JP2646925B2 - Automatic accompaniment device for electronic musical instruments - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic accompaniment apparatus for automatically setting a chord (chord) when performance information is given, and performing an automatic accompaniment based on the chord and a preselected rhythm type. In particular, it relates to an automatic accompaniment technique for performing an automatic accompaniment based on a fraction code.

[0002] The chord changes depending on its development type. Since the lowest chord (bass tone) of a chord is an important factor in determining the expression of a chord, a chord is often directly specified together with a chord name when specifying a chord. At this time, for example, CM7 / B (or CM7o)
Since the numerator of the fraction is expressed as a chord for playing a chord and the denominator is expressed as a pitch name for a bass performance as in nB), the expanded name of the chord, or more aggressively, the pitch name and the chord that the bass part should intend A method of separately specifying chords that accompaniment is intended to call is called a fraction code.

[0003]

2. Description of the Related Art When a fraction code is specified, two types of code information and base information are required. For example, a chord is specified on a first keyboard (hand keyboard) and a bass sound is specified on a second keyboard (foot keyboard). Was specified. In addition, Japanese Unexamined Patent Publication
Japanese Patent Application Publication No. 179690 discloses a technique of specifying a fraction code using only one key by distinguishing a code from its abbreviated type.

In the above-mentioned Japanese Patent Application Laid-Open No. 2-179690, a fraction code is generated by adding a predetermined number determined for each designated fraction code to a base pitch key code and shifting the bass sound. doing.

[0005]

However, Japanese Patent Application Laid-Open No.
In the method of -179690, any pitch in the bass pattern is shifted by a fixed width.
Except for a monotonous bass pattern, for example, the bass note keeps the same note name, there is a problem that the pitch after shifting does not match the scale or becomes a dissonant (different chord). For this reason, the base pattern itself had to be monotonous.

In view of the above problems, the present invention does not cause discomfort even if any bass pattern does not produce a dissonant tone (in the case of a performance in which a dissonant tone is actively incorporated, the dissonant tone may be generated). An object is to realize automatic accompaniment using a fraction code.

[0007]

SUMMARY OF THE INVENTION The present invention provides a chord detecting means for detecting a chord including a fractional chord from pitch information of a performance;
When the chord detecting means detects a fractional chord, a specifying means for specifying a specific note name relating to the denominator of the fractional chord as a specific bass note name, based on a numerator of the fractional chord detected by the chord detecting means. Base pattern generating means for generating a base pattern including a rhythm pattern and a pitch pattern of the bass performance, and changing means for changing the base pattern, wherein a predetermined one of the pitches of the pitch pattern of the base performance is Changing means for changing the pitch of the specified base note name specified by the specifying means, and automatic bass playing means for performing an automatic bass performance based on the bass pattern changed by the changing means. It is characterized by being provided.

[0008]

[Function] A keyboard is pressed or MIDI (Mu
When sical instrument digital interface information or the like is given, the chord detecting means detects a chord. When a fraction chord is detected, the specifying means specifies a pitch name corresponding to the denominator of the fraction chord. Then, the base pattern generating means generates a base pattern corresponding to the numerator of the fractional chord. Then, the converting means converts a predetermined pitch in the bass pattern so as to be the pitch name specified by the specifying means. Then, the automatic bass performance means plays the bass pattern corresponding to the fraction code.

For example, the chord detected by the chord detecting means is Dm / F, and the base pattern corresponding to Dm is shown in FIG.
If it is (A), the pitch name F is specified by the specifying means corresponding to the denominator F, the predetermined pitch D in the base pattern is converted into the specified F, and the pattern shown in FIG. Is played.

It is considered that all pitches of the same pattern are uniformly circulated as in the prior art.

Since the sound D needs to be shifted by a minor third in order to become the sound F, if all of them are shifted by a minor third, as shown in FIG. Occurs. In particular, the G♯ sound is a semitone different from the A sound, which is a constituent sound of Dm, and thus becomes a dissonant sound.

[0012]

FIG. 2 is a block diagram of an electronic musical instrument provided with an automatic accompaniment device according to an embodiment of the present invention.

A key 1 comprising a plurality of keys is connected to a key event detection circuit 2, which detects an ON event and an OFF event of each key of the keyboard.

