note on flag RIV・ ・ ・ Flag for switching the processing route in step (STEP) 4 which will be described later ROFCT ・ ・ ・ Constant that determines the number of times of checking of relative off STEP ・ ・ ・ Register for specifying the flow operation of MCP3 (1-5) Take a value) TF ... Previous zero-cross time data that became valid TFN (0, 1) ... Previous zero-cross time data immediately after the positive or negative peak value TFR ... Time storage register THLIM ... Frequency upper limit (constant) TLLIM ... Frequency lower limit (constant) TP (0, 1) ... Positive or negative previous cycle data T LAB (0, 1) ... Positive or negative absolute trigger level (note-on threshold) TRLRL ... Relative-on (re-production start) threshold TRLRS ... Resonance elimination threshold TTLIM ... -Lower frequency limit at the time of trigger TTP ... Cycle data extracted last time TTR ... Cycle register, TTU ... Constant (product of 17/32 and this cycle information tt) TTW ... Constant (31/16) And the current cycle information tt) VEL ... Information that determines the velocity (velocity), and is determined by the maximum peak value of the waveform at the start of sounding.

X: Flag indicating an abnormal or normal state b: Current positive / negative flag stored in the working register B (1 at the zero point next to the positive peak, at the zero point next to the negative peak) 0) c ... Current peak value (peak value) stored in working register C e ... Pre-previous peak value (peak value) stored in working register E h ... Stored in working register H The cycle data extracted before two times t ... the current zero-cross time stored in the working register T0 tt ... the current cycle information stored in the working register TOTO (operation of the interrupt processing routine) Figure 3 shows MCP
3 is a diagram showing an operation flowchart of an interrupt processing routine showing a process when an interrupt is issued by the interrupt signal INT from the zero-cross time acquisition circuit 204 (FIG. 2) in the pitch extraction digital circuit 2.

As described above, at the time when the interrupt signal INT is output from the zero-cross time acquisition circuit 204, the maximum or minimum peak value (absolute value) is held in the peak value acquisition circuit 203 of FIG. The zero-cross time acquisition circuit 204 has a positive / negative flag indicating 1 when the zero-cross time immediately after the peak value is generated and the immediately preceding peak value is the maximum (positive) peak value, and 0 when the minimum (negative) peak value. Is latched.

Therefore, the MCP 3 first sets a predetermined address read signal AR to the address decoder 4 at I1 in FIG. 3, and the zero cross time acquisition circuit 204 in FIG.
In response, a string number reading signal other than 3 is output.
From this circuit 204

[0057]

[Outside 3]

First, the string number indicating which string number the interrupt has occurred is M via the bus BUS.
It is output to CP3. Subsequently, the MCP 3 sets another address read signal AR in the address decoder 4, and instructs the zero-cross time acquisition circuit 204 to output a signal corresponding to the above-mentioned string number among the time read signals outside 4 to outside 5.

[0059]

[Outside 4]

[0060]

[Outside 5]

Is output. As a result, the circuit 204 outputs the above-mentioned time reading signal 6

[0062]

[Outside 6]

The zero-cross time information set in the latch corresponding to the string number designated by (i = 1 to 6) is output to the MCP 3 via the bus BUS.
This is set as the current zero-cross time t as shown in FIG. 3I1.

Subsequently, in I2 of FIG. 3, as described in the section "Operation of the pitch extraction digital circuit",
The positive / negative flag added to the most significant bit of the zero-cross time information is taken out and is set as the positive / negative flag b this time.

Thereafter, at I3 in FIG. 3, the MCP 3 outputs the peak value read signal outside 7 to the peak value fetch circuit 203 in FIG.

[0066]

[Outside 7]

(J = 1 to 12) is output. Here, although not particularly shown in the circuit 203, since there are 12 latches for holding the maximum peak value and the minimum peak value for 6 strings, the MCP 3 is based on the string number and the positive / negative flag b. , And select and output the above-mentioned peak value read signal outside 8. This allows the same circuit 20

[0068]

[Outside 8]

From 3 on, it is set to the latch designated by the signal 9 for reading the peak value.

[0070]

[Outside 9]

The maximum peak value or minimum peak value (absolute value) is output to the MCP3 via the bus BUS. This is set as the current peak value c, as indicated by I3 in FIG. After the above operation, at I4 of FIG. 3, the values of t, c and b obtained as described above are set in the registers T0, C and B (not shown) in the MCP3. Each time the interrupt process is performed, the zero-cross time information, the peak value information (absolute value), and the positive / negative flag information indicating the type of peak are written as one set in this register. In the routine, the information for each string is processed.

There are six registers T0, C, and B corresponding to the six strings, and the tone control processing described below with reference to FIGS. 4 to 9 is performed in a time-division manner for all six strings. However, for the sake of simplicity, the processing for one string will be described below. (Operation of Main Routine) FIG. 4 is an operation flowchart showing the processing of the main routine. Here, initialization (initialization) after power ON, note-off (silence) processing of musical sound, and STEP0 to STEP4 (or 5)
The process of selecting each process is performed. In the present embodiment, the tone control process is performed by the concept of step, as described later, and as described later, STEP0 → STEP1 → S.
TEP2 → STEP3 → STEP4 (→ STEP5) →
Tone control is performed in the order of STEP0. << Basic Operation >> In FIG. 4, first, by turning on the power (turning on the power), various registers and flags are initialized in M1, and the register STEP is set to 0. Further, in this case, as described in the explanation of the time constant conversion control circuit 202 (FIG. 2) in the section “Operation of the pitch extraction digital circuit”, in the initial state, the peak detection circuit 201 (FIG. 2) outputs the digital output D1. MCP3 so that the vibration at the rising edge of the waveform of
Via the bus BUS to the time constant conversion register CHTRR in the time constant conversion control circuit 202, the highest tone fret period CH
A peak hold signal output from the peak hold circuit in the peak detection circuit 201 with TIM set (see FIG. 10).
p ₀ or q ₀ ) is controlled so as to rapidly decay at the elapse of the highest tone cycle time.

Subsequently, in M2 of FIG. 4, it is judged whether or not the register described in the above-mentioned "operation of the interrupt processing routine" is empty, and if NO (hereinafter referred to as NO), M is determined.
In step 3, the contents of the registers B, C and T0 are read.
Next, in M4, some values of the register STEP are judged, and in M5, STEP0 and in M6, STEP is determined.
1, M7, STEP3 in M2, STEP3 in M8, and STEP4 in M9. In addition, the update to the next step is performed in each of STEP0 to STE as described later.
It is performed in the process of P4. << Note-off operation >> If the buffer is empty in M2, that is, if yes (hereinafter referred to as YES), M10
The process proceeds to the process from M19 to M19 where the normal note-off algorithm process is performed. This note-off algorithm is an algorithm for performing note-off when the peak value of the digital output D1 (FIG. 1) is below the OFF level for a predetermined off-time.

First, in M10, it is determined whether STEP = 0 or not. If YES, it is not necessary to perform note-off because it is an initial state in which no musical tone is generated, and the process returns to M2. On the other hand, in the case of NO, the routine proceeds to M11.

At M11, the digital output D at that time
The input peak value (instantaneous value) AD of 1 is directly read. This is because the MCP 3 outputs the peak value read signal to the peak value fetch circuit 203 (FIG. 2) via the address decoder 4.
Either 0 to 11

[0076]

[Outside 10]

[0077]

[Outside 11]

By giving this, the circuit 203 outputs the current instantaneous value of the digital output D1 to M via the bus BUS.
This can be achieved by outputting to CP3. And this value A
It is determined whether or not D is equal to or lower than the preset off level. If NO, note-off is not necessary, and therefore the process returns to M2, and if YES, the process proceeds to M12.

