JP2541021B2 - Electronic musical instrument - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention automatically reads accompaniment pattern data stored in advance as time passes, and automatically generates accompaniment sounds such as chord constituent tones, bass tones and percussion instrument sounds based on the data. The present invention relates to an electronic musical instrument having a built-in accompaniment device.

[0002]

2. Description of the Related Art Conventionally, this type of apparatus is disclosed in
-71293-71296 gazette, actual Kaihei 1-1012.
As disclosed in Japanese Patent Publication No. 99, etc., a key performance state quantity such as a key touch amount and a key press interval is detected, and the detected key performance state quantity is compared with a predetermined reference value to determine the same. When the key performance state amount exceeds the reference value, the accompaniment pattern data read from the accompaniment data memory is switched, and the accompaniment pattern can be automatically switched according to the key performance state during the performance of the music. The accompaniment sound suitable for the music being played is generated, and the accompaniment sound is not monotonous.

[0003]

However, in the above-mentioned conventional apparatus, since the reference value compared with the performance state quantity is constant, the musical composition of the musical composition, the habit of the performer, and the tone color of the musical performance sound. Change the detected key playing state amount, the accompaniment patterns are switched too frequently during the performance of the song, or the accompaniment patterns are rarely switched. There is a problem that does not become. The present invention has been made to address the above problems, and its purpose is to
An object of the present invention is to provide an electronic musical instrument in which accompaniment patterns can be switched at an appropriate frequency according to the performance state of a performance operation means such as a keyboard.

[0004]

In order to achieve the above-mentioned object, the structural features of the invention described in claim 1 are plural.
And the key or Ranaru performance operation means of, key in a different section within the
Performance state amount detecting means for detecting the difference in the number of times of each operation as a performance state amount, an accompaniment data memory storing a plurality of types of accompaniment pattern data, and a plurality of types of accompaniment pattern data.
When the performance state quantity detected by the performance state quantity detecting means and a predetermined reference value are compared and the performance state quantity belongs to one of the ranges with the reference value as a boundary, the reading means Switching means for switching the read accompaniment pattern data to different accompaniment pattern data, and generating a musical tone signal in response to the operation of a plurality of keys of the performance operating means and an accompaniment tone signal in accordance with the read accompaniment pattern data. Tone signal generating means , switching condition selecting means for selecting switching conditions by the switching means, and at least one of the detected playing state quantity and the reference value in accordance with the switching conditions selected by the switching condition selecting means. <br/> in that a changing means for.

[0005]

[0006]

[0007]

Further, the structural feature of the invention described in claim 2 is that the performance state quantity detection hand of the invention according to claim 1 is
Is the difference between the average key touches of keys in different sections.
Is replaced by a performance state quantity detecting means for detecting as a performance state quantity .

Further, the structural feature of the invention described in claim 3 is that the performance operation means comprises a plurality of performance operators.
Performance state quantity detection that detects performance state quantity by performance operation means
Accompaniment that stores output means and accompaniment pattern data of multiple types
Data memory and one of multiple types of accompaniment pattern data
The reading means for reading and the playing state quantity detecting means detect
Compare the performance state quantity that was issued and a predetermined reference value, and compare
Read when the playing state quantity belongs to one range with the same reference value as the boundary.
Different accompaniment pattern data read by the projecting means.
Switching means for switching to the accompaniment pattern data
A musical tone is transmitted according to the operation of a plurality of performance operators of the performance operation means.
Signal and the read accompaniment pattern data
Tone signal generating means for generating an accompaniment tone signal according to the data
In the electronic musical instrument provided with, a tone color selecting means for selecting a tone color of a tone signal generated in response to an operation of the performance operating means, and the detected playing state amount or the tone state amount according to the tone color selected by the tone color selecting means And a changing means for changing at least one of the reference values.

Further, the structural features of the invention described in claims 4 to 7 are such that the plurality of performance operators of the invention according to claim 3 are composed of a plurality of keys, and the invention is related to the invention of claim 3. <br/> The performance state quantity detecting means is used as the performance state quantity in a predetermined section.
Means to detect the number of key operations in the interval, within a predetermined interval
For detecting the average key touch amount of keys in different groups
Means for detecting the difference in the number of times each key operation is performed, and
Detect the difference between the average key touch amount of keys in different sections
It is configured by each means .

The structural feature of the invention described in claim 8 is the same as that of the invention according to claim 3 in that the performance operation means, accompaniment data memory, reading means, switching means and tone signal generating means are provided. At the same time, in the electronic musical instrument provided with the performance state quantity detecting means for detecting the performance state quantity by the performance operating means for each predetermined section, the switching means based on the performance state quantity of the previous section detected by the performance state quantity detecting means. A switching evaluation value storage means for storing a switching evaluation value indicating whether or not the read accompaniment pattern data has been switched, and the detected performance state quantity or the reference value in accordance with the stored switching evaluation value. The change means for changing at least one of them is provided.

[0012] The billing structural feature of the invention according to claim 9 to 12, and a plurality of performance operators and performance state quantity detecting means of the invention according to claim 8, wherein the claim 4-7 It is composed of the same kind as the invention according to.

Further, according to the thirteenth aspect of the present invention, there is provided a first performance operating means composed of a plurality of performance operators, and a second performance operating means composed of a plurality of performance operators.
Accompaniment data memory storing a plurality of types of accompaniment pattern data, reading means for reading one of a plurality of types of accompaniment pattern data, and operation of the read accompaniment pattern data for a plurality of performance operators of the second performance operation means. And a musical tone signal generating means for generating a musical tone signal according to the operation of a plurality of performance operators of the first performance operating means and for generating an accompaniment tone signal according to the converted accompaniment pattern data. An electronic musical instrument comprising: a first performance state quantity detecting means for detecting a performance state quantity by the first performance operating means; a second performance state quantity detecting means for detecting a performance state quantity by the second performance operating means; Selecting means for selecting one of the respective performance state quantities detected by the first or second performance state quantity detecting means; and the selected performance state quantity and a predetermined reference value. Compare to the play state quantity is in fact provided a switching means for switching to a different accompaniment pattern data accompaniment pattern data read by the reading means when belonging to one of the ranges the boundary of the reference value.

The structure of the invention described in claim 14 is the same as that of the invention according to claim 13, in which the first performance operating means, the second performance operating means, the accompaniment data memory, the reading means, and the converting means. In an electronic musical instrument provided with a means and a tone signal generating means, a first performance state quantity detecting means for detecting a performance state quantity by the first performance operating means and a second performance for detecting a performance state quantity by the second performance operating means. At least one of the playing state quantities detected by the state quantity detecting means and the first or second playing state quantity detecting means, or the playing state quantity detected by the first playing state quantity detecting means. Selection means for selecting one of the performance state quantities in consideration of the performance state quantity detected by the second performance state quantity detecting means and the selected performance state quantity are compared with a predetermined reference value to perform the same performance. In the amount of state is provided and switching means for switching to a different accompaniment pattern data accompaniment pattern data read by said reading means when belonging to one of the ranges the boundary of the reference value.

According to a fifteenth aspect of the present invention, the plurality of performance operators of the first performance operating means of the thirteenth or fourteenth aspect of the invention are provided with a plurality of keys for playing a melody. In addition to the above, the plurality of performance operators of the second performance operating means of the present invention are composed of a plurality of keys for playing chords.

[0016]

[Advantages and effects of the invention] Claims configured as described above
In the inventions 1 and 2 , the switching means compares the performance state quantity detected by the performance status quantity detection means with a predetermined reference value, and the performance status quantity is set to one of the reference values. When belonging to the range, the accompaniment pattern data read by the reading means is switched to different accompaniment pattern data, so that the pattern of the accompaniment sound signal generated from the musical sound signal generating means is automatically changed according to the performance state amount in the performance operation means. Can be switched. On the other hand, if the performer changes the switching condition using the switching condition selecting means, the changing means changes at least one of the detected playing state quantity and the reference value according to the changed switching condition. The criterion for switching in the means is changed according to the player's selection.

[0017] In the invention according to claim 1, 2 This is
The performance operation means is composed of multiple keys
Detecting means, a play state quantity, and detects the difference in the number of operations of the key at different intervals within and different in the interval key the difference between the average key touch weight respectively, the accompaniment pattern is in the key play, It is automatically switched according to the difference in the number of key operations or the difference in the average key touch amount.

As a result, according to the inventions according to claims 1 and 2 , the player has a peculiar habit even when playing music having different musical compositions and playing music with different tone colors. However, before the performance of the musical composition, the player appropriately selects the switching condition in accordance with the musical composition, the tone color and the habit, so that the difference in the number of key operations and the even key touch amount during the performance of the musical composition. depending on each Starring Kanade state quantity difference, accompaniment patterns become automatically switch at a suitable frequency, it is possible to play music with high quality.

[0019] Also in the invention according to the claims 3 to 7, the pattern of accompaniment sound signal generated from the comfortable sound signal generating unit, the performance state quantity of the performance operation means (e.g.,
It is automatically switched according to the key performance number, the average key touch amount, the key operation number difference, the average key touch amount difference, etc.). In this case, in particular, the changing means changes at least one of the detected playing state quantity and the reference value according to the timbre selected by the timbre selecting means. Is automatically changed according to the tone of the performance sound.

As a result, according to the inventions according to claims 3 to 7 , even if the performance state of the performance operating means changes due to a difference in the tone color of the performance sound or switching of the tone color during the performance ( For example, in the case of strings tone,
As the playing becomes slower and the key touches are less likely to change), the switching conditions are automatically switched according to the tone color of the performance sound, so the performer does not have to perform special operations during the performance. Accompaniment patterns can be automatically switched at an appropriate frequency according to performance state quantities such as the number of key operations, the average key touch amount, the difference in key operation numbers, and the difference in average key touch amount, and high quality music can be played. be able to.

