JP3334781B2 - Automatic accompaniment device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic accompaniment apparatus, and more particularly, to an automatic accompaniment apparatus that generates musical notes having a plurality of pitches specified by a user and generates them as arpeggio sounds.

[0002]

2. Description of the Related Art Conventionally, an electronic musical instrument capable of executing an automatic arpeggio performance has been known (for example, Japanese Patent Publication No. Sho 61-7).
No. 638). Automatic arpeggio performance means that when a user designates a plurality of pitches by pressing a key or the like, the tones of those pitches (or tones of pitches based on those pitches) are converted into one according to a predetermined arpeggio pattern.
The sound is dispersed and generated automatically.

FIG. 8 shows an example of a key press and an arpeggio pattern. Reference numeral 801 denotes an example of key depression, in which pitches of fa, ra, and do are designated. 802 to 807 are key press examples 8
An example of an arpeggio pattern when a key is pressed like 01 is shown. The key press example 8 of the arpeggio pattern UP of 802
This is an arpeggio pattern running up to a designated pitch such as 01. Similarly, UPOC2 803 is a pattern in which the specified pitch is run up one octave, 8
DOWN of 04 is a pattern that runs down the specified pitch, DOWNNOC of 805 is a pattern that runs down the specified pitch from one octave above, and UPDO of 806
WN1 is a pattern in which the specified pitch is run up and down (the highest pitch is pronounced once), UPN of 807
OWN2 is a pattern in which the specified pitch is run up and down (the highest pitched note is sounded twice). The user selects one of the arpeggio patterns 802 to 807 in advance, and then presses a plurality of keys on the keyboard. Arpeggio sounds are formed and sounded based on these arpeggio patterns in accordance with key depression.

[0004]

By the way, when performing such an automatic arpeggio performance, the touch information of the keyboard is
Touch information when each key is pressed is directly reflected in the arpeggio sound corresponding to the key. For this reason, for a beginner, for example, when three different keys are pressed with three fingers to generate an arpeggio sound, the touches of the three fingers are different, and the touch information of each arpeggio sound is different. Therefore, there is a disadvantage that the intensities of the generated arpeggio sounds are not uniform. On the other hand, a method of generating an arpeggio sound at a predetermined volume irrespective of each key touch is also conceivable, but in that case, the arpeggio performance becomes monotonous.

According to the present invention, in an automatic accompaniment apparatus that generates an arpeggio sound in a predetermined arpeggio pattern based on designation of a plurality of pitches, it is possible to reflect touch information on each arpeggio sound in various modes. The purpose is to be able to.

[0006]

To achieve this object, an automatic accompaniment apparatus according to claim 1 comprises a plurality of pitch information and
Pitch information for inputting touch information corresponding to each pitch information
Input means, and a plurality of sounds input by the pitch information input means
In accordance with high information and in a predetermined arpeggio pattern
Arpeggios that form a dispersed arpeggio sound
The sound forming means and the touch of the arpeggio sound are
Of the average value of the plurality of pieces of touch information input by the information input means.
Touch control means to be a touch, and the touch control means
Using the touch information of each controlled arpeggio sound,
For each arpeggio sound formed by the arpeggio sound forming means
And control means for executing such control .

[0007] The automatic accompaniment apparatus according to claim 2, double
Number of pitch information and touch information corresponding to each pitch information
Inputting means for inputting pitch information,
A predetermined arpeggiation according to the plurality of pitch information
Arpeggio sounds distributed according to the pattern
Arpeggio sound forming means and touch of arpeggio sounds
Of the touch information input by the pitch information input means.
Touch control with the maximum or minimum value touch in the information
Means and each arpeggio controlled by the touch control means
Using the touch information of the sound, the arpeggio sound forming means
A control means for executing control for each formed arpeggio sound
And a step .