At M12, it is judged whether or not the previous input crest value AD is below the off level, and if NO, then M1
7, the timer 303 in MCP3 is started, and M2
Return to. Then, the next time this process is performed again, M1
2 becomes YES, so the process proceeds to M13, where the timer 3
It is determined whether the value of 03 is the off-time OFTIM. As the off-time OFTIM, for example, the open string fret cycle CHTIO of the string being processed is set. If NO in M13, the process returns to M2 and the process is repeated. If YES, the process proceeds to M14 and the register STEP
0 is written in and the highest tone fret period CHTIM is set in the time constant conversion register CHTRR, and then the process proceeds to M16 via M15 (described later). That is, when the level of the digital output D1 is attenuated, it can be determined that the digital output D1 is not input and the string is no longer plucked when the input peak value AD below the off level continues for a time corresponding to the off-time OFTIM. The process of note-off is proceeded to.

In M16, the MCP3 causes the tone generation circuit 50 to
1 (FIG. 1), a note-off instruction is sent, and the tone generation is stopped. When the note-off is made in this way, the process always returns to STEP0.

In step M15, a YES determination is made in a normal state, but the register STEP has a value other than 0 even if the musical tone is not instructed by a process described later. (For example, due to noise input), in such a case M1
By returning to M2 after the processing of 4, M15, the initial setting is made to STEP0. (Processing Operation in STEP 0) Next, details of each routine for branching and performing the corresponding processing in the main routine of FIG. 4 will be described.

First, FIG. 5 shows M of the main routine of FIG.
The operation flow of the processing of step 0 (STEP 0) shown as 5
It is a row chart. In this process, pitch extraction
Initial setting for processing, etc., and transition to the next STEP1
Perform processing. Explanation will be given below using the basic operation explanatory diagram of FIG.
I do. Note that FIG. 11 has the same waveform as FIG. "basic action" Now, the main routine of FIG. 4 is M2 and M1.
By repeating the 0 loop,
Pitch extraction as described in the section "Chin's movement"
Interrupts from the digital circuit 2 (Fig. 1)
Waiting for data to be input to registers T0, C, B
It

Then, when data is input and the contents of the above registers are read through M2 to M3 in FIG. 4, M
4 through M5, that is, STEP 0 of FIG. In this state, for example, as shown in FIG. 11, the current zero-cross time t = t ₀ , the current positive / negative flag b = 0, the current peak value c is the minimum peak value from b = 0, and c = b ₀ (absolute value).
Is. In FIG. 11, b and b _{0 to} b ₃ are different symbols.

First, in S01 of FIG. 5, it is determined whether or not the value of the current peak value c is higher than the absolute trigger level (positive threshold value for note-on) TRLAB (b). Note that this determination is executed for each of positive and negative polarities (maximum peak value or minimum peak value) based on the value of the positive / negative flag b this time, and the positive absolute trigger level TRL is set.
Ab (1) and negative side absolute trigger level TRLAB (0)
Can be empirically set to different values in consideration of a case where an offset is superimposed on the digital output D1 (FIG. 1). The same value may be used in an ideal system. Figure 1
In the example of No. 1, this time the minimum peak value c = b ₀ (absolute value) and T
RLAB (b) = TRLAB (0) is compared, c =
b ₀ > TRLAB (0), that is, the determination is YES.

Next, after passing through S02 (described later), S0
Process 3 is executed. Here, first, the present positive / negative flag b is written in the flag MT, and 1 is written in the register STEP.
Is written to prepare for transition to the next step, and the zero-cross time t of this time is set as the previous zero-cross time data TFN (b) for subsequent processing. In the example of FIG. 11, MT = b = 0, as shown in FIG.
TFN (b) = TFN (0) = t = t ₀ .

Then, in S04, "Explanation of variables"
Other flags (excluding constant values) other than the above-mentioned flags shown in the section (4) are initialized. Further, in S05, the current peak value c is the previous peak value AM for the subsequent processing.
It is set as P (b) (absolute value), and the process returns to M2 of the main routine of FIG. In the example of FIG. 11, AMP (b) = AMP (0) = c = b ₀ as shown in the figure.

With the above processing, in the example of FIG. 11, the current positive / negative flag b = 0 of the register B is written in the flag MT as shown in the same figure (between STEP 0 → 1), and the previous zero-cross time on the negative side is written. Register TFN (0) to register T
The current zero-cross time data t = t _{0 of 0} is written, and the current minimum peak value c = b ₀ of the register C is written in the previous negative peak value AMP (0). << Resonance Removal Operation >> In S01 of FIG. 5, if the value of the current peak value c is equal to or less than the absolute trigger level TRLAB (b), the process of sound generation (note-on) does not proceed, and S
In 05, the current peak value c is simply set to the previous peak value AMP (b), and the process returns to the main routine of FIG. However, when picking one string causes another string to resonate, the vibration level of the other string gradually increases, and eventually the determination result of S01 in FIG. 5 becomes YES, The process moves to S02. However, in such a case, since the regular picking is not performed, it is not appropriate to shift to the operation of sounding (note-on). Therefore, the resonance is removed in the process of S02. That is, in the above case, since the current peak value c is not substantially larger than the previous peak value AMP (b), the difference c-AMP (b) is not larger than the resonance removal threshold TRLRS. Has
When it is determined that the resonance state has occurred, the process does not proceed to the tone generation process, and only the current peak value c is set to the previous peak value AMP (b) in S05, and the process returns to the main routine of FIG. On the other hand, when normal picking as shown in FIG. 11 is performed, the waveform rises sharply, and the peak value difference c-AMP (b) becomes the resonance elimination threshold value TR.
When the LRS is exceeded, the process proceeds from S02 to S03 as described above. << Relative-on Entry Operation >> In FIG. 5, A is a relative-on (start of re-sound generation) entry described later, and jumps to S06 from the flow of STEP4 described later. Then, in S06, the musical tone that has been output so far is once erased (note-off), and the process proceeds to S03 in order to start re-generation. The process for starting re-sounding is similar to that at the start of normal sounding and is as described above. Here, note-off processing of S06 is performed by the above M of FIG.
This is the same as the processing of 16. (Processing Operation of STEP1) Next, FIG. 6 is an operation flowchart of the processing of step 1 (STEP1) shown as M6 of the main routine of FIG. In this process, the initial setting for the pitch extraction process following STEP0 and the subsequent transition process to STEP2, or the double process (error process) when a strange waveform is input are performed. << Basic Operation >> First, after the initial setting for the first data is performed in STEP0, in the main routine of FIG. 4, the pitch extraction digital circuit 2 (FIG. 4) is repeated by repeating the loop of M2 → M10 → M11 → M2. The interrupt is applied again from 1) and waits for the next data to be input to the registers T0, C and B.

Then, when data is input and the contents of the above registers are read through M2 to M3 in FIG. 4, M is read.
4 through M6, that is, STEP 1 of FIG. In this state, for example, as shown in FIG. 11, the current zero-cross time t = t ₁ , the current positive / negative flag b = 1, and the current peak value is the maximum peak value c = a _{0 since} b = 1.

First, through S11 of FIG. 6 (which will be described later), in S12, the value of the peak value c at this time is exactly the same as that described in S01 of FIG. 5 in the section "Processing operation of STEP0". But the absolute trigger level TRLA
It is determined whether or not it is larger than B (b). In the example of FIG. 11, the maximum peak value c = a ₀ and TRLAB (b) = T this time.
RLAB (1) is compared, and c = a ₀ > TRLAB
(1) That is, the determination is YES.

Next, in step S13, 2 is written in the register STEP to prepare for transition to the next step. Subsequently, in S14, the current zero-cross time t of the register T0 is set as the previous zero-cross time data TFN (b) for the subsequent processing. Further, in S15, the current peak value c of the register c is set as the previous peak value AMP (b) for the subsequent processing, and the process returns to the processing of M2 of the main routine of FIG. In the example of FIG. 11,
As shown in the figure, TFN (1) = t = t ₁ , AMP
(1) = c = a ₀ . The contents of MT are not rewritten and remain 0. << Dubliation Processing Operation >> When the normal digital output D1 as shown in FIG. 11 is input, after the negative (positive) side minimum (large) peak value (absolute value) is extracted in STEP0, , STEP 1 oppositely, the maximum (small) peak value on the positive (negative) side is extracted. Therefore, S11 in FIG.
In this case, the positive / negative flag b = 1 (0) this time is STEP0.
Since it is different from the flag MT = 0 (1) set in step S1, the process proceeds to step S12 as described above.