[0021] Also in the invention according to the claims 8 to 12, the pattern of accompaniment sound signal generated by the easy tone signal generation means, playing state quantity of the performance operation means (e.g., a key operation number, the average key touch Volume, key operation number difference, average key touch amount difference, etc.). In this case, in particular, the evaluation value storage means stores the switching evaluation value indicating whether or not the accompaniment pattern data in the previous section has been switched, and the changing means detects the detection value according to the stored switching evaluation value. Since at least one of the played state quantity and the reference value is changed, the switching reference in the switching means is automatically changed according to whether or not the accompaniment pattern was previously switched.

As a result, according to the inventions of claims 8 to 12 , it is possible to prevent the accompaniment patterns from being continuously switched in a short period of time. This allows the performer to perform an appropriate accompaniment pattern according to the performance state quantity such as the number of key operations, the average key touch amount, the difference in the number of key operations, and the difference in the average key touch amount without performing any special operation during the performance. It can be switched automatically at any frequency, and high quality music can be played.

Further, in the invention according to claim 13 configured as described above, each performance state amount by the first and second performance operating means is respectively detected by the first and second performance state amount detecting means. Detected and these first or second
One of the playing state quantities is selected by the selecting means, and the switching means compares the selected playing state quantity with a predetermined reference value, and the playing state quantity belongs to one range with the same reference value as a boundary. At this time, the accompaniment pattern data read by the reading means is switched. As a result, the pattern of the accompaniment sound signal generated by the musical sound signal generating means is automatically switched during the performance in accordance with either the performance operation of the first performance operation means or the performance operation of the second performance operation means. .

Further, in the invention according to claim 14 configured as described above, the selection means includes at least one of the performance state quantities detected by the first or second performance state quantity detection means. Either one of the playing state quantities or the playing state quantity in which the playing state quantity detected by the second playing state quantity detecting means is added to the playing state quantity detected by the first playing state quantity detecting means is selected. The pattern of the accompaniment sound signal generated from the musical sound signal generating means is either a performance operation in the first or second performance operating means or both performance operations in the first and second performance operating means during performance. It is automatically switched according to.

In the case of the invention according to claims 13 and 14, for example, as in the invention according to claim 15 , the first
The plurality of performance operators of the performance operation means are composed of a plurality of keys for playing a melody, and the plurality of performance operators of the second performance operation means are composed of a plurality of keys for playing a chord. Therefore, the automatic switching of the accompaniment pattern is selectively performed according to the melody performance or the chord performance.

As a result, according to the thirteenth to fifteenth inventions, the player can select various performances such as melody performance and chord performance suitable for automatic switching of accompaniment patterns by using the selection means. Since the switching of the accompaniment pattern is controlled based on this selection, the frequency of switching the accompaniment pattern during the performance can be made appropriate.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the entire electronic musical instrument according to the present invention. This electronic musical instrument includes a left keyboard 11, a right keyboard 12, and an operation panel 20. The left keyboard 11 includes a plurality of keys and is used for playing a chord. The right keyboard 12 also comprises a plurality of keys and is used for playing a melody. The key press / release of each key is detected by opening / closing of a plurality of key switches provided for each key in the key press detection circuit 13, and a key touch (initial touch) accompanying the key press / release is detected. The key touch detection circuit 14 detects it.

On the operation panel 20, as shown in FIG.
A plurality of tone color selection switch groups 21, a plurality of accompaniment style switch groups 22, a tempo volume 23, a start switch 24a, a stop switch 24b, a selection switch 2
5, switching condition switch 26, determination area switch 27
And lamps 28a and 28b are provided.

The tone color selection switch group 21 includes a violin,
It corresponds to a plurality of tone colors such as guitar and piano, and each switch designates one tone color for a melody tone from the plurality of tone colors. The accompaniment style switch group 22 corresponds to a plurality of accompaniment styles such as march and rock, and each switch specifies one of the plurality of accompaniment styles. The tempo volume 23 sets the tempo of automatic accompaniment.
The start switch 24a instructs the start of the automatic accompaniment, and the stop switch 24b instructs the stop of the accompaniment. The selection switch 25 is used for playing the automatic accompaniment.
Whether or not to automatically switch the accompaniment pattern is selected according to the key performance state (key touch or key press number). The switching condition switch 26 switches the reference value of the key performance state quantity for automatically switching the accompaniment pattern in three steps. The judgment area switch 27 is
Whether to perform the switching in accordance with only the key playing state of the left keyboard 11, whether to perform the switching only in the key playing state of the right keyboard 12, or whether to perform the switching in accordance with the key playing state of both keyboards 11 and 12. It is something to choose. Lamp 28
Each of a and 28b is composed of a light emitting diode and indicates an accompaniment pattern (first or second pattern) during performance. The operation of each of the switches 21 to 27 is detected by the switch operation detection circuit 20a, and the lamps 28a and 28b are connected to the display control circuit 20.
Lighting and extinction are controlled by b.

The key press detection circuit 13a, the switch operation detection circuit 20a and the display control circuit 20b are connected to a bus 30, and the bus 30 includes a tone signal forming circuit 40, a microcomputer section 50 and an accompaniment data memory 60. It is connected. The tone signal forming circuit 40 is provided with a plurality of tone signal forming channels, and each tone signal forming channel includes a key code KC, a volume data VOL, a key-on signal KON, etc. supplied from the microcomputer unit 50 via the bus 30. Based on the control data, it forms a melody sound signal and an accompaniment sound signal, such as a piano and a clarinet, whose pitch changes, and a percussion instrument sound signal such as a drum or cymbal (defined as a part of the accompaniment sound signal in this case). Form and output. The output of the tone signal forming circuit 40 is an amplifier 4
1 is connected to the speaker 42.

The microcomputer unit 50 includes a bus 30.
Memory 51, tempo clock generator 52, CPU 53 and working memory 5 connected to
It consists of 4. The program memory 51 is composed of a ROM and stores a program corresponding to the flowcharts shown in FIGS. The tempo clock generator 52 is composed of a variable frequency oscillator, and is connected from the CPU 53 to the bus 30.
The tempo clock signal is generated at a cycle according to the tempo control data supplied via the. The period of the tempo clock signal corresponds to the timing obtained by dividing a quarter note into 24. After turning on the power switch (not shown), the CPU 53 continues to execute the “main program” corresponding to the flowchart shown in FIG. The "interrupt program" corresponding to the flowchart shown is executed by interruption. The working memory 54 is constituted by a RAM, and temporarily stores data necessary for executing the program.

The accompaniment data memory 60 is composed of a ROM, and has a style table STLTBL and a pattern table PTNT.
BL, performance data table PLDT, switching condition table CG
In addition to the CTBL and the coefficient table KTBL, it also has an area for storing other accompaniment data. As shown in FIG. 3, the style table STLTBL is a storage area STLTBL designated by a style number STLN representing each accompaniment style.
It is divided into (STLN), and each storage area STLTBL (STLN) stores the bar number BAR. Here, the bar number BAR indicates the bar number of one cycle of the first and second accompaniment patterns of each accompaniment style. In the automatic accompaniment apparatus of this embodiment, the first accompaniment pattern and the second accompaniment pattern are provided for each accompaniment style.
An accompaniment pattern is prepared, and the second accompaniment pattern has a greater sense of excitement (more colorful) than the first accompaniment pattern.

As shown in FIG. 4, the pattern table PTNTBL contains the accompaniment patterns (first and second accompaniment styles) of each accompaniment style.
Storage area corresponding to accompaniment pattern) PTNTBL (STLN, PTRN)
Each storage area PTNTBL (STLN, PTRN) is designated by a style number STLN and a pattern number PTRN (0 or 1). Each storage area PTNTBL (STL
N, PTRN) stores a sound addition flag ADD and volume data VOL for each track number TRKN (0 to 8). The track number TRKN represents a chord component sound sequence from 0 to 5, 6 represents a bass tone sequence, and 7, 8 represent a percussion instrument tone sequence. The pronunciation addition flag ADD is "0".
Indicates that the accompaniment sound of each track is a normal sound, and that "1" indicates that the accompaniment sound of each track is an additional sound. Note that the normal sound is always sounded in the first and second accompaniment patterns, and the additional sound is sounded only in the arrangement mode (arrangement flag ARNG = "1") in the first and second accompaniment patterns. Sound. Volume data VOL represents the relative volume of the accompaniment sound for each track.

As shown in FIG. 5A, the performance data table PLDT is a storage area PLDT (STLN, PTRN, PTRN, STRLN, PTRN, corresponding to each accompaniment style, each accompaniment pattern (first and second accompaniment pattern), and each track. TRKN), and each storage area PLDT (STLN, PTRN, TRKN) is specified by a style number STLN, a pattern number PTRN and a track number TRKN. Each storage area PLDT (STLN, PTRN, TRKN) has
Note data NOTE, tone color data TC and bar line data BARL
A series of performance data strings consisting of is stored for each track as time passes. As shown in FIG. 5B, the note data NOTE includes an identification code and an event time EVT.
, Key code KC, key touch KT and key-on time KOT
And a set of data consisting of In this case, the identification code is a note data NOTE
The event time EVT indicates the read timing of the same data NOTE by the time from the beginning of the bar, and the key code KC indicates the pitch of the accompaniment sound based on the major chord as a semitone from the C sound. It is expressed by the number of intervals (however, for percussion sounds, it indicates the type), and key touch KT
Represents the relative volume of the accompaniment sound, and the key-on time KOT represents the duration of the accompaniment sound. As shown in FIG. 5C, the timbre data TC includes an identification code, an event time EVT, and a timbre number.
It consists of one set of data consisting of VOIN. In this case, the identification code indicates that this set of data is the tone color data TC, and the event time EVT is the same data.
The TC read timing is represented by the time from the beginning of the bar, and the timbre number VOIN represents the timbre of the accompaniment sound (however, for percussion instrument sounds, it represents a subtle change of the same sound). The bar line data BARL is composed only of an identification code indicating that the accompaniment sound string is a phrase break.