[0008] The automatic accompaniment device according to claim 3 is a device for accommodating a plurality of sounds.
Input high information and touch information corresponding to each pitch information
Pitch information input means and the pitch information input means
According to a plurality of pitch information and a predetermined arpeggio pattern
To form a distributed arpeggio sound
The arpeggio sound forming means and the touch of the arpeggio sound
The last of the plurality of touch information input by the pitch information input means.
Touch control means for making the last or first touch,
Performance information of each arpeggio sound controlled by the touch control means
Information on each of the arpeggio sounds formed by the arpeggio sound forming means.
Control means for performing control on the lupeggio sound.
It is characterized by the following.

According to a fourth aspect of the present invention, there is provided an automatic accompaniment apparatus for inputting a plurality of pieces of pitch information, a plurality of pieces of pitch information input by the plurality of pieces of pitch information, and Arpeggio sound forming means for forming a dispersed arpeggio sound in accordance with an arpeggio pattern; touch control means for setting a touch of an arpeggio sound to a set value corresponding to the number of inputs inputted by the pitch information input means; Control means for controlling the arpeggio sounds formed by the arpeggio sound forming means using the touch information of each arpeggio sound controlled by the control means. An automatic accompaniment apparatus according to claim 5 , wherein a plurality of pitch information and performance information corresponding to each piece of pitch information are input, an arpeggio pattern specifying means for specifying an arpeggio pattern, and an arpeggio sound. A mode designating unit for designating a mode that defines a mode of control to be performed on the plurality of pitch information input by the pitch information inputting unit, and an arpeggio specified by the arpeggio pattern designating unit. Arpeggio sound forming means for forming a dispersed arpeggio sound according to a pattern;
A performance that generates performance information of each arpeggio sound formed by the arpeggio sound forming means based on only the performance information input by the pitch information input means and in a mode corresponding to the mode specified by the mode specifying means. Information generating means, and control means for executing control on each arpeggio sound formed by the arpeggio sound forming means using the performance information of each arpeggio sound generated by the performance information generating means. . An automatic accompaniment device according to a sixth aspect is characterized in that, in the fifth aspect , the performance information is touch information. The automatic accompaniment device according to claim 7 is
Pitch information input means for inputting a plurality of pitch information and touch information corresponding to each pitch information, arpeggio pattern specifying means for specifying an arpeggio pattern, touch mode specifying means for specifying an arpeggio touch mode, Arpeggio sound forming means for forming a dispersed arpeggio sound in accordance with a plurality of pieces of pitch information input by pitch information input means and in accordance with an arpeggio pattern designated by the arpeggio pattern designating means; Touch information for generating touch information of each arpeggio sound formed by the arpeggio sound forming means based on the touch information input by the information input means and in a mode corresponding to the arpeggio touch mode specified by the touch mode specifying means. Generating means, and touch information of each arpeggio sound generated by the touch information generating means Used, it is characterized in that a control means for performing a touch control for each arpeggio notes formed in the arpeggio forming means.

[0010]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a system configuration diagram of an automatic accompaniment apparatus according to the present invention. The automatic accompaniment device includes a central processing unit (CPU) 101, a read only memory (ROM) 1
02, random access memory (RAM) 103, performance operator 105, setting / display unit 106, tone generator 10
7 and a system bus line 108.

The CPU 101 controls the entire automatic accompaniment device. Details of the processing will be described later with reference to FIGS. The ROM 102 stores a control program executed by the CPU 101, a preset tone color parameter group, and the like. The RAM 103 is used as a work area of the CPU 101 and various buffers. Also,
The RAM 103 has an area for storing a user-defined tone color parameter group. A backup battery 104 is connected to the RAM 103, and the contents stored in the RAM 103 are retained even when the power of the automatic accompaniment device is turned off.

The performance operator 105 is a keyboard having a plurality of keys for the user to perform a performance operation. Setting / display unit 1
Reference numeral 06 denotes a liquid crystal display (LCD) for displaying various information provided on the panel and a panel switch group. The tone generator 107 generates and outputs a tone based on an instruction from the CPU 101. The CPU 101 detects a performance operation of the performance operator 105 or outputs an external MIDI signal.
(Musical Instrument Digital Interface) A MIDI signal from a device is input, and a musical tone generator 1
07 is instructed to generate a musical tone.