However, in some cases, a waveform as shown in FIG. 12 (a) or 12 (b) may be input in STEP1 after STEP0. In this case, after the minimum peak value b ₀ of the negative side it is extracted with STEP0, the minimum peak value b ₁ again negative in STEP1 is extracted I dub. Therefore, FIG.
In S11 of, the positive / negative flag this time becomes b = 0,
This matches the flag MT = 0 set in STEP0.
In this case, the process proceeds to S16 of FIG. 6 and the duplication process (error process) is performed.

In S16, it is determined whether or not the value of the peak value c is larger than the previous peak value AMP (b) of the same sign. Now, in the case of FIG. 12 (a), c = b ₁ > AMP
(B) = AMP (0) = b ₀ does not hold. In such a case, the minimum peak value b ₁ of this time is ignored as a strange waveform (hatched portion), STEP is not updated, the process returns to the process of M2 of the main routine of FIG. 4, and the next normal peak is input. Wait to be done.

On the other hand, in the case of FIG. 12 (b), c = b₁
> APM (b) = AMP (0) = b ₀Holds. this
In such a case, the minimum peak extracted in the previous STEP 0
Value b ₀Is ignored as a strange waveform (shaded area), S
Last negative cross set on TEP0
Time data TFN (0) and the previous peak value A on the negative side
This time, the contents of MP (0) will be changed by S14 and S15 in FIG.
The zero cross time t and the current peak value c
To change. That is, in the example of FIG. 12 (b), TFN (0) = t
= T₁, AMP (0) = c = b₁Becomes This dubli
After that, STEP is not updated (do not go through S13 in FIG. 6).
4) and returns to the process of M2 of the main routine of FIG.
Wait for a normal peak to be entered.

After the above operation, when a normal peak value is input, the above-described processing is performed by S11 → S12 → S13 → S14 → S15 in FIG. 6, for example, at t = t ₁ as shown in FIG. Then, the processing shifts to the processing of STEP2. (Processing Operation of STEP2) Next, FIG. 7 is an operation flowchart of the processing of step 2 (STEP2) shown as M7 in the main routine of FIG. In this process, the first pitch period detection for pitch extraction,
Velocity setting and transition processing to STEP 3, or error processing (double processing) when a strange waveform is input are performed. << Basic Operation >> First, after the processing in STEP1 is performed, in the main routine of FIG. 4, M2 → M10 → M1
By repeating the loop of 1 → M2, the pitch extraction digital circuit 2 (FIG. 1) again interrupts and waits for the next data to be input to the registers T0, C, and B.

Then, when data is input and the contents of the above registers are read through M2 to M3 in FIG. 4, M
M4, that is, STEP 2 in FIG. In this state, for example, as shown in FIG. 11, the current zero-cross time t = t ₂ , the current positive / negative flag b = 0, and the current peak value is the minimum peak value c = b _{1 since} b = 0.

First, after passing through S20 of FIG. 7 (described later)
In S21, the MCP 3 sets the open string fret cycle CHTIO of the currently processed string in the time constant conversion register CHTRR in the time constant conversion control circuit 202 of FIG. 2 via the bus BUS. This is the time constant conversion control circuit 20 in the section "Operation of the pitch extraction digital circuit".
As described in the description of No. 2, the peak detection circuit 201
After (Fig. 2) detects the vibration at the rising edge of the waveform of the digital output D1, the peak hold signal output from the peak hold circuit in the peak detection circuit 201 (Fig. 2) so as not to pick up the overtone of each pitch period. 10p ₁ , q ₂ etc.)
Indicates that the open string period of each string, that is, the minimum tone period CHTIO, is rapidly attenuated.

Next, in S22, it is determined whether or not the value of the current peak value c is larger than 7/8 times the previous peak value AMP (b) of the same sign. This processing will be described in detail later, but since the waveform picking the string normally attenuates smoothly and naturally, this determination becomes YES and the next S
After S23, the process proceeds to S24 (described later).

At S24, {(current zero-cross time t)-(previous zero-cross time data TFN of the same code)
(B))} is calculated to detect the first pitch period. Then, this result is referred to as STEP 3 described later.
It is set as the previous cycle data TP (b) in order to be used as a condition of note-on (start of sound generation) in. 11
In the above example, TP (0) = t-TFN as shown in FIG.
(0) = t ₂ −t ₀ .

At S24, the zero-cross time t of this time is set to the previous zero-cross time data T for the subsequent processing.
It is set as FN (b). In the example of FIG. 11, TFN (0) = t = t ₂ as shown in the figure. In addition, ST
The TFN (0) = t ₀ set in EP0 is erased as unnecessary because the previous cycle data TP (b) = TP (0) can be calculated.

Similarly, in S24, 3 is written in the register STEP to prepare for transition to the next step. Further, in S24, the current peak value c and the previous peak values AMP (0) and AMP (1) are set for the subsequent processing.
Among them, the largest value is set as the velocity VEL. The velocity VEL is used as a value for determining the volume of the musical sound as described later in STEP3. Similarly, the current peak value c is changed to the previous peak value AMP.
It is set as (b), and the process returns to the process of M2 of the main routine of FIG. In the example of FIG. 11, VEL = max {c, AM
P (0), AMP (1)} = max {b ₁ , b ₀ ,
a ₀ }, and AMP (0) = c = b ₁ . In addition,
AMP (0) = b ₀ set in STEP ₀ is no longer necessary and is erased because the velocity VEL can be calculated. << Dubliation Processing Operation >> When the normal digital output D1 as shown in FIG. 11 is input, after the positive (negative) side maximum (small) peak value is extracted in STEP1,
On the contrary, in STEP 2, the minimum (large) peak value on the negative (positive) side is extracted. Therefore, the sign of the peak value in STEP2 in this case is opposite to that in STEP1, and is the same as that in STEP0. In S20 of FIG. 7, the positive / negative flag b = 0 (1) is set in STEP0 this time. Flag MT = 0 (1), the process proceeds to S21 as described above.

However, the above-mentioned "STEP1 processing operation"
In the same manner as described in the description of the section “Operation of the double processing”, the waveform may be doubled in some cases, and the waveform as shown in FIG. 13 or 14 may be input after STEP1. In this case, after the maximum peak value a ₀ on the positive side is extracted in STEP 1, the maximum peak value a ₀ on the positive side is again extracted in STEP 2.
₁ is dub and extracted. Therefore, in S20 of FIG. 7, the positive / negative flag this time becomes b = 1, which coincides with the flag MT = 0 set in STEP0. In this case, the process proceeds to S25 of FIG. 7 and the duplication process (error process) is performed.
It should be noted that peaks hatched with simple diagonal lines in FIGS. 13 and 14 are peak hold signals p ₀ , p ₁ and q of FIG. 13 or 14 generated from the peak hold circuit in the peak detection circuit 201 of FIG. _This is the part that was not detected as a peak because it was not caught at ₀ mag.

In S25, first, the double flag DUB is set to 1 (described later), and then the process proceeds to S26, in which it is determined whether or not the value of the current peak value c is larger than the previous peak value AMP (b) of the same sign. To be done.