The switching condition table CGCTBL is shown in FIG. 6 (A).
, The three-stage switching sensitivity SENS (= 0 to 0 set by the switching condition switch 26
Corresponding to 2), 8 types of reference values RDNT, RDVL, RVUP, RVDW,
It stores RNUP, RNDW, LVUP, LVDW. The reference value RDNT is two consecutive sections on the right keyboard 12 (in the case of the present embodiment,
(Each section corresponds to the length of one bar) The accompaniment pattern is switched according to the difference in the number of keys pressed.
The reference value RDVL is for switching the accompaniment pattern according to the difference between the average key touch amounts of two consecutive sections on the right keyboard 12. The reference values RVUP and RVDW are used to switch the accompaniment pattern in the direction of raising the accompaniment pattern and in the opposite direction with respect to the magnitude of the average key touch amount in one section on the right keyboard 12. The reference values RNUP and RNDW are used to switch the accompaniment pattern in the direction in which the accompaniment pattern is raised and in the opposite direction, with respect to the number of keys pressed in one section on the right keyboard 12. The reference values LVUP and LVDW are for switching the accompaniment pattern in the direction of raising the accompaniment pattern and in the direction opposite thereto with respect to the magnitude of the average key touch amount in one section on the left keyboard 11.

As shown in FIG. 6 (B), the coefficient table KTBL has eight reference values RDNT, RDVL, RVUP, RVDW, RNUP, RNDW,
Table coefficient TK (i) (i corresponding to LVUP and LVDW respectively)
= 0 to 7) is stored. It should be noted that these table coefficients TK (i) are multiplied by the respective reference values when the timbre of the melody is a strings timbre.

In the area for storing other accompaniment data,
A chord detection table for detecting a chord and a conversion table for converting an accompaniment tone into a constituent tone of the chord based on the detected chord are provided.

Next, the operation of the above embodiment will be described with reference to the flow chart. First, when a power switch (not shown) is turned on, the CPU 53 executes step 1 in FIG.
At 00, the execution of the “main program” is started, and at step 102, the tone signal forming circuit 40 and the working memory 54 are set to the initial state. Particularly, in the initialization process, the pattern number PTRN is set to "0" representing the first accompaniment pattern, the switching evaluation coefficient CF is set to "1", and the tone color coefficient K (i) (i = 0 to 7) are set to "1" respectively. Each of these coefficients CF, K (i) is used to evaluate the automatic switching of accompaniment patterns, and the switching evaluation coefficient CF is set to "1" when the accompaniment patterns are switched in the previous section. When the accompaniment pattern was not switched in the previous section, `` 0.
5 '', the tone color coefficient K (i) is set to the table tone color coefficient TK (i) when the melody tone color is a strings type tone color, and the melody tone color is a string type tone color. It is set to "1" when the tone color is other than the tone color. After this initial setting, the CPU 53 continues to execute the circulation processing consisting of steps 104 to 110.

When one of the keys on the left keyboard 11 and the right keyboard 12 is pressed and released during the circulation processing consisting of steps 104 to 110, the CPU 53 determines "YE" at step 104.
If "S" is determined, the "key event routine" is executed in step 106. This "key event routine"
Step 120, as shown in detail in FIG.
In this routine, the generation of the melody sound is controlled and the performance chord is detected in accordance with the performance of the left keyboard 11 and the right keyboard 12.

When one of the keys on the right keyboard 12 is depressed, the CPU 53 proceeds to "Y" at steps 122 and 124.
ES ”, and a key-on process is executed at step 126. In this key-on processing, a key-on signal KON representing a key depression, a key code KC representing a key name depressed,
And a key touch signal KT indicating the strength of the key touch detected by the key touch detection circuit 14 is output to the tone signal formation circuit 40, and the formation circuit 40 has a pitch represented by the key code KC, Further, a melody sound signal having a volume proportional to the key touch signal KT is formed, and the melody sound signal is output to the speaker 42 via the amplifier 41. The tone color of this melody tone signal is determined by operating the tone color selection switch group 21 by the process described later. After the processing of step 126,
In step 128, a value representing the strength of the key touch is set as the key touch detection value VEL, and in step 130, the detection value VEL is added to the key touch amount RVSM for the right keyboard 12, and the right keyboard is also added. "1" is added to the number of pressed keys RNSM for 12. The key touch amount RVSM is a variable for accumulating the strength of each key touch on the right keyboard 12 over one measure together with the key touch amount LVSM for the left keyboard 11 which will be described later. The number of keys pressed RNSM is the left keyboard 1
This is a variable for accumulating the number of key presses on the right keyboard 12 over one measure together with the number of key presses for 1 LNSM.

If any of the keys on the right keyboard 12 is released, "YE" is entered in steps 122 and 124, respectively.
S "and" NO "are determined, and a key-off process is executed in step 132. In this key-off process, the key code KC and the key-off signal that represent the released key name.
KOF is output to the tone signal forming circuit 40,
0 stops the formation of the melody tone signal of the pitch represented by the key code KC. As described above, when any key of the right keyboard 12 is pressed and released, a melody sound corresponding to the pressed and released key is emitted from the speaker 42 and the right keyboard 12 is released.
The key touch amount RVSM and the number of pressed keys RNSM are updated according to the key pressed.

On the other hand, when one of the keys on the left keyboard 11 is pressed or released, the CPU 53 makes a "NO" determination at step 122 and executes chord detection processing at step 134. In this process, the chord detection table in the accompaniment data memory 60 is referred to in accordance with the combination of the keys pressed on the left keyboard 11, the chord is detected, and the root note and type of the detected chord are detected. Is stored as a chord root note CRT and a chord type CTP. After the process of step 134, if the key operation of the left keyboard 11 is related to the key depression, "Y" in step 136 is executed.
Based on the determination of “ES”, the key touch amount LVSM and the number of key presses LNSM for the left keyboard 11 are newly set by the processing of steps 138 and 140 similar to the processing of steps 128 and 130. Updated in response to key presses.

If any switch or volume on the operation panel 20 is operated during the circulation process consisting of steps 104 to 110 in FIG. 8, the CPU 53
In step 108, “YES” is determined, and step 1
At 10, a "switch event routine" is executed. This "switch event routine" comprises steps 150 to 174, as shown in detail in FIG.
In the same routine, each switch and volume 21-27
In response to the operation, the tone color of the melody sound is set, and the generation of the accompaniment sound is controlled.

When any one of the tone color selection switch groups 21 is operated, the CPU 53 determines the tone color number data VOIN representing the operated tone color selection switch 21 in step 154 by the determination processing in step 152. Output to 40. As a result, the tone signal forming circuit 40 controls the tone color of the melody tone. Step 15
After the processing of 4, if the tone color of the melody tone selected by the tone color selection switch group 21 is a strings tone color, the CPU 53 makes a "YES" determination at step 156.
On the basis of the determination of, it is determined in step 157 that the coefficient table KTBL
(FIG. 6 (B)), the tone color coefficient K (i) (i = 0 to 7) is stored in the table KTBL as a table tone color coefficient value.
Set to TK (i) (i = 0 to 7) respectively. If the tone color of the melody tone selected by the tone color selection switch group 21 is other than the strings tone color, step 15
Based on the determination of "NO" in No. 6, each timbre coefficient K (i) is set to "1" in step 158.

When any of the accompaniment style switch groups 22 is operated, the CPU 53 sets the value representing the accompaniment style switch 22 operated in step 161 as the style number STLN by the determination processing of step 152, and 162 pointers for addressing performance data storage areas PLDT (STLN, PTRN, 0) to PLDT (STLN, PTRN, 8) for each track designated by style number STLN and pattern number PTRN in 162 , Current bar CBAR and current timing CTIM. In addition,
The current bar CBAR has a progressive bar value of 0 in each pattern.
.About.n-1 (n is the number of bars corresponding to the repetition period of each pattern), and the current timing CTIM represents each progress timing obtained by dividing a quarter note into 24 within each bar.
In this case, the process of step 162 is usually unnecessary in connection with the process of step 202 of FIG. 12 described later, but even if the accompaniment style switch group 22 is operated during the automatic accompaniment operation. , It is necessary to start the automatic accompaniment with the newly specified accompaniment style from the correct progress position.

When the tempo volume 23 is operated,
The CPU 53 executes tempo setting processing in step 160 based on the determination processing in step 152. In this tempo setting process, the tempo control data corresponding to the operation position of the tempo volume 23 is transmitted to the tempo clock generator 5.
The generator 52 outputs a tempo clock signal to the CPU 53 at a cycle corresponding to the tempo control data. The period of the tempo clock signal corresponds to the timing obtained by dividing a quarter note into 24.

When the start switch 24a is operated,
The CPU 53 sets the run flag RUN to “1” in step 166 and executes a “pattern activation routine” in step 168 to start the operation of the automatic accompaniment in the determination processing in step 152. When the stop switch 24b is operated, the CPU 53 sets the run flag RUN to “0” in step 170 by the determination process in step 152, and
In step 2, the tone signal forming circuit 40 is silenced to stop the automatic accompaniment operation. Run flag RU
N is "0" to indicate that automatic accompaniment is stopped, and "1"
Indicates that the accompaniment is operating.

