JP3243856B2 - Pitch extraction type electronic musical instrument - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pitch extraction type electronic musical instrument for extracting a pitch from waveform data input from a microphone or the like, generating musical tone data based on the extracted pitch, and generating a tone.

[0002]

2. Description of the Related Art Conventionally, a pitch extraction type electronic device in which a pitch is extracted from waveform data input by a microphone or a guitar (string), tone data is generated based on the extracted pitch, and a tone is generated by tone generating means. Instruments are known.

Incidentally, the pitch of waveform data input from a microphone or the like is not always stable from the beginning, and it takes some time for the pitch to stabilize.

Therefore, in a conventional pitch extraction type electronic musical instrument, when a pitch is extracted from input waveform data,
The frequency band is shifted little by little, the pitch is extracted each time within the range of each shifted frequency band, and it is confirmed whether or not the pitch is stabilized, thereby outputting a stable pitch.

[0005]

For this reason, it takes time to extract the pitch. For example, a voice (singing voice) input from a microphone is output, and a tone is generated based on the pitch extracted from the voice waveform data. In the case where an instrument sound is generated by an external sound source by generating data, there has been a problem that a gap occurs between the voice and the instrument sound.

The present invention has been made under such circumstances, and an object thereof is to enable a pitch to be rapidly extracted from input waveform data.

[0007]

To achieve the above object, the present invention provides a waveform input means for inputting waveform data,
Pitch extracting means for extracting a pitch from the waveform data input by the waveform input means, wherein the pitch extracting type electronic musical instrument for generating musical tone data based on the pitch extracted by the pitch extracting means and generating a tone ; and multiple operators, said pitch extraction means and a frequency band specifying means for specifying in accordance with the frequency band to the operation of the operator in extracting pitch.

[0008]

When waveform data is input from the waveform input means,
The pitch extracting means extracts a pitch from the input waveform data.

In this case, the pitch extracting means shifts the frequency band little by little, without extracting the pitch each time within the range of each frequency band, and designates the pitch by the frequency band specifying means in accordance with the operated operator. The pitch is extracted within the range of the specified frequency band.

Therefore, the pitch can be quickly extracted from the input waveform data.

[0011]

Next, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an outline of a microphone type electronic musical instrument to which a pitch extracting type electronic musical instrument according to an embodiment of the present invention is applied.

In the microphone type electronic musical instrument shown in FIG. 1, waveform data such as sound input from the microphone 1 is amplified by the amplifier 2 and is transmitted through the mixer 3 to the sound system S.
It is pronounced from S.

The waveform data (analog data) input from the microphone 1 is output to an A / D converter 4 via an amplifier 2, converted into digital data by the A / D converter 4, and sent to a pitch extractor 5. Is output. The pitch extractor 5 extracts a pitch from the input waveform data and outputs the extracted pitch to the CPU 6 via the bus line BUS.

As shown in FIG. 2, the microphone 1 is provided with eight button switches SW1 to SW8 to be operated by the index finger to the little finger on the front side, and one button switch to be operated by the thumb on the back side. Button switch SW9 is provided.

The button switches SW1 to SW8 are used to shift and designate a frequency band in units of one octave when the pitch extractor 5 performs pitch extraction. That is, as shown in FIG. 3, the button switch SW2 is used to designate a basic frequency band corresponding to one octave of C4 to B4. And the button switch SW1 is C5 ~ one octave above the basic frequency band.
The button switch SW3 designates a frequency band corresponding to one octave of C3 to B3, one octave below the basic frequency band, and the button switch SW4 designates a frequency band corresponding to one octave range of B5. It is used to designate a frequency band corresponding to one octave of C2 to B2 two octaves below the basic frequency band. The button switch SW5 specifies a frequency band corresponding to one octave of G5 to F6 shifted one octave + 5 degrees above the basic frequency band, and the button switch SW6 shifts 5 degrees above the basic frequency band. A frequency band corresponding to one octave of G4 to F5 is designated, and the button switch SW7 is shifted from the basic frequency band by -1 octave + 5 degrees (that is, shifted by 1 octave-5 degrees). Specify the frequency band corresponding to the octave range, and press the button switch S
W8 is G2 shifted from the fundamental frequency band by -2 octaves +5 degrees (that is, shifted down by 2 octaves-5 degrees).
It is used to designate a frequency band corresponding to one octave of F3.