Now, referring to FIG. 13, STEP0 (t = t
_0), STEP1 (after t = t ₁₎ of the process, if the STEP2 in t = t ₂ is executed, c = a _1> AMP
(B) = AMP (1) = a ₀ does not hold. That is, FIG.
The determination result of S26 is NO. In this case,
Ignore the maximum peak value a ₁ this time as a strange waveform (the cross hatched area in the figure),
Does not update and returns to the process of M2 of the main routine of FIG. 4 and waits for input of the next normal peak. And
At t = t ₃ , the minimum peak value c = b ₁ is input, so that S20 of FIG. 7 becomes YES, and the processing of S21 → S22 → S23 → S24 is performed as in the case of FIG. At t = t ₃ in FIG. 16, the process proceeds to the next STEP3. The previous cycle data TP (0) set in S24 of FIG. 7 is the difference between the current zero-cross time t ₃ and the previous zero-cross time t ₀ set in STEP 0, as shown in FIG. In addition, S described later
The starting point of the next period data Tx calculated in TEP3 is the peak (c = a ₁ ) hatched with cross hatched lines as shown in the figure, so the previous zero-cross time set in STEP 1 is ignored. TFN (1) = t
_{Is 1} .

On the other hand, in the case of FIG. 14, conversely to the above, c = a
₁ > AMP (b) = AMP (1) = a ₀ holds. That is, the determination result of S26 in FIG. 7 is YES. In such a case, the maximum peak value a extracted in the previous STEP1
Ignoring ₀ as a strange waveform (the hatched portion of the cross hatched in the figure), the previous zero-cross time data TFN (1) set in STEP1 and the previous peak value AMP (1) on the positive side. The contents of S2 in FIG.
9, the zero crossing time t of this time and the peak value c of this time
Replace with and change. That is, in the example of FIG. 14, TFN (1) = t = t ₂ and AMP (1) = c as shown in FIG.
= A ₀ . After this duplication processing, STEP returns to the processing of M2 of the main routine without updating, and waits for the input of the next normal peak value. Hereinafter, the processing after the minimum peak value c = b ₁ is input at t = t ₃ is the same as in the case of FIG. However, the peak extracted in STEP 1 (peak c = a ₀ hatched with cross hatched lines in FIG. 14) is ignored and changed to the peak of c = a ₁ , so TP calculated in STEP 3 described later The starting point of the period data T _y next to (0) is ST
The previous zero-cross time TFN (1) = t ₂ set in the _double processing of EP2 is set, which is different from the case of FIG.

As described above, when the waveform is doubled as shown in FIG. 13 or 14, the peak with the smaller peak value is ignored as a strange waveform and error processing is performed. Next, the branch of S22 of FIG. 7 for the processing of other cases of the duplication processing will be described.

When the processing of STEP2 of FIG. 7 is executed, the normal waveform picking the strings naturally attenuates smoothly. Therefore, in S22, the current peak value is the same as the previous peak value AMP (b ) Is greater than 7/8 times, the determination in S22 is YES and the next S23
Proceed to.

However, in some cases, c> (7/8) × A
MP (b) may not hold. In the first case, for example, by picking a string near the bridge, the waveform may not be a normal but smooth decay waveform, and the determination result of S22 may be NO.
However, even in such a case, S24 in FIG.
Must be processed normally. Then, in this case, since the waveform is normal, the above-described duplication does not occur, and since the branch from S20 to S25 in FIG. 7 has not been performed before that, the value of the duplication flag DUB remains 0. .. Therefore, if DUB = 1 is not established in S27 of FIG. 7, S is re-established regardless of the determination result of S22.
Returning to the processing of 24, the processing described in the above "basic operation" is performed. The double flag DUB is the ST of FIG.
In the process of S04 of EP0, the value is initialized to 0.

On the other hand, as a second case in which S22 of FIG. 7 is not established, there is a case where the above-mentioned duplication occurs in the waveform. This case will be described below with reference to FIG.

Now, as in the case described with reference to FIG. 14, FIG.
As shown in, STEP0 (t = t ₀), STEP1
(T = t₁), T = t₂At the dub
Re-processing is performed, c = a₀Peak (cross diagonal line in the figure)
Hatched peaks) are removed and c = a₁The pea
Ku (peaks with vertical lines in the figure) left
To do. The peaks with a simple hatching (c =
a₁) Is not originally detected as in FIG. 13 or FIG.
It is a peak.

When the duplication occurs as described above, the next t
= T ₃ , the positive / negative flag is b =
Since it becomes 0, the flag MT =
Matches 0. Therefore, the process proceeds from S20 to S21 of FIG. 7 to S22. However, the current minimum peak value c = b ₁ detected at t = t ₃ is the same as the previous minimum peak value AMP (0) =
It is far away from b ₀ and has large attenuation. Therefore, as shown in FIG. 15, the determination result of S22 of FIG. 7 may be NO.

In the above case, t = t ₂ before that
The dubli flag D
The value of UB is 1. Therefore, the determination result of S27 of FIG. 7 is NO, and the process proceeds to S29 via S28 (described later).

In S29, in order to acquire a normal waveform after t = t ₃ in FIG.
The previous zero-cross time data TFN (0) set in STEP0 and the previous peak value AMP on the negative side
The contents of (0) are changed by replacing the current zero-cross time t and the current peak value c in S29 of FIG. That is, in the example of FIG. 15, TFN (0) =
Since t = t ₃ and AMP (0) = c = b ₁ , the hatched peak (c = b ₀ ) of the horizontal line in the figure is ignored. Note that the double flag DUB is reset to 0 in S28 of FIG. 7 for the subsequent processing. After these operations, the STEP value is not updated and the process returns to the process of M2 of the main routine of FIG. 4 to wait for the input of the next peak.

In the above case, as shown in FIG. 15, at t = t ₄ and t = t ₅ , STEP 2 of FIG. 7 is repeated, and then the process proceeds to STEP 3. Such STE
Since there are various patterns for the repeated operation of P2, a detailed description thereof will be omitted, but a normal waveform can be obtained as a whole flow, and the next STEP3
TFN (0), AMP for use in
After operating so that (0), TFN (1), and AMP (1) are effectively determined, the process proceeds to STEP3. In the case of FIG. 15, TP (0) = t ₅ −t ₃ , and the starting point of the next cycle data Ti calculated in STEP 3 described later is TFN (1) = t ₄ . (Processing Operation of STEP3) Next, FIG. 8 is an operation flowchart of the processing of step 3 (STEP3) shown as M8 in the main routine of FIG. In this processing, note-on (start of sound generation) processing, extraction of pitch cycle for pitch setting at note-on, correction of the extracted pitch cycle according to the present invention, calculation of velocity, ST
Processing for shifting to EP4 and error processing when a strange waveform is input are performed. << Basic Operation >> First, after the processing in STEP3 is performed, in the main routine of FIG. 4, M2 → M10 → M1
By repeating the loop of 1 → M2, the pitch extraction digital circuit 2 (FIG. 1) again interrupts and waits for the next data to be input to the registers T0, C, and B.

Then, when data is input and the contents of the above registers are read through M2 to M3 in FIG. 4, M
4 through M8, ie, STEP 3 in FIG. In this state, for example, as shown in FIG. 11, the current zero-cross time t = t ₃ , the current positive / negative flag b = 1, and the current peak value is the maximum peak value c = a _{1 since} b = 1.

First, through S30, S31 and S32 of FIG.
After that (described later), the velocity VEL is set in S33.
Calculate Now, in "Basic operation of STEP 2", "Basic"
As described in the section of “Operation”, in S24 of FIG.
The peak value of the last three times, b in the example of FIG.₀, A₀, B₁
The maximum of each value (absolute value) of Velocity VEL
It is stored in. Therefore, in S33 of FIG.
Judges the greater of the lossy VEL and the current peak value c
Then, a musical tone is generated by the musical tone generating circuit 501 (FIG. 1).
New velocity VEL In the example of FIG.
Is VEL = a ₀, C = a₁Therefore, VEL = max [a
₀, A₁] = A₀Becomes

After the above operation, MT ← b is set in S33 of FIG. 8 (described later), and then in S34.
The pitch cycle is detected by calculating {(current zero-cross time t)-(previous zero-cross time data TFN (b) with the same sign)}, and the previous cycle data TP (b)
Set as. In the example of FIG. 11, as shown in FIG.
P (1) = t ₃ −t ₁ .