When the selection switch 25 is operated, the CPU
53 is step 1 based on the determination processing of step 152.
At 65, the switching selection flag CNGF is inverted for each operation (switching from "1" to "0" or from "0" to "1"). In this case, the switching selection flag CNGF is set to “
"1" represents a mode in which the accompaniment pattern is automatically switched according to the number of key touches and the number of pressed keys on the left and right keyboards 11 and 12, that is, the pressed state, and "0" represents a mode in which the switching is prohibited. When the CPU 26 is operated, the CPU 53 determines in step 163 the switching sensitivity SE in the determination processing of step 152.
NS is a value corresponding to the operation position of the switch 26 (0 to 2)
Set to. When the determination key range switch 27 is operated, the CPU 53 causes the determination process of step 152 to
In step 164, the determination key range flag RNG is set to a value (0 to 2) corresponding to the operation position of the switch 27. The judgment key range flag RNG is set to "0" by the left keyboard 1
1, the right keyboard 12 is represented by "1", and both keyboards 11 and 12 are represented by "2".

Next, the operation of automatic accompaniment based on the operation of the switches on the left and right keyboards 11 and 12 and the operation panel 20 will be described in detail. First, the case where the switching selection flag CNGF is set to "0" will be described.

As described above, by operating the start switch 24a, the CPU 53 executes the "pattern activation routine" of step 168 of FIG. In this "pattern activation routine", as shown in detail in FIG. 11, its execution is started in step 180, and in step 182, the current timing CTIM, the current bar CBAR and the section flag PERF are set to the initial value " It is set to "0". Each of these variables CTIM and CBAR is as described above, and the section flag PERF is increased by "1" for each bar at the start of the automatic accompaniment operation, and represents the progress position of the accompaniment performance. . After the processing of step 182, step 1
Each key touch amount RVS for the left and right keyboards 11 and 12 at 84
M, LVSM and number of keys pressed RNSM, LNSM are initialized to "0",
In step 186, the section key touch amount QRV (0) to QRV (2), QLV
(0) to QLV (2) and number of section keys QRN (0) to QRN (2), QLN (0) to Q
LN (2) is initialized to "0". The section key touch amounts QRV (0) to QRV (2), QLV (0) to QLV (2) are the left and right keyboards 11, 1
Each key touch amount RVS for the past 3 sections per bar for 2
Indicates the total value of M and LVSM, and the number of section keys QRN (0) to QR
N (2), QLN (0) to QLN (2) are 1 for the left and right keyboards 11 and 12.
It represents the total value of the number of keys pressed RNSM, LNSM for the past 3 sections for each measure. Next, in step 188, the “pattern switching routine” is executed, and in step 190, the “pattern activation routine” ends.

The "pattern switching routine" is shown in FIG.
Steps 200-212, as detailed in FIG.
In step 202, the performance data storage area PLDT (S
Nine pointers for addressing TLN, PTRN, 0) to PLDT (STLN, PTRN, 8) are newly set. Next, CPU5
3 controls the lighting of the lamp 28a if the pattern number PTRN is "0" by the processing of steps 204 to 210, and the lamp 28 if the pattern number PTRN is "1".
The lighting of “b” is controlled, and the execution of the “pattern switching routine” ends. As a result, the lamps 28a and 28b are turned on in accordance with the set pattern number PTRN (initial value is "0", and then switched to "1" or switched to "0" again). .

In this state, when the tempo clock generator 52 outputs a tempo clock signal to the CPU 53 at each timing of dividing a quarter note into 24, the CPU 53 outputs the "main program" (FIG. 8) at each timing. The execution is interrupted, the “interrupt program” is started to be executed in step 220 of FIG. 13, and it is determined “YES” based on the run flag RUN set to “1” in step 222, and step 224 is executed. The processes of 240 to 240 are executed. In this case, switch selection flag
Since CNGF is set to "0", the "automatic conversion routine" of step 238 is not executed.

The CPU 53 executes steps 224, 228,
By the process at 230, the “reproduction routine” is repeatedly executed at step 226 while incrementing the variable i by “1” from “0” to “8”. This "reproduction routine"
As shown in detail in FIG. 14, execution starts at step 250, and at step 252, the style number STL
N, a pattern number PTRN, and a set of performance data pointed to by a pointer for each track are sequentially read from each storage area PLDT (STLN, PTRN, i) specified by a variable i indicating each track. The processing after 254 is executed.

In this case, if the one set of performance data is bar line data BARL, it is judged "YES" in step 254, the pointer of the track is advanced in step 266, and the program is executed again in step 25.
2 and the next performance data on the same track is read. Also, the set of performance data is the note data NO.
If it is TE and the event time EVT is equal to the current timing CTIM, steps 254, 256, 2
At 58, “NO”, “YES”, and “YES” are determined.
The process of the "note routine" of No. 2 is executed to control the generation of musical tones. If the set of performance data is the timbre data TC and the event time EVT is equal to the current timing CTIM, steps 254 and 25 are executed.
6, NO at "NO", "YES", "N"
It is determined to be "O", and the tone color of the generated musical tone is controlled in step 264. After the processing of steps 262 and 264, the pointer of the track is advanced in step 266, and the program returns to step 252 again to read the next performance data of the same track. On the other hand, the one set of performance data is note data NOTE or tone color data TC, and their event time EV
If T is not equal to the current timing CTIM, both are determined to be "NO" in steps 254 and 256, and the "reproduction routine" ends in step 268. As a result, the note data NOTE and the tone color data TC stored in the storage areas PLDT (STLN, PTRN, 0) to PLDT (STLN, PTRN, 8) are sequentially read by the pointer instruction, and their event time EVT is The generation of a musical tone and the tone color of the generated musical tone are controlled each time it becomes equal to the current timing CTIM.

Next, the generation of these musical tones and the control of the tone color of the generated musical tones will be described in detail. First, the control of the timbre will be described. In the process of step 264, the timbre number VOIN and the variable i in the timbre data TC are output to the tone signal forming circuit 40. Tone signal forming circuit 40
Is the tone color of the accompaniment sound of the track specified by the variable i, based on these tone color numbers VOIN and the variable i.
Set to the tone specified by VOIN.

Next, the generation of a musical sound will be described. The arrangement flag ARNG is determined by the determination processing in step 260.
= “1”, or a sound addition flag PTNTBL (STL) in the pattern table PTNTBL specified by the style number STLN, the pattern number PTRN, and the variable i representing each track.
Step 2 only if N, PTRN, i) .ADD is "0"
The process of "note routine" 62 is executed to control the pronunciation of the accompaniment sound. This "note routine" is shown in Figure 1.
As shown in FIG. 5, Steps 270 to 286 are performed, and if the variable i is equal to or less than “6”, in Step 272, “YE
S ”, and the key code KC that constitutes the read note data NOTE in step 274 is determined based on the detected chord root CRT and chord type CTP.
The key code is converted into a key code KC representing a chord component sound or a bass sound corresponding to the chord played in Step 1. If the variable i is “7” or more, “NO” is determined in the step 272, and the conversion process in the step 274 is not performed. This is because the variable i indicates a track of a chord component sound sequence and a base sound sequence by 0 to 6 and a track of a percussion sound sequence by 7 and 8 (see FIG. 4). next,
In steps 276 and 278, the volume data VOL (see FIG. 4) in the pattern table PTNTBL (STLN, PTRN, i) specified by the style number STLN, the pattern number PTRN and the variable i, and the read note data NOTE The volume VOL and the key-off time K are calculated by executing the following mathematical formula 1 based on the key-on time KOT and the key touch KT (see FIG. 5)
OFT (i) is calculated.

[Equation 1] VOL = PTNTBL (STLN, PTRN, i) .VOL + KT KOFT (i) = TIME + KOT In this case, the time TIME is the "count routine" described later.
Represents an absolute elapsed time which sequentially counts up from 0 to 5000, and the key-off time KOFT (i) defines the end timing of the generated musical sound based on the absolute time. After the processing of the step 278,
If the key-off time KOFT (i) becomes larger than "5000" as a result of the above equation 1 by the processing of steps 280 and 282, the same time KOFT (i) is decremented by "50" from the value.
By subtracting “00”, the value is changed to a value less than “5000”.

Next, at step 284, the converted key code KC (however, if the variable i is 7 or 8, the key code KC is not converted), the volume VOL, the key-on signal KON and the variable i are the tone signal forming circuit. It is output to 40. The tone signal forming circuit 40 forms an accompaniment tone signal of the track designated by the variable i based on the key code KC, the volume VOL, the key-on signal KON and the variable i, and outputs it to the speaker 42 via the amplifier 41. Begin to. In this case, the pitch of the accompaniment sound signal is designated by the key code KC (when the variable i is 7 or 8, the type of percussion instrument sound is designated by the key code KC). The tone color of the accompaniment tone signal is set by the tone color number VOIN, and the volume of the accompaniment tone is specified by the volume VOL. As a result, a series of accompaniment sounds composed of chord constituent sounds, bass sounds, and percussion sounds are emitted from the speaker 42.

Again, returning to the "interrupt program" of FIG. 13, after the processing of the above steps 226 to 230,
At step 232, a “key-off routine” is executed, and at step 234, a “count routine” is executed. As shown in detail in FIG. 16, the "key-off routine" includes steps 290 to 300. In steps 292 to 296, the key-off time KOFT (i) matching the time TIME while changing the variable i from 0 to 8 As a result, at step 298, the variable i relating to the key-off time KOFT (i) and the key-off signal are output to the tone signal forming circuit 40. Since the musical sound signal forming circuit 40 ends the formation of the accompaniment sound signal of the track represented by the variable i, the sounding of the accompaniment sound corresponding to the signal from the speaker 42 ends.