The button switch SW9 has a built-in pressure sensor and is used for inputting after-touch information and the like.

The switch detector 7 in FIG. 1 is used to turn on / off the button switches SW1 to SW9 provided on the microphone 1.
Off is detected and output to the CPU 6. Further, in addition to various control programs, the ROM 8 includes, as shown in FIG.
A parameter table T in which frequency bands (parameters) to be designated are recorded corresponding to the switch numbers of the button switches SW1 to SW8 is formed. Note that the parameter table T includes button switches SW1 to SW8.
All frequency bands (corresponding to the range of G1 to F6) as the frequency bands to be designated when none of them are turned on
Is also recorded.

The CPU 6 searches the parameter table T for a frequency band corresponding to the button switch that has been turned on, and outputs the frequency band to the pitch extractor 5. And CPU
6 generates musical tone data based on the pitch extracted by the pitch extractor 5 and outputs it to the external sound source 10 via the MIDI output unit 9. The external sound source 10 generates tone waveform data based on the tone data from the CPU 6 and outputs the tone waveform data to the mixer 3. The mixer 3 mixes the musical sound waveform data from the external sound source 10 and the audio waveform data from the microphone 1 and outputs the mixed sound waveform data to the sound system SS.

The RAM 11 is used as a work area when performing various processes. The timer 12
Outputs a clock having a frequency corresponding to the tempo information designated by the panel switch 13 and
A clock for starting the process performed in step 6 every predetermined time is also output. The display 14 displays various performance-related information.

FIG. 5 is a block diagram showing an example of the configuration of the pitch extractor 5, which includes a fundamental wave extraction filter 51, a first zero cross detection circuit 52a, a second zero cross detection circuit 52b,
First time interval measuring circuit 53a, second time interval measuring circuit 5
3b and a pitch determination circuit 54.

The fundamental wave extraction filter 51 is a filter for extracting a fundamental wave from the waveform data input from the microphone 1, and the CPU 6 designates a filter coefficient range (frequency band) for extracting the fundamental wave. The first zero-cross detecting circuit 52a detects each zero-cross when the fundamental wave extracted by the fundamental-wave extracting filter 51 changes from negative to positive, and the first time interval measuring circuit 53a detects the time interval of the zero-cross. Measure. Further, the second zero-cross detecting circuit 52b detects each zero-cross when the fundamental wave extracted by the fundamental-wave extracting filter 51 changes from positive to negative, and the second time interval measuring circuit 53b detects the time of the zero-cross. Measure the interval. Then, when the time intervals measured by the first time interval measuring circuit 53a and the second time interval measuring circuit 53b do not change and become stable, the pitch determining circuit 54 determines the pitch based on the stable time intervals. Thus, the pitch is extracted.

Next, performance control by the CPU 6 will be described.

FIG. 6 is a flowchart showing the main processing in the performance control.
Initialize the type, tempo, rhythm, etc. of the instrument sound set in Step 3 (Step S1). Next, panel switch processing such as applying vibrato based on an instruction from the panel switch 13 is performed (step S2). Then, predetermined microphone switch processing is performed on the button switches SW1 to SW9 provided on the microphone 1 (hereinafter, these switches are referred to as microphone switches) (step S3). Then, it is started at regular intervals by the timer 12, and performs a microphone pitch process to cope with a change in the pitch of the waveform data input from the microphone 1 (step S4), and returns to step S2.

Next, the microphone switch processing will be described in detail. The microphone switch processing is roughly divided into a microphone switch on event processing and a microphone switch off event processing. ,
The microphone switch process (step S3) is activated, and is activated only when any of the microphone switches has an ON event or an OFF event.