Then, after passing through S35 to S38 of FIG. 8 (described later), at S39, the previous cycle data TP (b) obtained at S34 and the TP (set at S24 of FIG. 7 are set. b) is almost the same as the previous cycle data 12 of the opposite polarity.

[0118]

[Outside 12]

It is determined whether or not Then, if the determination result is YES, it is determined that the pitch period has started to be stably extracted, and after S301 (described later),
In step S302, note-on processing (start of sound generation) is performed. In the example of FIG. 11, the previous cycle data TP (1) on the negative side
= T ₃ −t ₁ and the previous cycle data TP (0) on the positive side = t ₂
It is determined that −t ₀ is almost the same, and the process proceeds to the note-on process. The case where the determination result is NO will be described later.

Next, in S302, S33 in FIG.
Volume information is generated on the basis of the velocity VEL calculated in step 1, and the previous pitch period TP (b) is set to 1.02.
The corresponding pitch information is generated based on the correction value obtained by the multiplication. That is, the pitch period is corrected to be 2% longer. As described above, it is a great feature of this embodiment that pitch information is generated after the pitch period is corrected at the start of sound generation, and as a result, as described in the section "Outline of operation of this embodiment". , It is possible to prevent the pitch period from being extracted at the beginning of sound generation and the pitch information of a higher pitch to be specified. The volume information and the pitch information generated by the above processing are output to the musical tone generation circuit 501 via the MIDI-BUS and the interface MIDI of FIG. Then, in the circuit 501, musical tones having a volume and a pitch corresponding to the above information are generated in real time.

Since the processing of FIG. 8 is performed for each of the 6 strings of the guitar, the above processing is also performed independently for each of the 6 strings. With the above note-on processing,
After setting the parameters used in the next STEP4 in S38 and S301 of FIG.
After 02, the process returns to M2 of the main routine of FIG.
The process proceeds to the next STEP4. That is, in S38, S3
The previous cycle data TP (b) extracted in 4 is set as the cycle data TTP previously extracted, and the previous zero-cross time data set in S24 of STEP2 of FIG.

[0122]

[Outside 13]

This time is set in TFR, the current zero-cross time data t is set as the last valid zero-cross time data TF, the waveform number counter HNC is cleared to 0, the value of the register STEP is updated to 4, and the note The ON flag ONF is set to 2 (sound generation state), the constant TTU is set to 0 (minimum MIN), the constant TTW is set to maximum MAX, and the previous amplitude value AMRL1 for the relative off check is cleared to 0. It Each of these parameters will be described later in STEP 4. << Operation in Case of Improper Cycle >> When the previous cycle data TP (b) is detected in S34 of FIG. 8, this pitch cycle has a value larger than the cycle when the corresponding string is played at the highest fret. And has a value less than the open string period of the string.

Therefore, as a constant called the upper frequency limit THLIM, a pitch period of a half tone above the pitch determined by the highest fret of the string currently being processed is set, and as a constant called the lower frequency limit TTLIM, the same string is set. Set a pitch cycle of about 5 semitones below the pitch determined by the open string state of S.
36, S37, the previous cycle data TP (b) obtained in S34 is larger than THLIM, and TTLI
It is determined whether or not it is smaller than M. Then, if both of the above determination results are YES, the process proceeds to S39 and the cycle determination process described above is performed.

If the determination results in S36 and S37 are NO, the previous cycle data TP (b) extracted in S34 is not an appropriate value. Therefore, in such a case, from S36 or S37 to M2 of the main routine of FIG.
The process returns to step 3 and STEP 3 is repeated.

Next, in S39 of FIG. 8, when the previous cycle data TP (b) obtained in S34 and the previous cycle data 14 having a different polarity are different from each other,
Overtone

[0127]

[Outside 14]

It is highly possible that the pitch period is not extracted in a stable manner due to the fact that the pitch period has been extracted incorrectly by mistakenly extracting the pitch period. Therefore, in such a case, S
The determination result in 39 is NO, the process returns to M2 of the main routine of FIG. 4 and STEP3 is repeated.

Here, when STEP3 is repeated by the above operation, in a normal waveform, the peaks newly detected via M2 and M3 in FIG. 4 switch their polarities alternately and the value of b is 0. Alternately reversed at 1, and as shown in FIG.
In S33, the value of the flag MT is alternately changed, and similarly in S34, TP (b) is newly calculated and TF (b) is calculated.
The contents of N (b) are also rewritten. Therefore, S36, S
The determination of 37 is performed on the pitch cycle that is obtained most recently, and further, the determination at S39 is the pitch cycle that is obtained most recently and the pitch cycle on the opposite polarity side that is one cycle before (about half a cycle before). When the pitch period can be extracted stably, the process shifts to the note-on process.

In each case, in S33 of FIG.
The velocity VEL is updated to correspond to the newly detected peak. << Operation of Noise Removal Process >> The process of S31 in FIG. 8 is a process for coping with the case where noise occurs at the rising portion of the waveform. Now, for example, as shown in FIG.
In EP0, ₁ , ₂ , the peak a ₀ due to noise,
When b ₀ , a _1, etc. are detected, if a cycle of these noises is detected and an instruction to start sound generation is given, a completely strange musical sound is sounded.

Therefore, in S31 of FIG. 8, if a continuous peak value changes greatly, it is determined that noise is occurring and the abnormality detection flag X is set to 1, and S
By making a NO determination at 35, note-on is prevented based on the noise portion.

Specifically, if the value obtained by ⅛ of the current peak value c is smaller than the previous peak value AMP (b) having the same sign as that, it is judged to be normal, and X = 0 is set. Let X = 1. Then, when it is determined in S35 that X = 0 is not satisfied, M2 of the main routine of FIG.
The process returns to step 3 and STEP 3 is repeated. In this case, since the previous peak value AMP (b) is sequentially updated in S32 of FIG. 8, the processing in S31 is performed for the most recently detected peak value and the peak value of the same sign immediately before that, When the continuous peak values do not change significantly, the process shifts to the note-on process. FIG.
In this example, both t = t ₃ and t = t ₄ are set to X = 1 at S31, so no note-on occurs. At t = t ₅ , it is determined that a normal peak is input for the first time, and X = 0. , Note on at t = t ₅ . And in this case,
The continuous pitch period TP (b) and the outer 15 are not normal values.

[0133]

[Outside 15]

It is << Operation of Duplication Processing >> The determination processing of S30 of FIG. 8 is a determination for duplication processing. Now, when the normal waveform D1 as shown in FIG. 11 is input, the current positive / negative flag b = 1 at t = t ₃ does not match the flag MT = 0, and the process proceeds to S31 as described above.

However, the above-mentioned "processing operation of STEP 1"
Or "Operation of debris processing" in "Processing operation of STEP2"
Similarly to the case described in the description of the section (1), when the waveform is doubled, the determination result of S30 in FIG. 8 is NO.

If the peak value c of the dubbed peak is smaller than the peak value AMP (b) immediately before it having the same sign as that of the dubbed peak, the determination result of S303 in FIG. Is ignored and the process returns to M2 of FIG. 4, and then STEP3 is repeated. This is based on the same idea as in FIG.

On the other hand, when the peak value c of the doubled peak is larger, the determination result of S303 is YE.
The process proceeds to S and the process proceeds to S304. Then, in S304, the immediately preceding peak is ignored, the content of AMP (b) is reset to the current peak value c, and the velocity VEL is recalculated using the value, and then M2 in FIG. 4 is set. Return to STE
Repeat P3. This is based on the same idea as in the case of FIG.