The "count routine" is shown in detail in FIG. 17 and its execution begins at step 310,
By the processing of steps 312 and 318, the time TIME and the carreton timing CTIM are counted up by “1”, and when the time TIME reaches “5000” by the processing of steps 314 and 316, it is cleared to “0”,
The caleton timing CTIM is set to steps 320 and 322.
Is cleared to "0" when one bar time is reached by the processing of. As a result, every time the tempo clock signal generator 52 generates a tempo clock signal, that is, two quarter notes are generated.
At each of the four divided timings, the time TIME is sequentially increased by “1” from “0” to “4999”. Note that this “4999” has no meaning in the value itself, and may be any value as long as it is an extremely large value as compared with other variables representing time. The Caleton timing CTIM increases by "1" at each timing in each bar section. The current bar CBAR is processed by steps 320, 324 to 328, and every one bar time over one cycle of the pattern specified by the style number STLN and the pattern number PTRN.
From "0" to the number of measures "STLTBL (STLN) .BAR-1""1"
It increases in steps.

As described above, the start switch 2
By the operation of 4a, the automatic accompaniment starts to operate, and the “interrupt program” is executed each time the tempo clock signal generator 52 generates a tempo clock signal (at the timing of dividing a quarter note into 24). The performance data in the accompaniment data memory 60 specified by the style number STLN and the pattern number PTRN is repeatedly read, and the sounding of the accompaniment sound is controlled.

Next, the case where the switching selection flag CNGF is set to "1" by operating the selection switch 25 will be described. In this case, "YES" is determined in step 236 of the "interrupt program" in FIG. 13, and the "automatic conversion routine" is executed in step 238.

As shown in FIG. 18, this "automatic conversion routine" consists of steps 340 to 356. In this routine, only when the current timing CTIM is "0" (CTIM = 0), that is, one measure. For each time, based on the determination of “YES” in step 342, step 34
Each process of 4 to 354 is executed, and in other cases, the substantial process is not executed. In step 344, the section key touch amounts QRV (0), QRV (1), for the right keyboard 12
QRV (2) is the section key touch amount QRV (1), QRV (2) and right keyboard 1
The key touch amount for 2 is updated to RVSM,
The key touch amount RVSM is initialized to "0". Step 3
In 46, the number of section keys for the right keyboard 12 QRN
(0), QRN (1), QRN (2) are sequentially updated to the number of section keys QRN (1), QRN (2) and the number of key presses RNSM for the right keyboard 12, and the number of key presses RNSM Is initially set to "0". Step 3
In 48, the section key touch amount QL for the left keyboard 12
V (0), QLV (1), QLV (2) is the section key touch amount QLV (1), QLV (2)
And the key touch amount LVSM for the left keyboard 12 are sequentially updated, and the key touch amount LVSM is initialized to "0". At step 350, the number of section keys QLN (0), QLN (1), QLN (2) for the left keyboard 12 is changed to the number of section keys QLN (1), Q.
The number of pressed keys LNSM for LN (2) and the left keyboard 12 is sequentially updated, and the number of pressed keys LNSM is initialized to "0".
As a result, the left and right keyboards 11 for each of the past three sections for each bar,
Each section key touch amount for 12 QRV (0) ~ QRV (2), QLV (0)
~ QLV (2) and number of section keys pressed QRN (0) ~ QRN (2), QLN (0) ~ QL
N (2) will be calculated.

Next, at step 352, the "conversion determination routine" is executed. As shown in FIG. 19, this "conversion determination routine" consists of steps 360-380,
In the routine, first, in step 362, the up / down index UD is initialized to “0”, and then in step 36.
At 4, it is determined whether the section flag PERF is "2" or more. While the section flag PERF is less than "2",
When it is determined to be “NO” in step 364, step 3
At 66, "1" is added to the flag PERF, and step 3
The processes after 68 are not executed. This is the section flag PE
RF is initially set to "0" at the start of the automatic accompaniment operation, and until the lag PERF becomes "2", the section key touch amount and the section press of the past three sections for performing the switching evaluation of the accompaniment pattern are performed. This is because the number of keys has not been calculated. When three sections have passed, the section flag PERF is "2", and thereafter, in this "conversion determination routine",
Based on the determination of “YES” in step 364, the processes of steps 368 to 378 are executed.

The process of steps 368 to 378 is a process of calculating the up / down index UD according to the playing state of the left and right keyboards 11 and 12, and the judgment key range flag RNG indicates that the accompaniment pattern depends on the playing state of the right keyboard 12 only. If it is "1" indicating switching, "NO" in step 368.
Then, based on the determination of “YES” in step 376, only the processes of steps 370 to 374 are executed.

In step 370, the "first operation routine" shown in FIG. 20 is executed. In the "first operation routine", its execution is started in step 390, and the right keyboard 12 is executed by the determination processing in step 392.
If none of the number of key presses QRN (0) to QRN (2) in the past three sections for “0”, that is, if there is a key press on the right keyboard 12 in any of the past three sections, Step 39
It is determined to be “YES” in Step 2, and steps 394 to 398.
In step 400, 402, an index X (0) based on the difference in the number of pressed keys in two consecutive sections is calculated, and in step 400, 402, an index X (1) based on the difference in the average key touch amount in two consecutive sections is calculated. It is calculated. Further, if any one of the number of section keys QRN (0) to QRN (2) of the past three sections for the right keyboard 12 is “0”, that is, the right keyboard 12 is pressed in any of the past three sections. If there is no key, it is determined to be "NO" in step 392, and both the indexes X (0) and X (1) are set to "0" in step 404.

The exponent X (0) is calculated based on the following expression 3 in step 396 based on the judgment of “YES” in step 394 when one of the two logical operations based on the following expression 2 is established. Calculated. Also, 2 based on the following equation 2
If both logical operations are not satisfied, step 394.
Is determined to be "NO" at step 398 and the index X (0)
Is set to "0".

[Equation 2] QRN (2) ≧ QRN (1)> QRN (0) QRN (2) ≦ QRN (1) <QRN (0)

[Equation 3] X (0) = CF x {QRN (2) -QRN (1)} / {K (0) x RDNT (SENS)} As a result of these operations, the number of pressed keys in each section over the past three sections QR
If N (0), QRN (1), QRN (2) increase sequentially, the index X (0)
Is a positive value that represents the difference between the number of times the key was pressed two times before and the number of times the key was pressed QRN (1), QRN (2). Number of keys pressed in each section over the past three sections QRN
When (0), QRN (1), QRN (2) decrease sequentially, the index X (0)
Is a negative value that represents the difference between the number of key presses QRN (1) and QRN (2) two times before and the previous time. In this case, the switching evaluation coefficient CF is set to "0.5" if the accompaniment pattern was previously switched, and is set to "1" if the accompaniment pattern was not previously switched. If was previously switched, the absolute value of the index X (0) will be small. Further, the tone color coefficient K (0) is set to "0.8" when the melody tone color is a strings tone color (see FIG. 6 (B)), and is set to "0" when the melody tone color is other than that. Since it is set to "1",
When the melody tone is a strings tone, the index
The absolute value of X (0) becomes large. On the other hand, if the number of keys pressed in each section QRN (0), QRN (1), QRN (2) over the past three sections does not decrease or increase sequentially, the index X (0) is "0". Is set to.

The index X (1) is calculated in steps 400 and 402.
Is calculated based on the following equation 4.

[Formula 4] AVL1 = {QRV (0) + QRV (1)} / {QRN (0) + QRN (1)} AVL2 = {QRV (1) + QRV (2)} / {QRN (1) + QRN (2)} X (1) = CF x (AVL2-AVL1) / {K (1) x RDVL (SENS)} As a result of these calculations, the section key touch amount QRV (0), QRV (1), QR
If V (2) is on the increase, the index X (1) represents the difference between the average key touch amount AVL1 over the last and the previous-previous section and the average key touch amount AVL2 over the previous-and-previous section. It will be a positive value. If the section key touch amount QRV (0), QRV (1), QRV (2) is decreasing, the index X (1) is the average key touch amount AVL1, AVL2
It is a negative value that represents the difference between. Also in this case, the switching evaluation coefficient CF and the tone color coefficient K (1) have the same influence on the absolute value of the index X (1).

In step 372 (FIG. 19), the "second operation routine" shown in FIG. 21 is executed. In the "second operation routine", its execution is started at 410, and if the previous section key press number QRN (2) for the right keyboard 12 is not "0" by the judgment processing of step 412, That is, if there is a key press on the right keyboard 12 in the previous section, it is determined as “YES” in the same step 412, and the average key touch amount in the previous section in steps 414 to 424.
The index X (2) is calculated on the basis of the AVL, and the indexes X (3) are calculated on the basis of the previous section key press number QRN (2) in steps 426 to 434. If the previous section key press number QRN (2) for the right keyboard 12 is "0", that is, if there is no key press on the right keyboard 12 in the previous section, "NO" is given in step 412. It is determined and both the indexes X (2) and X (3) are set to "0" in step 436.

In the calculation of the index X (2), the average key touch amount AVL in the previous section is calculated in step 414, and this average key touch amount AVL is calculated as the switching condition table C.
If the reference value RVUP (SENS) read from GCTBL (FIG. 6 (A)) is multiplied by the tone color coefficient K (2) value K (2) × RVUP (SENS) or more, then “YES” is determined in step 416. It is determined and the index X (2) is set to “1” in step 420. The average key touch amount AVL is the value K (3) × the reference value RVDW (SENS) read from the switching condition table CGCTBL multiplied by the tone color coefficient K (3).
If it is less than RVDW (SENS), in step 418, "YE
S ”, and the index X (2) is“ − ”in step 422.
It is set to 1 ”. If the average key touch amount AVL is between the above two values K (2) × RVUP (SENS) and K (3) × RVDW (SENS),
Both of the steps 416 and 418 are determined to be “NO”,
In step 424, the index X (2) is set to "0". In this case, the tone color coefficients K (2) and K (3) are set to "0.8" and "1.2" when the tone color of the melody tone is a strings tone tone (see FIG. 6 (B)) and the melody tone. Since it is set to "1" in other tones, when the melody tone is a strings type tone, even if the average key touch amount AVL changes little, the index X (2 )
Tends to be positive (= 1) or negative (= -1).