In the microphone switch-on event processing,
As shown in the flowchart of FIG. 7, the switch number of the microphone switch having the off event is stored in the buffer BUF.
(Step S301). Next, buffer B
It is determined whether or not the switch number in the UF matches the switch number of the microphone switch related to the previous switch-on set in the register SWN (step S30).
2). As a result, when they do not match, that is, when another microphone switch is newly turned on, the detection parameter of the frequency band (frequency band corresponding to one octave sound range) corresponding to the switch number in the buffer BUF is changed to the parameter. The data is read from the table T and output to the pitch extractor 5 (step S303). As a result, the pitch extractor 5 shifts the frequency band little by little, and does not extract the pitch each time within the shifted frequency band, but within the frequency band corresponding to the given one octave sound range. Since pitch extraction is performed, pitch extraction is quickly performed.

Next, it is determined whether or not the content of the register SWN is "0" (step S304). as a result,
If the content of the register SWN is not “0”, it means that note-on (sound generation) is currently performed based on the pitch in the register TPIT, as will be apparent from the following description. The external sound source 10 is instructed via the MIDI output unit 9 to turn off the note based on the pitch (step S305), and
Proceed to 306. On the other hand, if the content of the register SWN is "0", it means that nothing is currently being pronounced. In this case, the process immediately proceeds to step S306, and the switch No. in the buffer BUF is stored in the register SWN. set.

Next, it is determined whether or not the previous extraction pitch set in the register DPIT is a pitch in a frequency band (sound range) corresponding to the switch No. in the register SWN (step S307). As a result, the register S
If the pitch is not within the frequency band (tone range) corresponding to the switch number in the WN, the pitch extraction status of the pitch extractor 5 is checked (step S308).
It is determined whether or not is extracting the pitch (step S
309). As a result, if the pitch extractor 5 has extracted the pitch, the extracted pitch is set in the register DPIT (step S310), and the process proceeds to step S312.
On the other hand, if the pitch extractor 5 has not extracted the pitch,
That is, if no waveform data is input from the microphone 1, the pitch previously set in the register DPIT is
The value is changed in octave units so as to enter the frequency band (sound range) corresponding to the switch No. in the register SWN (step S311), and the process proceeds to step S312. On the other hand, when it is determined in step S307 that the previously extracted pitch set in the register DPIT is a pitch in the frequency band (sound range) corresponding to the switch No. in the register SWN, steps S308 to S311 are performed. Skip,
Proceed to step S312.

In step S312, the register DPIT
Is set in the register TPIT. And
The external sound source 10 is output via the MIDI output unit 9 so as to turn on the note having the pitch set in the register TPIT.
And exit.

In step S302, buffer B
If it is determined that the switch No. in the UF matches the switch No. of the microphone switch related to the previous switch-on set in the register SWN, the process is terminated as it is, and the buffer B is reset due to some error.
When the switch number in the UF matches the switch number set in the register SWN, the processing after step S303 is not performed.

Next, the microphone switch off event process will be described with reference to the flowchart of FIG.

In the microphone switch off event processing,
The switch number of the microphone switch having the off event is set in the buffer BUF (step S351). Next, the switch No. in the buffer BUF and the register SW
It is determined whether or not the switch number of the microphone switch related to the previous switch-on set to N matches with the switch number (step S352). As a result, when they do not match, that is, when another microphone switch is newly turned on,
Since the note has already been turned off in step S305 in FIG. 7, the process ends.

On the other hand, when they match, that is, when a predetermined microphone switch is turned off and another microphone switch is not newly turned on, first, a parameter when the microphone switch is not operated, that is, pitch extraction of the entire frequency band is performed. The range parameters are read from the table T and output to the pitch extractor 5 (step S353). Next, based on the pitch in the register TPIT, the MIDI output unit 9 turns off the note that has been turned on (produced).
Is instructed to the external sound source 10 through (step S35)
4). Then, “0” is set in the register SWN (step S355), and the process ends.