After the above processing, a normal peak is input, so that the determination result in S30 is YES, and further,
Each determination result of S35, S36, S37 and S39 is Y.
When the status becomes ES, the note-on process is performed, and the pronunciation of a musical sound is started. (Processing operation of STEP4) Next, FIG. 9 is an operation flowchart of the processing of step 4 (STEP4) shown as M9 of the main routine of FIG. In this processing, pitch extraction / change processing, relative on / relative off processing, processing when the pitch cycle is not appropriate, double processing, and the like are performed. Although this processing is not directly related to the present invention, it is important for performing pitch control of musical tones,
The explanation will be made sequentially. First, in the pitch extraction / change process, there are routes for only pitch extraction and routes for actually changing the pitch, and normally, they are alternately repeated every time a new peak is input. << Route Operation >> First, S40, S41, S42, S
The routes shown in 63 to S67 will be described. S40
At, the waveform number counter HNC> 3 is determined, and if YES, the process proceeds to S41. In S41, it is determined whether or not the relative-on threshold value TRLRL <(current peak value c-previous peak value AMP (b) of the same sign), and if NO, the process proceeds to S42 (if YES, it will be described later). To). In S42, this time positive / negative flag b = flag M
In other words, it is determined whether T, that is, the pitch change side, and if YES, the process proceeds to S43.

By the way, in the initial state, the waveform number counter HNC is 0 (see S301 in FIG. 8), so that a negative determination is made in S40 and the operation proceeds to S42. Then, for example, in the case of the waveform input as shown in FIG. 11, t = t ₄
At b = 0 and MT = 1 (S33 of STEP3 of FIG. 8)
It has been rewritten in), so S42 to S63
Go to.

In S63, it is determined whether or not the register RIV = 1 to check whether or not the peaks having the same polarity are continuously input (whether the peaks are double), and if YES (when the peaks are double). S6)
Proceed to step 8 to perform double processing (described later), and NO
If (no duplication), the process proceeds to S64, where the following process is performed.

That is, at S64, the current peak value c is input to the previous peak value AMP (b), and the previous amplitude value AMRL1 is input to the amplitude value AMRL2 of the previous two times for the relative off processing (described later). At first, AMR
The content of L1 is 0. (S301 of STEP3 of FIG. 8
reference).

Further, in S64, the previous peak value 16 with a different sign and the current peak value c

[0143]

[Outside 16]

Whichever is larger is the previous amplitude value AM.
Input to RL1. In other words, there are two positives in the cycle,
Amplitude value AM is larger than negative peak value.
It is set to RL1.

Then, in S65, the waveform number counter H
It is determined whether NC> 8, and here the waveform number counter (zero cross counter not on the pitch change side) HNC
Is incremented by 1 and is upgraded. Therefore, the upper limit of the waveform number counter HNC is 9. And S6
After 5 or S66, the process proceeds to S67.

In S67, the register RIV is set to 1, the content of the time storage register TFR is subtracted from the current zero-cross time t, and the result is input to the period register TTR. This cycle register TTR has cycle information TTR =
shows the _{t-TFR = t 4 -t 2} . Then, the zero-cross time t of this time is saved in the time storage register TFR,
After that, the process returns to M2 of the main routine of FIG.

As described above, the route is processed as follows according to the example of FIG. That is, MT = 1
≠ b, RIV = 0, AMP (0) ← c = b ₂ , AMRL
2 ← AMRL1 = 0, AMRL1 ← max {AMP
(1) = a ₁ , c = b ₂ (whichever is larger)}, H
NC ← {HNC + 1} = 1, RIV ← 1, TTR ← {t
−TFR} = {t ₄ −t ₂ } and TFR ← t = t ₄ . Therefore, the difference in the time information from the previous zero-cross point t = t ₂ (change point of STEP2 → 3) of the same polarity to the current zero-cross point t = t ₄ in the cycle register TTR, that is,
This means that the cycle information has been obtained. Then, the process returns to M2 of the main routine of FIG. 4 and waits for the next peak to be input. << Route Operation >> Next, a case where the route has proceeded to S40 to S62 will be described. Since the waveform number counter HNC = 1 now (see S66), S40 to S
The process proceeds to 42 (S40 will be described later).

In S42, for example, in the case of FIG.
Since MT = 1 and b = 1, the determination result is YES, and the process proceeds to S43.
In S43, it is determined whether or not the register RIV = 1. Since the register RIV has already been set to 1 in the route (see S67), the determination in S43 is YES, and the process proceeds to S44. The dubbing process when the determination result in S43 is NO will be described later.

In S44, it is determined whether or not the register STEP = 4. If YES, the process proceeds to S45 (NO is described later). In S45, it is determined whether or not the current peak value c <60H (H indicates hexadecimal notation). If a large peak value is input, the determination result is NO and the process proceeds to S47. On the other hand, if the value is less than 60H, the determination result is YES and the process proceeds to S46.

At S46, the amplitude value (peak value) of the previous two times
AMRL2-Previous amplitude value (peak value) AMRL1 ≤
(1/32) x the amplitude value (peak value) AMRL2 of the previous two times
If YES, proceed to S47 if YES,
The relative off counter FOFR is set to zero.
In the case of NO, the process proceeds to S74 and the relative off process is performed. The process of this relative off will be described later.

In S48, the period is calculated. Specifically, (the current zero-cross time t-the previous zero-cross time data TF) is the register T as the current cycle information tt.
Set to OTO. Then, the process proceeds to S49.

In S49, it is determined whether or not the current cycle information tt> frequency upper limit THLIM (upper limit after sound generation starts), and if YES, the process proceeds to S50 (if NO, it will be described later). The upper frequency limit THLIM of S49 is ST of FIG.
It is the same as the upper limit of the permissible range of the frequency at the time of triggering (at the start of sound generation) used in S36 of EP3 (thus, it is the minimum period and corresponds to the pitch period two to three semitones above the highest fret). ..

Next, in S50, the following processing is performed.
That is, the register RIV is set to 0, the current zero-cross time t is input as the previous zero-cross time data TF,
Also, the previous peak value AMP (b) is input to the peak value e two times before, and the current peak value c is further changed to the previous peak value AM.
Input to P (b).

After the processing of S50, the process proceeds to S51,
In S51, the lower limit of frequency TLLIM> current cycle information t
When it is YES, that is, when the current cycle is equal to or lower than the pitch extraction range lower limit during note-on, the process proceeds to S52. In this case, the lower frequency limit TLLIM is
For example, it is set one octave below the open string scale.
That is, the lower frequency limit TTLIM (S
37), the allowable range is widened. By doing so, it becomes possible to cope with the frequency change due to the operation of the tremolo arm or the like.

With the above operation, the process proceeds to S52 only when the upper and lower limits of the frequency are within the allowable range.
If not, the process returns from S49 and S51 to the process of M2 of the main routine of FIG. 4 to wait for the input of the next peak.

Next, in S52, the cycle data TTP is input to the cycle data h extracted two times before, and the cycle information tt of this time is input to the cycle data TTP previously extracted.

Then, in S53, the current peak value c is written in the velocity VEL, and the process proceeds to S54. In S54, it is determined whether or not the no-change level NCHLV> (previously before peak value e-current peak value c), and if YES, the process proceeds to S55. That is, the previous peak value of the same polarity (e
= When AMP (b) and the current peak value c are greatly changed, the difference exceeds NCHLV,
In such a case, if the pitch is changed based on the extracted period information, there is a high possibility that an unnatural pitch change will occur. Therefore, if a NO determination is made in S54, the process returns to the process of M2 of the main routine of FIG. 4 without waiting for the processes of S55 and thereafter, and waits for the input of the next peak.

Next, in the case of YES in S54, it is determined whether or not the relative off counter FOFR = 0. When the relative off process which will be described later is being performed, the relative off counter FOFR is not 0, and in such a case, the determination of NO is made in S55 without performing the process of changing the pitch (see S61). The process returns to M2 of the main routine of FIG. If YES is determined in S55, the process proceeds to S56 and S57.