A switch event detection circuit 4 is connected to a switch circuit 3 including various switches such as a tone color selection switch and a rhythm selection switch, and the detection circuit 4 detects an ON event and an OFF event of each switch. The ROM 5 stores a chord / base sound conversion table 50,
Code pattern memory 51, base pattern memory 52
And a rhythm pattern memory 53. code·
The bass sound conversion table 50 stores the chord pattern memory 5
1, the basic (Cmaj) chord pattern stored in the base pattern memory 52 and the basic (Cmaj) base pattern stored in the base pattern memory 52 are combined with the chord (chord) TYPE (type: m, sus4
7, dim, etc.) and ROOT (root note) are used to read chord sounds and bass sounds corresponding to TYPE and ROOT using a table reference function. As will be described later, in the present embodiment, for the bass sound, the name of the bass sound read out from the table 50 is used. The processing for further conversion is performed.

The chord pattern memory 51 stores chord performance patterns for each style such as rock and jazz in a form corresponding to Cmaj. The base pattern memory 52 also
The base pattern is stored as Cmaj for each style. The rhythm pattern memory 53 also stores a rhythm pattern indicating a percussion instrument performance pattern for each style. When a style number is designated by the switch 3, a chord pattern, a base pattern, and a rhythm pattern corresponding to the designated style number are stored in the respective memories 51 to 51.
53. In this embodiment, each pattern is stored for two measures, and the data of each pattern is read out every eighth beat.

The CPU 6 has a tempo clock circuit 7 to 8
A clock interrupt occurs every minute. When this interrupt occurs, the CPU 6 reads out the chord, bass and rhythm pattern data from the pattern memories and performs the automatic accompaniment process. The program memory 8 composed of a ROM stores processing programs such as channel assignment and tone color selection, in addition to programs to be described later. Also, RA
The working memory 9 composed of M is used as a work area when executing these programs.

The tone generator 10 includes a sound source, and a tone signal output from the tone source is guided to a sound system 11, where it is acoustically output as a tone.

Next, the operation of the above electronic musical instrument will be described with reference to FIG.
This will be described with reference to the following flowchart.

FIG. 3 shows a main flowchart. When the power is turned on, initialization of various registers and the like assigned to the working memory 9 is performed (n1).
The presence / absence of an event is determined at the input unit. That is, first, the presence or absence of a key event is determined (n2), and subsequently, the presence or absence of a switch event is determined at n4. If there is a key event, the flow branches to a key event subroutine (n3).
If there is a switch event, the flow branches to a switch subroutine (n5).

FIG. 4 shows the n3 key event subroutine.

First, the register MOD is examined. The register MOD is set to one of data 0, 1, and 2. MOD = 0 indicates the normal mode, and the entire key range is allocated for the melody. MOD = 1 indicates the Finguard 1 mode. In this mode, the left key range is assigned for accompaniment, and the automatic accompaniment of a chord having one key code depressed in the left key range as a root note is performed. MOD = 2 indicates the Finguard 2 mode. In this mode, automatic accompaniment is performed by detecting chords from the key codes of all keys depressed in the left key range.

In the above n10, when MOD = 0,
That is, in the normal mode, the entire key range becomes the target range of the melody sound, so that it is not necessary to perform processing such as chord detection from the key pressed state. Therefore, when MOD = 0, n
The program proceeds to step S17, where sound generation and mute processing are performed, and the process returns.

When n10 and MOD = 1 or 2,
It is determined whether or not the position of the key where the event has occurred is in the left key area (n11). If it is a right key area key event, the process proceeds to n17. If it is in the left key range, a code is detected at n12, and on condition that the code is correctly detected, the root note of the detected code is set in the register RT and the type of the detected code is set in the register TP. Then, it is determined whether or not MOD = 2 (n15), and in the fine guard 2 mode, the code of the tone name of the lowest key in the left key range is set in the register LN. The lowest note in the key press sound corresponds to the specific bass note name of the present invention.

If the code is not correctly detected in n13, for example, if only two key presses are detected, the process returns without performing the processing of n14 and below. If MOD = 1 even if the code is correctly detected, the process returns without performing the process of n16. This is n16
Is a preparation step for bass sound control according to the present invention, but such control is not required in the Finguard 1 mode.

FIG. 5 shows the switch subroutine of n5 in FIG.

The type of the switch having the event is set to n20.
And n21. When there is an ON event of the style selection switch, the style number of the operated switch, that is, the selected style number is set in the register STYL (n22). Subsequently, the register RUN is determined. RUN is a register that is inverted by operating a start / stop switch as described later, and RUN = 1 indicates an automatic accompaniment mode. Therefore, when RUN = 1, the accompaniment channel of the sound source is first silenced (n24), and a tone corresponding to the style number set in the register STYL is set to each channel of the sound source corresponding to the accompaniment track (n2).
5).