Next, the details of the microphone pitch process activated by the timer 12 at regular intervals will be described with reference to the flowchart of FIG.

The pitch extraction status of the pitch extractor 5 is checked (step S401), and it is determined whether or not the pitch extractor 5 is currently extracting a pitch (step S40).
2). As a result, if the pitch extractor 5 has not extracted the pitch, the process ends. On the other hand, if the pitch extractor 5 has extracted the pitch, the extracted pitch is stored in the register D.
Set to PIT (step S403), register S
It is determined whether or not the content of WN is “0” (step S404). As a result, the content of the register SWN becomes “0”.
If this is the case, it means that there is no microphone switch that is currently turned on, and in this case, the process ends immediately.
On the other hand, if the content of the register SWN is not “0”, it means that there is a microphone switch that is currently turned on. In this case, furthermore, a new extraction pitch in the register DPIT and a new extraction pitch in the register TPIT It is determined whether or not the pitch of the currently sounding musical tone matches (step S4).
05). As a result, if they match, the process ends.

On the other hand, if they do not match, the register TPI
An instruction is issued to the external sound source 10 via the MIDI output unit 9 to turn off the note that has been turned on (produced) based on the pitch in T (step S406). Next, the pitch in the register DPIT is set in the register TPIT (step S407), and the external tone generator 10 is instructed via the MIDI output unit 9 to note on the note having the pitch set in the register TPIT (step S40).
8), end.

[Applied Modifications] The present invention is not limited to the above-described embodiments. For example, a peak detection method other than the zero-cross detection method, an auto-correlation detection method, and
Also, the present invention can be applied to a pitch extractor based on t Fourier (Transform) method.

In the case of the peak detection method, a fundamental wave extraction filter is used as in the case of the zero-cross detection method. Therefore, a parameter indicating the filter coefficient range of the fundamental wave extraction filter may be specified as frequency band information. In the case of the autocorrelation detection method, ∫W (t) · W (t +
A parameter indicating the range of Δt may be specified as frequency band information for the calculation unit that performs the autocorrelation calculation Δt). In the case of the FFT method, a parameter indicating a conversion range may be specified as frequency band information for an operation unit that performs an FFT operation.

[0039]

As described above, according to the pitch extraction type electronic musical instrument of the present invention, the pitch can be quickly extracted from the input waveform data. Therefore, for example, in the case of outputting a voice (singing voice) input from a microphone, generating musical sound data based on a pitch extracted from the voice waveform data, and generating an instrument sound by an external sound source,
It is possible to eliminate a problem such as a shift between the voice and the instrument sound.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an outline of a microphone type electronic musical instrument to which a pitch extraction type electronic musical instrument according to an embodiment of the present invention is applied.

FIG. 2 is a diagram showing a button switch specially provided for a microphone.

FIG. 3 is a diagram for explaining a function of a button switch specially provided in a microphone.

FIG. 4 is a diagram showing data contents of a parameter table.

FIG. 5 is a block diagram illustrating a configuration example of a pitch extractor.

FIG. 6 is a main flowchart showing an outline of performance control.

FIG. 7 is a flowchart showing a microphone switch-on event process.

FIG. 8 is a flowchart showing a microphone switch off event process.

FIG. 9 is a flowchart illustrating a microphone pitch process.

[Explanation of symbols]

 Reference Signs List 1 microphone 3 mixer 5 pitch extractor 6 CPU 7 switch detector 8 ROM 10 external sound source 11 RAM SS sound system T parameter table

Claims (1)

(57) [Claims] 1. Waveform input means for inputting waveform data,
Pitch extraction means for extracting a pitch from the waveform data input by the waveform input means, wherein the pitch extraction type electronic musical instrument for generating and generating musical tone data based on the pitch extracted by the pitch extraction means , and multiple operators, pitch extraction type electronic musical instrument, characterized in that said pitch extraction means and a frequency band specifying means for specifying in accordance with the frequency band to the operation of the operator in extracting pitch.