In S56 and S57, the two-wave three-value matching condition is determined. In S56, the current cycle information tt × 2 ⁻⁷ > | this cycle information tt-previously previous cycle data h |
If YES, the process proceeds to S57, and in S57, the current cycle information tt × 2 ⁻⁷ > | the current cycle information tt−contents of the cycle register TTR | is determined, and if YES, S58.
Proceed to. That is, in S56, in the example of FIG. 11, the current cycle information tt = t ₅ -t ₃ (see S48) is approximately the value of the previous period data _{h = TTP = t 3 -t 1} ( see S52) it is determined whether match, the S57, the current cycle information tt = t ₅ -t _3, the period TTR = t overlapping therewith
_It is determined whether or not it substantially matches _4- t ₂ (see S67). The limit range is set to 2 ⁻⁷ · tt, and the value changes depending on the cycle information. Of course,
This value may be a fixed value, but a better result can be obtained by the technique adopted in this embodiment.

In the next S58, it is determined whether or not this cycle information tt> constant TTU, and if YES, the process proceeds to S59, in which it is determined whether this cycle information tt <constant TTW, and if YES, to S60. move on. Note that S58,
The case where NO is determined in S59 will be described later.

In S60, it is determined whether or not the register STEP = 4. If YES, the process proceeds to S61. In S61, the musical sound generating circuit 5 from the MCP3 in FIG.
The pitch is changed to 01 (based on the current cycle information tt), and the process proceeds to S62. In the pitch changing process in S61, it is not necessary to correct the pitch cycle (see S302 in FIG. 8) as at the start of sounding.

In S62, the time constant is changed corresponding to the current cycle information tt, and the constant TTU is (17/3).
2) × rewritten to the current cycle information tt, and the constant T
TW is rewritten as (31/16) × current cycle information tt.

As will be described later, STEP = 5 only when the relative off process is performed. In that case, the process directly proceeds from S60 to S62 and the pitch is not changed in S61. Make a constant change.

The processing of changing the time constant is the time constant conversion register CH in the time constant conversion control circuit 202 of FIG.
It means that the MCP 3 of FIG. 1 sets the cycle data based on the value of the current cycle information tt in the TRR. This is as already described in the section "Detailed Operation" of "Operation of Pitch Extraction Digital Circuit".

When the process of S62 is completed, the process returns to the process of M2 of the main routine of FIG. As described above, in the route, the following processing is performed in the example of FIG. In other words, HNC = 1, MT = 1 = b, it is determined that the RIV = 1, FOFR ← 0, tt ← t-TF = t 5 -t 3, RI
V ← 0, TF ← t = t ₅ , e ← AMP (1) = a ₁ , A
MP (1) ← c = a ₂ , h ← TTP = TP (1) = t ₃
−t ₁ , TTP ← tt = t ₅ −t ₃ , VEL ← c = a ₂
Further, the pitch is changed according to tt when the following three conditions are satisfied: TTP≈TTR≈tt, TTU <tt <TTW, and AMP (0) −c <NCHLV.
After that, TTU ← (17/32) × tt, TTW ← (3
1/16) × tt is set.

With the above operation, in the route, the pitch is changed with respect to the actual tone generating circuit 501 (FIG. 1), the route is processed by the following zero cross interrupt (detection of the next peak), and similarly, the following zero cross interrupt is performed. The route is processed. In this way, the root simply extracts the cycle (see S67).
However, on the route, actual pitch change (see S61) and time constant change processing (see S62) are performed. << Relative-ON Processing Operation >> After the waveform number counter HNC is incremented to exceed 3 in S66 of the route in STEP 4 of FIG. 9, a YES determination is made in S40, and the process proceeds to S41 and the relative-on is performed. The condition of is detected.

This is c-AMP (b)> TRLRL, and when the current peak value c exceeds the threshold value TRLRL and is larger than the previous peak value AMP (b), that is, this is the string. This happens when the same string is picked again (by the tremolo playing method or the like) after the operation. In this case, the determination result of S41 is YES, and the process proceeds from S41 to S78 to perform the relative on process.

In S78, the time constant conversion control circuit 202
The period CHTIM of the highest-pitch fret (for example, 22 fret) is set in the time constant conversion register CHTRR (FIG. 2). After the above processing, the process proceeds to S06 in STEP0 of FIG. 5, notes off the musical tone being sounded, and then starts re-sounding. According to the normal performance operation, the STE of FIG.
In S41 of P4, a NO determination is made, and the process proceeds to S42,
Proceed to the above route or route. << Relative Off Processing Operation >> Next, the relative off processing will be described with reference to FIG. Relative off refers to performing a muffling operation in accordance with an operation of shifting from a state of fret operation to an open string state without picking.

In this case, the amplitude level of the waveform sharply drops, and the difference between the peak value (peak value) AMRL2 of the previous time and the peak value (peak value) AMRL1 of the previous time is (1/32).
・ Beyond AMRL2, the process proceeds from S46 to S74 in STEP4 of FIG.

Then, the relative off counter FOFR
Advances from S74 to S75 so as to count up until exceeds the constant ROFCT. Continuously, from S75 to S48
Go to step S49 to S55, but FOFR = 0
Therefore, the determination result of S55 is NO, and immediately before the relative off process is started, the process returns to the process of M2 of the main routine of FIG. 4 without changing the pitch.

Then, the peaks at the time of relative off are input one after another, and if NO is determined in S74, that is, when the value of FOFR becomes 3 (ROFCT is 2) in the example of FIG. 17, S74 to S76. Move to.

However, if the result of the determination in S46 is YES, the process proceeds from S46 to S47 to reset the FOFR. Therefore, unless the condition of S46 is continuously satisfied for the number of times specified by ROFCT,
No relative off processing is performed. ROFCT
The value of is large for strings with high pitch,
Relative-off processing can be performed on any of the strings after a substantially constant time has elapsed.

Next, when proceeding from S74 to S76, the relative off counter FOFR is reset and the register ST
EP is set to 5, and the process proceeds to S77 to instruct the tone generation circuit 501 (FIG. 1) to perform note-off.

In the state of STEP 5, the pitch extraction processing is executed in the same manner as in STEP 4, but since the process proceeds from S60 to S62 without going through S61, the pitch change is instructed to the tone generating circuit 501. Not done. However, the time constant changing process is performed according to the cycle extracted in S62.

Then, in the state of STEP 5, the relative on process is accepted (S41, S78), but in other cases, it is detected in the main routine of FIG. 4 that the vibration level has decreased. By doing M14
Then, STEP becomes 0 and the initial state is restored.

It should be noted that AMRL1 and AM used in S46
RL2 is made in S64, and the peak with the larger level (one of the maximum peak and the minimum peak) in one cycle is set to this value, and the maximum peak a in FIG.
When k is always larger than the maximum peak bk −1, an + 1 and an + 2, an + 2 and an + 3, an +
Any difference between 3 and an + 4 exceeds a predetermined value.

At this time, in the processing of the route, the minimum peaks bn + 1, bn + 2, bn + 3 have extremely decreased, so that a NO determination is made in S54 and the processing returns to the processing of M2 of the main routine of FIG. , Pitch change processing is not performed. << Processing Operation When Pitch Cycle Is Inappropriate >> Next, when the pitch cycle is inappropriate, that is, when the pitch is being extracted, S58
Alternatively, the processing when an overtone having an octave relationship, that is, a cycle having a higher octave or a cycle having a lower octave is continuously detected in S59 will be described.

Now, the constant TT of S58 in STEP4 of FIG.
U is set to the minimum value 0 in S301 of STEP3 of FIG. 8, and the constant TTW is set to the maximum value MAX in the same manner.
When this flow is passed for the first time, the determination of YES is always made in S58 and S59, but thereafter, in S62,
The constant TTU is set to (17/32) .tt (period information of high octave for one octave), and the constant TTW is similarly (31/16) .tt (period information of low tone for one octave) at S62. Is set.