In the calculation of the index X (3), the section key press number QRN (2) for the right keyboard 12 in the previous section is the reference value read from the switching condition table CGCTBL (FIG. 6 (A)). RNUP (SENS) multiplied by tone color coefficient K (4) K (4) × RNUP
If (SENS) or more, it is determined to be "YES" in step 426, and the index X (3) is set to "1" in step 430. The number of section keys QRN (2) is the switching condition table CG
If the reference value RNDW (SENS) read out from CTBL is multiplied by the tone color coefficient K (5) K (5) × RNDW (SENS) or less, step 42
It is determined to be “YES” in step 8, and the index is set in step 432.
X (4) is set to "-1". The number of section keys QRN (2)
Is between the above two values K (4) × RNUP (SENS) and K (5) × RNUP (SENS), both are determined to be “NO” in steps 426 and 428, and the index X ( 3) is set to "0". In this case as well, the degree of influence of the tone color coefficients K (4) and K (5) on the index X (3) is the same as in the above case. The index X (3) is positive (= 1) or negative (=-) even if the change in 2) is poor.
It becomes easy to become 1).

After executing the "first and second operation routines" (FIG. 19) in steps 370 and 372, step 37
In step 4, the calculated indexes X (0) to X (3) are added and set as the up / down index UD, and after the determination of “NO” in step 374, the “conversion determination routine” is executed in step 380. finish. Again, returning to the "automatic conversion routine" of FIG. 18, the "conversion routine" is executed at step 354 of the routine. This "conversion routine"
23, as shown in detail in FIG.
6 and the up / down index UD is evaluated in steps 462 and 464. This up / down indicator UD is "1"
If it is above, based on the evaluation that the key playing state is in the up state (exciting state), "YE" is determined in step 462.
It is determined to be "S", and the "up routine" is executed in step 466. If the up-down UD is "-1" or less, the key playing state is evaluated as being in the down state, and "YES" is determined in step 464, and the "down routine" is determined in step 468. To be executed. On the other hand, if the up / down index UD indicates a value between "-1" and "1", it is determined that both of the steps 462 and 464 are "NO" based on the evaluation that the key playing state has not changed. Is determined, the switching evaluation coefficient CF is set to "1" indicating that the pattern is not automatically switched in step 474 without executing the "up routine" and the "down routine".

As shown in FIG. 24, the "up routine" consists of steps 480 to 492. In this routine, if the current arrangement flag ARNG is "0",
When it is determined to be “NO” in step 482, step 4
At 84, the same flag ARNG is changed to "1". If the current arrangement flag ARNG is “1”, step 4
Based on the determination of “YES” in 82, only when the current pattern number PTRN is “0”, it is determined to be “YES” in step 486, and the pattern number PTRN is changed to “1” in step 488. And step 49
At 0, the arrangement flag ARNG is changed to "0".

After such "up routine" processing,
The program proceeds to the "pattern switching routine" of step 470 of FIG. 23, and in this routine, as described above, the automatic accompaniment pattern is switched according to the changed pattern number PTRN. In addition, after the processing of step 470, the switching evaluation coefficient CF indicates that the pattern is automatically switched in step 472 "0.
Set to 5 ".

As a result, when the up-down index UD is "1" or more, as shown in FIG. 7, the automatic accompaniment pattern currently playing is the first or second accompaniment pattern (PTRN = 0,1).
In the normal mode (ARNG = 0), the automatic accompaniment pattern is the arrangement mode of the first or second accompaniment pattern (A
RNG is changed to 1). If the automatic accompaniment pattern currently being played is the first accompaniment pattern (PTRN = 0) arrange mode (ARNG = 1), the automatic accompaniment pattern is the second accompaniment pattern (PTRN = 1) normal mode (ARNG).
= 0).

As shown in FIG. 25, the "down routine" comprises steps 500 to 512. In this routine, if the current arrangement flag ARNG is "1",
When it is determined to be “NO” in step 502, step 5
At 04, the same flag ARNG is changed to "0". If the current arrangement flag ARNG is “0”, step 5
Based on the determination of “YES” in 02, only when the current pattern number PTRN is “1”, it is determined to be “YES” in step 506, and the pattern number PTRN is changed to “0” in step 508. And step 51
At 0, the arrangement flag ARNG is changed to "1".

After such "down routine" processing,
The program proceeds to the "pattern switching routine" of step 470 of FIG. 23, and in this routine, as described above, the automatic accompaniment pattern is switched according to the changed pattern number PTRN. In addition, after the processing of step 470, the switching evaluation coefficient CF indicates that the pattern is automatically switched in step 472 "0.
Set to 5 ".

As a result, the up / down index UD is "-1".
In the following cases, as shown in FIG. 7, the automatic accompaniment pattern currently playing is the first or second accompaniment pattern (PTRN = 0,1).
) Arrange mode (ARNG = 1), the automatic accompaniment pattern is changed to the normal mode (ARNG = 0) of the first or second accompaniment pattern. If the automatic accompaniment pattern currently being played is the second accompaniment pattern (PTRN = 1) normal mode (ARNG = 0), the automatic accompaniment pattern is the first accompaniment pattern (PTRN = 0) arrange mode (ARNG). = 1).

As can be understood from the above description,
If the determination range flag RNG is "1" indicating that the accompaniment pattern is switched only by the playing state of the right keyboard 12,
During the automatic accompaniment operation, the key playing state amount of the right keyboard 12, that is, the number of keys pressed by the right keyboard 12 in a plurality of consecutive sections QRN
The automatic accompaniment pattern is automatically switched according to the difference between (0) to QRN (2) and the difference between each average key touch amount AVL1, AVL2, and the average key touch amount of the right keyboard 12 in the predetermined section.
The automatic accompaniment pattern is switched according to the AVL and the number of section keys QRN (2). And in this case, the number of keys pressed QRN (0)
~ QRN (2) difference, average key touch amount AVL1, AVL2 difference, average key touch amount AVL and section key press number QRN (2) corresponding reference value RDNT (SENS) , RDVL (SEN
S), RVUP (SENS), RVDW (SENS), RNUP (SENS), RNDW (SENS)
Can be switched in three steps by operating the changeover condition switch 26, so that depending on the performer's individuality or musical idea,
The performer can arbitrarily select the automatic accompaniment pattern switching condition. Further, in the evaluation of the switching, the tone color coefficient K (i) (table tone color coefficient TK is set to different values depending on whether the tone color of the melody tone selected by the tone color selection switch group 21 is scrings type or not.
(i)) and the switching evaluation coefficient CF, which is set to a different value depending on the presence / absence of the previous accompaniment pattern switching, the conditions for automatic switching of the accompaniment pattern are melody tone and Change control is performed according to the past pattern switching state.

Next, a case will be described in which the determination key range flag RNG is set to "0" indicating that the accompaniment pattern is switched only by the performance state of the left keyboard 11. In this case, "YES" is determined in step 368 of the "conversion determination routine" (FIG. 19), and only the "third calculation routine" in step 347 is executed.

As shown in FIG. 22, the execution of this "third calculation routine" is started at step 440, and the determination processing of step 442 determines the previous section key press count QLN for the left keyboard 11. If (2) is not "0", that is, if the left keyboard 11 is pressed in the previous section,
In the same step 442, it is determined as “YES”, and in steps 444 to 452, the up / down index UD is calculated based on the average key touch amount AVL of the previous section for the left keyboard 11. If the section key press count QLN (2) is “0”, that is, if there is no key press on the left keyboard 11 in the previous section, it is determined to be “NO” in step 442, and step 454 is performed. Since the execution of the lever routine is completed, the up / down index UD is set in the step 362 (see FIG. 1).
It is maintained at "0" by the initial setting of 9).

In the calculation of the up / down index UD, the average key touch amount AVL for the left keyboard 11 in the previous section is calculated in step 444, and this average key touch amount is calculated.
AVL value obtained by multiplying the reference value LVUP (SENS) read from the switching condition table CGCTBL (Fig. 6 (A)) by the tone color coefficient K (6).
If K (6) × LVUP (SENS) or more, it is determined to be “YES” in step 446, and the up / down index UD is set to “1” in step 450. Average key touch amount AVL
Is the reference value read from the switching condition table CGCTBL
Value obtained by multiplying LVDW (SENS) by tone color coefficient K (7) K (7) × LVDW (SENS)
If it is less than or equal to “YES” in step 448, the up / down index UD is “−” in step 452.
It is set to 1 ”. If the average key touch amount AVL is between the above two values K (6) × LVUP (SENS), K (7) × LVDW (SENS),
Both are determined as “NO” in steps 446 and 448,
In step 424, the up / down index UD is maintained at "0" as described above. In this case, the timbre coefficients K (6), K (7)
Are set to "0.8" and "1.2" when the melody tone is a strings tone (see Fig. 6 (B)).
And "1" when the melody tone color is other than that.
Since the melody tone is a strings tone, the up / down index UD is positive (= 1) or negative (= -1) even if the average key touch amount AVL changes little. ).