FIG. 2 shows the appearance of the setting / display unit 106 shown in FIG. The setting / display unit 106 includes an LCD 209 for displaying various information, and panel switches 201 to 208. Arpeggio on / off switch 201, arpeggio hold switch 20
2, an arpeggio type switch 203, an arpeggio tempo switch 204, an arpeggio subdivide switch 205, tone switches 206-1 to 206-n, a data entry operator 207, and an arpeggio touch mode switch 208.

By pressing the arpeggio on / off switch 201, the arpeggio mode can be switched on and off. When the arpeggio mode is on, a key press event of the performance operator (keyboard) 105 or MID
When there is an event such as a note-on / off event from the I device, an arpeggio sound is generated in accordance with the event (a musical tone at a pitch for which a sound is instructed is dispersed in a predetermined arpeggio pattern and generated). When the arpeggio mode is off, the sound is generated according to the sounding instruction timing. In the following, the term “key press or key release” (without “keyboard 105”) includes not only the key press and key release of the keyboard 105 but also an event input from an external MIDI device.

The arpeggio hold switch 202 is a switch for designating whether or not to use the hold function when performing an arpeggio performance. When the arpeggio hold switch 202 is pressed simultaneously when the arpeggio mode is turned on by the arpeggio on / off switch 201, the hold function is turned on. When the arpeggio mode is turned on by pressing only the arpeggio on / off switch 201 without simultaneously pressing the arpeggio hold switch 202, the hold function is turned off and the arpeggio mode is turned on. When the arpeggio hold switch 202 is pressed while the arpeggio mode is on with the hold function on, the hold function is turned off (the arpeggio mode on remains unchanged). When the hold function is on,
When the user once depresses the keyboard to start an arpeggio performance and then releases all the keys that have been depressed, the arpeggio performance is continued at the currently released pitch. Next, when there is a new key press, the continued arpeggio performance ends, and the arpeggio performance according to the new key press starts. When the hold function is off, the arpeggio performance ends when the user once depresses the keyboard to start an arpeggio performance and then releases all of the keys pressed. Regardless of whether the hold function is on or off, after the user has pressed the keyboard once and started playing the arpeggio,
When a part of the depressed keys is released, an arpeggio performance corresponding to the key being depressed at that time is performed.

An arpeggio type switch 203, an arpeggio tempo switch 204, and an arpeggio subdivide switch 205 are used to change the type of arpeggio performance (ie, the arpeggio pattern as shown in FIG. 8), the tempo, and the subdivide. Switch to perform. The sub-divide means designation of a shuffle or triplet. When the arpeggio type switch 203 is pressed, the mode changes to the arpeggio type editing mode, the current arpeggio type setting is displayed on the LCD 209, and the set value can be changed by operating the data entry operator 207. Pressing the arpeggio type switch 203 again cancels the arpeggio type editing mode. The same applies to the arpeggio tempo switch 204 and the arpeggio sub-divide switch 205. When these switches are depressed, the mode changes to the arpeggio tempo edit mode and the arpeggio sub-divide edit mode, respectively. can do. Again these switches 204,2
When 05 is pressed, each editing mode is canceled.

The tone switches 206-1 to 206-n are:
This is a switch for selecting a tone. Tone switch 2
When any one of the buttons 06-1 to 206-n is pressed, the corresponding timbre is set.

Arpeggio touch mode switch 208
Is a switch for selecting a touch mode that defines in what form touch control is performed when performing touch control of an arpeggio sound described later with reference to FIG. When the arpeggio touch mode switch 208 is pressed, the mode changes to the arpeggio touch mode editing mode, the current setting of the arpeggio touch mode is displayed on the LCD 209, and the set value can be changed by operating the data entry operator 207. Pressing the arpeggio touch mode switch 208 again cancels the arpeggio touch mode editing mode.