Therefore, when the octave is suddenly raised (this occurs when the vibrating string is muted so as to stop the vibration with a finger), or when the octave is lowered (this is when the peak of the waveform is missed, etc.). Happen)
When the pitch occurs, it is unnatural when the pitch is changed, so the process branches so as not to change the pitch.

That is, when tt does not exceed TTU in S58, that is, when it becomes smaller than the value TTU which is 17/32 times the cycle extracted last time, the process proceeds to S76. In other words, when a high octave sound is extracted, it is considered that the mute operation has been performed, and the process proceeds from S58 to S76 without outputting the high octave sound, and the sound is produced by the processing of S76 and S77 as in the case of the relative off. To stop.

Further, in S59, when tt does not exceed TTW, that is, when it becomes larger than the value TTW which is 31/16 times the previously extracted period, it proceeds to S60 without proceeding to S60 of the main routine of FIG. Return to processing.

This state is shown in FIG. Normally, if the waveform is very small near note-off, the waveform will come on due to crosstalk of the hexa pickup or body resonance due to other picking. Then, for example,
The input waveform may be as shown in FIG. 18, and the input waveform one octave below may be continuously detected.

In such a case, if no processing is performed,
Suddenly, the sound below the octave is output, making it extremely unnatural. Therefore, in S56 and S57, Tan + 2≈T
Even if an + 3≈Tbn + 2 is detected, Tan + 3> T
Since it becomes an + 1 × (31/16), the process returns from S59 to the process of M2 of the main routine of FIG. 4 without changing the pitch. << Operation of Double-Processing >> Next, the process when the waveform is double-extracted, that is, when the peaks of the same polarity are continuously detected will be described.

First, in the route in which the determination result of S42 in STEP4 of FIG. 9 is NO, in the case of YES in S63, the process proceeds to S68 and the duplication process is performed. That is, S6
If YES in step 3, the process proceeds to step S68, and it is determined whether or not the current peak value c> previous peak value AMP (b) having the same sign, and if YES, the process proceeds to step S69.

At S69, the previous peak value AMP (b) is rewritten to the current peak value c, and the routine proceeds to S70. S7
If 0, this time peak value c> previous amplitude value (peak value) AM
Whether or not it is RL1 is determined, and if YES, the process proceeds to S71, where the current peak value c is set as the previous amplitude value (peak value) AMRL1.

If NO is determined in S68, the process returns to M2 of the main routine of FIG. Therefore, it is considered that the peak of the overtone is not spread only when the peak of the new input waveform is large.
The peak value of the new waveform is registered.

When NO in S70 and when the processing in S71 ends, the process similarly returns to the main routine. FIG. 19 shows an example of the above double processing. In the case of this example, the state of MT = 0 is shown. In general, the fundamental wave period and the period of the overtone component have a non-integer multiple relationship, so the phase of the overtone will shift and the zero cross of the same polarity will be detected. I have to In the case of the example in this figure, a double state occurs at a place where "double" is indicated.

At this time, go from S42 to S63 and set Y
Determine ES and go to S68. In S68, in this case, (an + 2) and (an + 3) are compared, and (a
Only when n + 3) is larger than (an + 2), go to S69 and rewrite AMP (1). Then, the previous amplitude value (peak value) AMRL1 and the current amplitude information (peak value c) are compared in S70, and if YES, S71.
Go to, and set the current peak value c to the previous amplitude value (peak value)
AMRL1.

Next, in the route in which the determination result of S42 in STEP4 of FIG. 9 is YES, NO in the next S43.
In such a case, the process proceeds to S72 and the duplication process is performed in the same manner as above.

That is, if NO in S43, the process proceeds to S72, in which the current peak value c> the previous peak value AMP having the same sign.
Whether or not (b) is determined, and if YES, the process proceeds to S73, where the previous peak value AMP (b) is rewritten to the current peak value c, and then the process returns to M2 of the main routine of FIG.

If NO is determined in S72, the process immediately returns to M2 of the main routine of FIG. Therefore, also in this case, the peak value of the new waveform is registered only when the peak of the new input waveform is large.

FIG. 20 shows an example thereof. In this example MT
= 1 is shown. In this case, in the process of STEP4 at the time of zero cross where "Dubli" in the figure is written, the process proceeds from S42 to S43, and YES is determined in S43 and S72 is executed.
Go to. Here, the sizes of (an + 3) and (an + 2) are compared, and if (an + 3) is larger than (an + 2), a YES determination is made in S72, and AMP (1) is set to
The value of (an + 3) is set, and if it is the opposite, no change processing is performed.

By the way, in the case of the above-mentioned duplication processing, since the extracted time data is not used at all, the cycle information T
An + 3 does not change at all. Further, naturally, the pitch change based on the cycle data is not performed. {Other embodiment of the present invention} As described above, in the present embodiment, the present invention is applied to the electronic guitar (guitar synthesizer), but the present invention is not limited to this. A musical tone generator of the type that detects a performance operation as an input waveform signal and generates an acoustic signal different from the original signal based on it.When the input waveform signal rises, the pitch period deviates from the true pitch. Various applications are possible as long as they have a tendency to be extracted.

[0194]

According to the present invention, when the pitch extracting means extracts the pitch information from the input waveform signal, the pitch information correcting means corrects the original pitch period only when the input waveform signal rises. Therefore, even if the frequency corresponding to the pitch information is extracted with a deviation from the true value at the rising edge of the input waveform signal, it is possible to specify the pitch that accurately corresponds to the actual pitch, and the accurate pitch can be specified. It becomes possible to start the pronunciation of musical sounds at high pitches.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an electronic musical instrument according to the present invention.

FIG. 2 is a block diagram of a pitch extraction digital circuit.

FIG. 3 is a diagram showing an operation flowchart of an interrupt processing routine.

FIG. 4 is a diagram showing an operation flowchart of a main routine.

FIG. 5 is a diagram showing an operation flowchart of STEP0.

FIG. 6 is a diagram showing an operation flowchart of STEP1.

FIG. 7 is a diagram showing an operation flowchart of STEP2.

FIG. 8 is a diagram showing an operation flowchart of STEP3.

FIG. 9 is a diagram showing an operation flowchart of STEP 4 (5).

FIG. 10 is a schematic operation explanatory diagram of the present embodiment.

FIG. 11 is a diagram illustrating the basic operation of this embodiment.

12A and 12B are operation explanatory diagrams of the duplication process in STEP 1. FIG.

FIG. 13 is an explanatory diagram of the operation of the double processing in STEP2.

FIG. 14 is an operation explanatory diagram of the duplication process in STEP 2;

FIG. 15 is an operation explanatory diagram of the duplication process in STEP 2;

FIG. 16 is an operation explanatory diagram of noise removal processing in STEP3.

FIG. 17 is an operation explanatory diagram of a relative off process in STEP4.

FIG. 18 is an explanatory diagram of a processing operation in STEP 4 when the pitch cycle is inappropriate.

FIG. 19 is an explanatory diagram of the operation of the duplication process on the route.

FIG. 20 is a diagram illustrating the operation of the duplication process on the route.

[Explanation of symbols]

 1 Pitch extraction analog circuit 2 Pitch extraction digital circuit 3 Central control unit (MCP) 4 Address decoder 5 Music tone generator 201 Peak detection circuit 202 Time constant conversion control circuit 203 Crest value capture circuit 204 Zero cross time capture circuit 501 Music tone generation circuit D1 digital output

Claims (1)

[Claims] 1. A pitch extracting means for extracting pitch information from an input waveform signal, a pitch information correcting means for correcting pitch information at the rising edge of the input waveform extracted by the pitch extracting means, and a rising edge of the input waveform. At the time, a musical tone is generated with a pitch according to the pitch information corrected by the pitch information correcting means, and after the rising time, a musical tone is generated with a pitch according to the pitch information extracted by the pitch extracting means. An electronic musical instrument comprising: a musical tone generating means.