After the execution of this "third calculation routine" (step 378 in FIG. 19, FIG. 22), the "conversion routine" is executed in step 354 of the "automatic conversion routine" in FIG. In this "conversion routine" (Fig. 23),
As described above, if the calculated up-down index UD is “1” or more, the key playing state of the left keyboard 11 is evaluated to be in the up state (climbing state), and FIG.
The "up routine" is executed, and the automatic accompaniment pattern is switched in the direction of rising (see FIG. 7). If the up / down index UD is "-1" or less, the left keyboard 1
Based on the evaluation that the key playing state of 1 is the down state, the "down routine" of FIG. 25 is executed, and the automatic accompaniment pattern is switched in the opposite direction to the above (see FIG. 7).
If the up / down index UD is a value between “1” and “−1”, the “up routine” and the “up routine” are evaluated based on the evaluation that the key playing state of the left keyboard 11 has not changed. The "down routine" is not executed either, and the automatic accompaniment pattern is not changed.

As can be understood from the above description,
If the determination range flag RNG is “0” indicating that the accompaniment pattern is switched only by the performance state of the left keyboard 11,
During the automatic accompaniment operation, the automatic accompaniment pattern is switched according to the key playing state amount of the left keyboard 11, that is, the average key touch amount AVL of the left keyboard 11 in the predetermined section. Then, in this case, the reference values LVUP (SENS) and LVDW (SENS) corresponding to the average key touch amount AVL and used for the evaluation of the switching are
By operating the switching condition switch 26, the switching can be performed in three steps, so that the player can arbitrarily select the switching condition of the automatic accompaniment pattern according to the player's individuality or musical idea. Further, in the evaluation of the switching, the timbre coefficient K (i) (table timbre coefficient TK Since (i)) is taken into consideration, the condition for automatic switching of the accompaniment pattern is changed and controlled according to the tone color of the melody tone. In this case, the tone color coefficients K (6) and K (7) are respectively set to "0.8" and "1.2" when the melody tone color is a strings tone color (see FIG. 6 (B)) and the melody tone. Since it is set to "1" at other times than the tone of the tone, when the tone of the melody tone is a strings tone,
Even if the change in the average key touch amount AVL is small, the up / down index UD tends to be positive (= 1) or negative (= -1),
The pattern switching is facilitated.

Further, a case will be described in which the determination range flag RNG is set to "2", which indicates that the accompaniment pattern is switched depending on the performance state of both the left keyboard 11 and the right keyboard 12. In this case, the "conversion judgment routine"
Both are “NO” in steps 368 and 376 of (FIG. 19).
It is determined that the up / down index UD is calculated according to the playing state of the right keyboard 12 by the processing of the steps 370 to 374, and the up / down index UD is calculated according to the playing state of the left keyboard 11 by the processing of the step 378. UD is fixed. Then, by the processing of the "conversion routine" of FIG. 23, the switching of the accompaniment pattern is controlled according to the corrected updown index UD, so that the accompaniment pattern is changed according to the performance state of the left keyboard 11 and the right keyboard 12. Change controlled. In this way, the value of the judgment area flag RNG is “0”
A keyboard for controlling the switching of the accompaniment pattern is selected according to "2", and the determination flag RNG is selectively set by the operation of the determination area switch 27. Therefore, the performer intends to use the accompaniment pattern for the switching. You can select the desired keyboard.

In the above embodiment, the right keyboard 12
As the key playing state quantity of, the difference in the number of keys pressed in the two sections QRN (2) to QR
N (1) Detects the difference AVL2-AVL1 between the average key touch amounts in the two sections, the average key touch amount AVL in the predetermined section (QRV (2) / QRN (2)) and the number of key presses QRN (2) in the predetermined section Then, it was used to control the switching of the accompaniment pattern (see “conversion judgment routine” in FIG. 19), but from among these four key playing state quantities, one to three key playing state quantities were detected as necessary. It may be used for switching the accompaniment pattern. In the above embodiment, the average key touch amount AVL (QRV (2) / QRN (2))
And the number of keys pressed, QRN (2), the timbre coefficient that represents the timbre of the melody.
Reference values RVUP (SENS), RVDW (SENS), RNUP (SENS), RNDW (SEN) selected by K (2) to K (5) and switching condition switch 26
Although the indexes X (2) and X (3) are calculated by taking only S) into consideration (see “Second operation routine” in FIG. 21), the indexes X (0), X
As in the case of (1), the switching evaluation coefficient CF representing the switching state of the previous accompaniment pattern is set to the average key touch amount AVL.
(QRV (2) / QRN (2)) and the number of key presses QRN (2), the index X
(2) and X (3) may be calculated. Further, in the above embodiment, the difference in the number of pressed keys QRN (2) -QRN (1), the difference in the average key touch amount AVL2-AVL1, the average key touch amount AVL (= QRV
(2) / QRN (2)) and the number of key presses QRN (2), the tone color coefficient K (0)
~ K (5), switching evaluation coefficient CF and reference value RDNT (SENS), RDV
L (SENS), RVUP (SENS), RVDW (SENS), RNUP (SENS), RNDW (SEN
S) is added to calculate the indices X (0) to X (3) and the up / down index UD is calculated (“conversion judgment routine” in FIG. 19 and “first and second calculation in FIGS. 20 and 21”). Then, the up / down index UD is compared with the reference values “1” and “−1” to decide the switching of the accompaniment pattern (see “conversion routine” in FIG. 23). The reference values "1" and "-1" are coefficients corresponding to the timbre of the melody tone, a switching evaluation coefficient indicating the previous switching state of the accompaniment pattern, and a switching condition switch 2.
It may be changed according to the value selected in 6. Furthermore, both the indexes X (0) to X (3) and the up / down index UD side, and the reference values “1” and “−1” side are set to the coefficient corresponding to the timbre of the melody tone and the previous accompaniment pattern. It may be changed according to the switching evaluation coefficient indicating the switching state and the value selected by the switching condition switch 26.

In the above embodiment, the left keyboard 11
The average key touch amount AVL in a predetermined section as the key playing state amount of
Only (= QLV (2) / QLN (2)) was detected and used for accompaniment pattern switching control (“conversion judgment routine” in FIG. 19, “third calculation routine” in FIG. 22 and “conversion routine in FIG. 23”). "Conversion routine"), as the key playing state quantity of the left keyboard 11, as in the case of the right keyboard 12, the difference in the number of keys pressed in the two sections,
The difference between the average key touch amounts in the two sections and the number of keys pressed in the predetermined section may be detected and used for the accompaniment pattern switching control. Also in the case of the left keyboard 11, the detected difference in the number of key depressions, the difference in the average key touch amounts, the average key touch amount and the number of key depressions are calculated by the coefficient corresponding to the timbre of the melody sound or the accompaniment sound, It may be changed according to the switching evaluation coefficient indicating the switching state of the accompaniment pattern and the value selected by the switching condition switch 26. Furthermore, the reference value “1”, “−1” side with which the detected difference in the number of pressed keys, the difference in the average key touch amount, the average key touch amount and the number of pressed keys are compared,
Alternatively, for both the detection value side and the reference value side, a coefficient corresponding to the timbre of the melody tone, a switching evaluation coefficient indicating the switching state of the previous accompaniment pattern, and a switching condition switch 26.
It may be changed according to the value selected in.

Further, as the key playing state quantity of the left and right keyboards 11 and 12, the following quantity may be detected as the key playing state quantity. Performance frequency over a specified section of a chord type such as major or minor specified by the left keyboard 11 Number of key presses over a specified section of a black or white key on the right keyboard 12 Average across

In the above embodiment, the section for detecting the playing state quantity is one bar, but the length of this section is one.
It may be shorter or longer than the bar. For example, 2
For each beat, 2 bars may be set as the predetermined section. Moreover, you may determine the said area | region by absolute time. Further, the section may be variable according to the tempo of the automatic accompaniment set by the tempo volume 23. In this case, it is effective to lengthen the section if the tempo is fast and shorten the section if the tempo is slow.

In the above embodiment, the first and second accompaniment patterns (PTRN0,1) are used as the accompaniment pattern data for each accompaniment style in the normal mode (ARNG = 0) and the arrange mode (ARNG = 1). Divided into two types of pronunciation,
Although four types of accompaniment patterns are realized, four types of independent accompaniment patterns may be prepared for each accompaniment style. Further, the number of types is not limited to four, and may be different.

Further, in the above embodiment, the accompaniment data memory 60 stores the predetermined performance data, but the memory 60 is composed of the RAM so that the performer can write arbitrary data. You can
It is possible to write arbitrary data from an external recording medium such as a magnetic tape or a magnetic disk.

[Brief description of drawings]

FIG. 1 is an overall block diagram of an electronic musical instrument having an automatic accompaniment device according to an embodiment of the present invention.

FIG. 2 is a detailed view of an operation panel of FIG.

FIG. 3 is a data format diagram of a style table in the accompaniment data memory of FIG.

FIG. 4 is a data format diagram of a pattern table in the accompaniment data memory of FIG. 1;

FIG. 5 is a data format diagram of a performance data table in the accompaniment data memory of FIG. 1;

6A is a data format diagram of a switching condition table in the accompaniment data memory of FIG. 1, and FIG. 6B is a data format diagram of a tone color coefficient table in the memory.

FIG. 7 is an explanatory diagram for explaining a manner of changing an automatic accompaniment pattern.

FIG. 8 is a flowchart of a “main program” executed by the microcomputer unit of FIG. 1;

FIG. 9 is a detailed flowchart of a “key event routine” of FIG. 8;

10 is a detailed flowchart of the "switch event routine" of FIG.

FIG. 11 is a detailed flowchart of a “pattern activation routine” in FIG. 10;

FIG. 12 is a detailed flowchart of the “pattern switching routine” of FIGS.