On the other hand, when it detects the start / stop switch event (n21), by inverting the RUN (n26), as a result, it resets the clock CLK when it becomes RUN = 1, set the F _H in RT (N28). The clock CLK counts for two measures, that is, 64 times with an accuracy of 1/8 beat, and then returns to 0 again.
Is a register that starts counting from. F _H set to RT is a default value indicating that no code has been detected. At n28, CLK is reset and RT is set to an initial value to prepare for automatic accompaniment. If it is detected in n27 that RUN = 0, the flow advances to n29 to mute the accompaniment channel of the sound source and return.

Six tracks for accompaniment are prepared, tracks 0 to 2 are allocated for chord performance, track 3 is allocated for bass performance, and tracks 4 and 5 are allocated for rhythm performance.

If the event switch is neither the style selection switch nor the start / stop switch, it is determined that the mode switch has been operated, and the flow advances to the mode switch subroutine of n30.

FIG. 6 shows the mode switch subroutine.

First, the mode number of the operated switch, that is, the selected mode number is set in MOD. As described above, when MOD = 0, normal mode, when MOD = 1, Finguard 1 mode,
When MOD = 2, the mode is the Finguard 2 mode. n4
At the time of the MOD = 0 is 1, mute the accompaniment for the channel of the sound source (n42), and then returns set the F _H to RT. When MOD = 1 or 2, automatic accompaniment is performed. Therefore, the tone output corresponding to the depressed sound in the left key range of the sound source, which has been output up to that point, is muted. Preparation processing for automatic accompaniment such as setting a tone corresponding to STYL to the channel (n45) is performed and the routine returns.

FIG. 7 shows an interrupt flowchart executed when there is an interrupt (every eighth beat) from the tempo clock circuit 7.

On condition that RUN = 1, first, RT
= See whether or not the F _H. When a chord is detected in the left key range, the root of the detected chord should be set in RT at n14 in FIG.
To n52. In n52, the first track number 0 assigned for chord performance is set in the register TRK. Then, in n53, the chord performance is performed on all the chord performance tracks (n5).
3). Similarly, a bass performance is performed at n56, and n5
At 8, the rhythm performance for the track numbers 4 and 5 is performed (n58). When the chord performance, the bass performance, and the rhythm performance are completed for each of the tracks 0 to 5, the CLK is advanced by one (n61), and it is determined whether or not two measures have been completed (n62), and the two measures have been completed. Is set to 0
And return.

FIG. 8 shows the chord playing subroutine of n53 in FIG.

First, at n70, the style number set to STYL and the style number T from the code pattern memory 51 in which the code pattern is stored in Cmaj.
Track number set in RK and current C
The code data corresponding to the numerical value of LK is read and set in the register KC. The code pattern memory 5
The code data stored in 1 is stored as key code data. If the code data read from the code pattern memory 51 is FF _H , it means that the sound is muted. This determination is made in n71 and KC = FF
_{If H} , a key-off signal and a track number of TRK are output to the sound source (n78) and the process returns. When the read code pattern data is not FF _H , KC
Module function mo that divides the contents of by 12 and returns the remainder
d, and the result is set in the register NT (n
72). Since the divisor 12 is the number of semitones of one octave, the value set in NT is the corresponding number of the note name code where C is 0. In the next n73, the function NCTBL is executed, and the pitch correction value D of the pitch name code obtained in n72 is obtained by referring to the chord tone conversion table of the table 50. FIG. 9 shows a chord tone conversion table having a two-dimensional array of NT and TP for each STYL number. Function NCTB
Since the arguments of L are STYL, TP, and NT, FIG.
The pitch correction value D can be obtained from the table shown in FIG. In n74, the corrected code pattern data is obtained by adding D to KC. Since the chord pattern data is stored in the chord pattern memory 51 in Cmaj, in order to convert the chord pattern data into data having an actual pitch, the root note data is set in n75.
Add the contents of T. Further, if the addition result of n75 does not fall within the predetermined octave, octave increase / decrease processing (sound range restriction processing) is performed (n76). This range-of-range limiting technique is disclosed in, for example, JP-A-63-193199. Thereafter, the set data of the key-on signal, KC, and TRK are output to the sound source and the process returns (n77).

FIG. 10 shows a bass performance subroutine.