FIG. 13 is a flowchart of an “interrupt program” executed by the microcomputer unit of FIG. 1;

FIG. 14 is a detailed flowchart of a “reproduction routine” shown in FIG.

FIG. 15 is a detailed flowchart of a “note routine” of FIG. 14;

FIG. 16 is a detailed flowchart of a “key-off routine” in FIG. 13;

FIG. 17 is a detailed flowchart of the “count routine” of FIG.

FIG. 18 is a detailed flowchart of an “automatic conversion routine” of FIG.

FIG. 19 is a detailed flowchart of the “conversion determination routine” of FIG. 18.

FIG. 20 is a detailed flowchart of the “first calculation routine” of FIG.

FIG. 21 is a detailed flowchart of the “second operation routine” of FIG.

FIG. 22 is a detailed flowchart of the “third calculation routine” in FIG.

23 is a detailed flowchart of the "conversion routine" of FIG.

FIG. 24 is a detailed flowchart of the “up routine” of FIG. 23.

FIG. 25 is a detailed flowchart of the “down routine” of FIG. 23.

[Explanation of symbols]

11 ... left keyboard, 12 ... right keyboard, 13 ... key detection circuit, 1
4 ... Key touch detection circuit, 20 ... Operation panel, 22 ... Accompaniment style switch group, 24a ... Start switch, 24
b ... Stop switch, 25 ... Selection switch, 26 ... Switching condition switch, 27 ... Judgment area switch, 40 ...
Tone signal forming circuit, 50 ... Microcomputer section, 6
0 ... Accompaniment data memory.

Claims (15)

(57) [Claims] 1. A play-like and a plurality of keys or Ranaru performance operation means, the difference between the number of operations of the key at different intervals in
Performance state amount detecting means for detecting as a state quantity, accompaniment data memory storing a plurality of types of accompaniment pattern data, reading means for reading out one of the plurality of types of accompaniment pattern data, and detection by the performance state amount detecting means Switching the accompaniment pattern data read by the reading means to a different accompaniment pattern data when the performance state quantity is compared with a predetermined reference value and the performance state quantity belongs to one range with the same reference value as a boundary. Means, a musical tone signal generating means for generating a musical tone signal in response to the operation of a plurality of keys of the performance operating means, and a musical accompaniment signal in accordance with the read accompaniment pattern data, and switching by the switching means. Switching condition selecting means for selecting a condition, and the switching condition selected by the switching condition selecting means. Electronic musical instrument characterized by comprising a changing means for changing at least one of the detected playing state amount or the reference value Flip. 2. A plurality of keys or Ranaru performance operation means, the difference between the average key touch of the key at different intervals in
The performance state amount detecting means for detecting the performance state amount, the accompaniment data memory storing a plurality of types of accompaniment pattern data, the reading means for reading out one of the plurality of types of accompaniment pattern data, and the performance state amount detecting means. The detected performance state quantity is compared with a predetermined reference value, and when the performance state quantity belongs to one of the ranges with the same reference value as a boundary, the accompaniment pattern data read by the reading means is switched to different accompaniment pattern data. The switching means, the musical tone signal generating means for generating a musical tone signal in response to the operation of a plurality of keys of the performance operating means and the accompaniment tone signal in accordance with the read accompaniment pattern data, and the switching means. Switching condition selecting means for selecting a switching condition, and switching selected by the switching condition selecting means Electronic musical instrument characterized by comprising a changing means for changing at least one of the detected playing state amount or the reference value in accordance with the matter. 3. A performance operating means composed of a plurality of performance operators, a performance state quantity detecting means for detecting a performance state quantity by the performance operating means, an accompaniment data memory storing a plurality of types of accompaniment pattern data, and A reading means for reading out one of a plurality of types of accompaniment pattern data is compared with the performance state quantity detected by the performance state quantity detecting means and a predetermined reference value, and the performance state quantity is bordered by the same reference value. Switching means for switching the accompaniment pattern data read by the reading means to different accompaniment pattern data when belonging to one of the ranges, and generating and outputting a musical tone signal in response to operation of a plurality of performance operators of the performance operating means. And a musical tone signal generating means for generating an accompaniment tone signal according to the generated accompaniment pattern data. A tone color selecting means for selecting a tone color of a tone signal generated in response to an operation, and a changing means for changing at least one of the detected playing state amount and the reference value according to the tone color selected by the tone color selecting means. An electronic musical instrument characterized by having and. 4. The plurality of performance operators are composed of a plurality of keys, and the performance state quantity detecting means is a means for detecting the number of times the key is operated in a predetermined section as the performance state quantity. The electronic musical instrument according to claim 3 . 5. The plurality of performance operators are composed of a plurality of keys, and the performance state amount detecting means is a means for detecting an average key touch amount of the keys in a predetermined section as the performance state amount. The electronic musical instrument according to claim 3 . 6. The plurality of performance operators are composed of a plurality of keys, and the performance state quantity detecting means is a means for detecting a difference in the number of times each key is operated in different sections as the performance state quantity. The electronic musical instrument according to claim 3, wherein 7. The means for configuring the plurality of performance operators with a plurality of keys, and means for detecting a difference between average key touch amounts of the keys in different sections with the performance state quantity detecting means as the performance state quantity. The electronic musical instrument according to claim 3, wherein the electronic musical instrument comprises: 8. A performance operating means comprising a plurality of performance operators, a performance state amount detecting means for detecting a performance state amount by the performance operating means for each predetermined section, and an accompaniment storing a plurality of types of accompaniment pattern data. A data memory and a reading means for reading one of the plurality of types of accompaniment pattern data,
Accompaniment pattern data read by the reading means when the playing state quantity detected by the playing state quantity detecting means is compared with a predetermined reference value and the playing state quantity belongs to one range with the same reference value as a boundary. Switching means for switching between different accompaniment pattern data, and a musical sound signal for generating an accompaniment sound signal in response to the operation of a plurality of performance operators of the performance operation means and an accompaniment sound signal in accordance with the read accompaniment pattern data. In an electronic musical instrument provided with signal generating means, a switching evaluation value indicating whether or not the switching means has switched the read accompaniment pattern data based on the performance state quantity of the previous section detected by the performance state quantity detecting means. Switching evaluation value storage means for storing, and the detected performance state according to the stored switching evaluation value. Or electronic musical instrument, characterized in that a changing means for changing at least one of the reference values. 9. The apparatus according to claim 1, wherein the plurality of performance operators are composed of a plurality of keys, and the performance state amount detecting means is a means for detecting the number of times the key is operated in the section as the performance state amount. Item 8. The electronic musical instrument according to Item 8 . 10. The plurality of performance operators are composed of a plurality of keys, and the performance state amount detecting means is a means for detecting an average key touch amount of the keys in the section as the performance state amount. The electronic musical instrument according to claim 8 . 11. A means for configuring the plurality of performance operators with a plurality of keys, and using the performance state quantity detecting means as means for detecting a difference in the number of times each key is operated in the different sections as the performance state quantity. The electronic musical instrument according to claim 8, which is configured. 12. The plurality of performance operators are composed of a plurality of keys, and the performance state amount detecting means detects the difference between the average key touch amounts of the keys in different sections as the performance state amount. The electronic musical instrument according to claim 8, which is configured by means. 13. A first performance operating means comprising a plurality of performance operators, a second performance operating means comprising a plurality of performance operators, an accompaniment data memory storing a plurality of types of accompaniment pattern data, and the plurality of types. Reading means for reading one of the accompaniment pattern data, converting means for converting the read accompaniment pattern data according to the operation of a plurality of performance operators of the second performance operating means, and the first performance operating means. An electronic musical instrument comprising: a musical tone signal generating means for generating a musical tone signal in response to an operation of a plurality of musical performance operators and for generating an accompaniment tone signal in accordance with the converted accompaniment pattern data. To detect the playing state quantity by
A performance state quantity detecting means, and a second means for detecting a performance state quantity by the second performance operating means
Playing state quantity detecting means, selecting means for selecting one of the playing state quantities detected by the first or second playing state quantity detecting means, the selected playing state quantity and a predetermined reference value And a switching means for switching the accompaniment pattern data read by the reading means to different accompaniment pattern data when the same playing state amount belongs to one range with the same reference value as a boundary. . 14. A first performance operating means comprising a plurality of performance operators, a second performance operating means comprising a plurality of performance operators, an accompaniment data memory storing a plurality of types of accompaniment pattern data, and the plurality of types. Reading means for reading one of the accompaniment pattern data, converting means for converting the read accompaniment pattern data according to the operation of a plurality of performance operators of the second performance operating means, and the first performance operating means. An electronic musical instrument comprising: a musical tone signal generating means for generating musical tone signals in response to operations of a plurality of musical performance operators and for generating musical accompaniment tone signals in accordance with the converted accompaniment pattern data. To detect the playing state quantity by
A performance state quantity detecting means, and a second means for detecting a performance state quantity by the second performance operating means
Performance state quantity detecting means and at least one of the performance state quantities detected by the first or second performance state quantity detecting means, or the performance detected by the first performance state quantity detecting means. The selection means for selecting one of the playing state quantities in which the playing state quantity detected by the second playing state quantity detecting means is added to the state quantity, and the selected playing state quantity is compared with a predetermined reference value. Then, the electronic musical instrument is provided with switching means for switching the accompaniment pattern data read by the reading means to different accompaniment pattern data when the same performance state amount belongs to one of the ranges with the same reference value as a boundary. 15. A plurality of performance operators of the first performance operating means are composed of a plurality of keys for playing a melody, and a plurality of performance operators of the second performance operating means play chords. 14. The method according to claim 13, wherein the plurality of keys are provided.
Or the electronic musical instrument according to 14 .