First, from the base pattern memory 52 to ST
Style number set to YL, current CL
The base pattern data corresponding to the numerical value of K is read and set in KC (n80). Subsequently, it is determined whether the read data corresponds to the data FF _H to be silenced (n81), and if yes, the process proceeds to n91 to output a key-off signal and TRK track number data to the sound source and return. Base pattern data is FF _H
Otherwise, at n82 and n83, n72 and n72 in FIG.
The same processing as n73 is performed. That is, the tone name code of the read base pattern data is obtained (n82), and ST
The function NCTBL with YL, TP, and NT as arguments is executed to obtain a pitch correction value D (n83). The table used at this time is a bass sound conversion table. Also, n
At 84, base pattern data to which the correction value has been added is determined.

Next, the MOD is judged (n85).
If = 0 or 1, the process proceeds to n86, and the root tone data of the detection code set in RT is added to the base pattern data to which the pitch correction value D has been added, similarly to n75 in FIG. However, when MOD = 2, a module function mod with a divisor of 12 is executed, and the value added to the base pattern data is changed according to the result. That is, KC
If mod12 = 0 is not satisfied, that is, if the pitch name of the bass tone is not the pitch name of the root note, the process proceeds to n86. If the pitch name of the bass tone is the tone name of the root tone, the process proceeds to n88. At n86, the root note data of the detected code is added to the base pattern data as described above. Therefore, in combination with the processing of n84, the pitch is converted into a pitch corresponding to the detected chord here. On the other hand, when KCmod12 = 0, n8
In step 4, the lowest tone set in LN is added to the base pattern data to which the pitch correction value D has been added. Therefore, when the pitch name of the bass tone is the tone name of the root note of the detected chord, the bass tone is converted to the tone name of the lowest tone among the key-depression sounds.

After the above processing is completed, the same gamut restriction processing as in n76 of FIG. 8 is performed (n89), and a key-on signal, KC and TRK data are output to the sound source and the routine returns. (N90).

FIG. 11 shows a rhythm performance subroutine.

The STYL style number, TRK track number, CLK
Was set in the register RN reads the rhythm pattern data corresponding to the number of clocks (n100), unless the read data is FF _H, and outputs key-on signal, the track number of rhythm number and TRK of RN to the sound source ( n102) Return. It does not do anything at the time of the RN = FF _H.

In the embodiment, the base pattern data is sequentially read from the memory by preparing the base pattern memory 52. However, the base pattern data is created on a program to generate the base pattern data in real time. You can also make it. In addition, the specific bass note name is the lowest note name of the key pressed sounds, but is not limited thereto, and may be, for example, a sound pressed on another keyboard.

In the embodiment, the lowest note of the key pressed chord is used as the denominator of the fractional chord. A specific example of chord determination in this case will be described. If the key is depressed as shown in FIG. 1D, this chord is Am7, so the numerator is Am7. And since the lowest note is the G sound, the denominator is G, ie the fraction code Am
/ G.

[0044]

As described above, the predetermined pitch in the bass pattern is converted to the pitch of the pitch name corresponding to the denominator of the fraction code, so that any base pattern does not become a dissonant tone. in the exception of the intended base pattern dissonance from), forming a free base pattern discomfort
I can . As a result, a base pattern having various variations can be applied to the fractional chords, and the automatic accompaniment using the fractional chords and the complicated variation performance are compatible, and a rich performance can be achieved.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an operation of the present invention.

FIG. 2 is a block diagram of an embodiment of the present invention.

FIG. 3

FIG. 8 and

FIG.

FIG. 11 is a flowchart showing the operation of the embodiment.

FIG. 9 is a configuration diagram of a chord tone conversion table.

[Explanation of symbols]

 50-chord / bass tone conversion table 51-chord pattern memory 52-bass pattern memory 53-rhythm pattern memory CNVT-part corresponding to pitch conversion means

Claims (1)

(57) [Claims] 1. A chord detecting means for detecting a chord including a fractional chord from pitch information of a performance, and when a chord detecting means detects the fractional chord, a specific note name relating to a denominator of the fractional chord is specified. Specifying means for specifying as a base note name; base pattern generating means for generating a base pattern including a rhythm pattern and a pitch pattern of a bass performance based on a numerator of a fractional chord detected by the chord detecting means; Changing means for changing a predetermined pitch among the pitches of the pitch pattern of the bass performance to be the specific bass note name specified by the specifying means; and An automatic accompaniment device for an electronic musical instrument, comprising: automatic bass performance means for performing an automatic bass performance based on the changed bass pattern.