FIG. 3 shows a memory map of the RAM 103. Reference numeral 301 in FIG. 3A indicates the entire memory map.
On the RAM 103, a CPU working area 311,
Tone parameter buffer 312, ARPBUF 313,
And a tone color parameter group 314 are set. The timbre parameters stored in the timbre parameter buffer 312 are:
This is the currently set tone. Tone parameter group 31
4, n types of tone color parameters corresponding to the tone color switches 206-1 to 206-n in FIG. 2 are set. The timbre parameters defining one timbre include a timbre basic parameter group 321, an arpeggio type 322, an arpeggio tempo 323, and an arpeggio subdivide 324.

FIG. 3B shows the contents of the ARPBUF 313. This buffer is a buffer for storing key codes KC of a plurality of pitches at which a key is pressed and an arpeggio is performed, and touch information TOUCH corresponding to each key code. The key code is stored in the KC portion, and the corresponding touch information is stored. Here, it is assumed that up to m pieces can be stored. The actual value of m is appropriately determined according to the maximum pronunciation class and the processing capacity of the apparatus.

Next, the operation of the CPU 101 of FIG. 1 will be described with reference to the flowcharts of FIGS.

FIG. 4 shows a main program that operates after the power of the automatic accompaniment device is turned on. Step 4
At 01, various areas are initialized. Next, step 40
In step 403, an event is detected. The event detection processing in step 402
Processing such as detection of a key press or key release event of No. 05, detection of a note on / off event input from a MIDI device, and detection of an operation event of a panel switch group in FIG. The tone color selection process of step 403 is to read the corresponding tone color parameter from the tone color parameter group 314 of FIG.
This is a process of setting to 12. After step 403, arpeggio related setting processing (FIG. 5) is performed in step 404, sound generation processing (FIG. 6) is performed in step 405, and the process returns to step 402.

FIG. 5 shows the details of the arpeggio-related setting processing in step 404. In step 501, an arpeggio mode on / off setting according to an operation event of the arpeggio on / off switch 201 and an on / off setting of a hold function according to an operation event of the arpeggio hold switch 202 are performed. Next, in step 502, arpeggio related setting mode management is performed. When one of the arpeggio type switch 203, the arpeggio tempo switch 204, the arpeggio subdivide switch 205, and the arpeggio touch mode switch 208 is operated, the arpeggio type edit mode, the arpeggio tempo edit mode, and the arpeggio subdivide This is processing for setting or canceling the edit mode or the arpeggio touch mode edit mode.

Next, in step 504, it is determined whether or not the mode is the arpeggio type editing mode. If the mode is the arpeggio type editing mode, an arpeggio type setting process is performed in step 505, and the process returns. If the mode is not the arpeggio type edit mode, it is determined in step 506 whether or not the mode is the arpeggio tempo edit mode. If the mode is the arpeggio tempo editing mode, an arpeggio tempo setting process is performed in step 507, and the routine returns. If not in the arpeggio tempo edit mode, step 508
To determine whether the mode is the arpeggio subdivide editing mode. If the mode is the arpeggio subdivide editing mode, an arpeggio subdivide setting process is performed in step 509, and the routine returns. If the mode is not the arpeggio subdivide editing mode, it is determined in step 510 whether or not the mode is the arpeggio touch mode editing mode. If the mode is the arpeggio touch mode editing mode, an arpeggio touch mode setting process is performed in step 511, and the process returns. If the mode is not the arpeggio touch mode editing mode, the process returns as it is.

Steps 505, 507, 509, 5
The eleventh arpeggio type setting process, the arpeggio tempo setting process, the arpeggio subdivide setting process, and the arpeggio touch mode setting process respectively correspond to the currently set arpeggio type (arpeggio pattern),
This is a process of displaying an arpeggio tempo, an arpeggio sub-divide, and an arpeggio touch mode, and changing the current setting value according to the operation of the data entry operator 207.

FIG. 6 shows the details of the tone generation process in step 405 of FIG. In step 601, the key depression of the keyboard 105 (M
(Including key-on of IDI input), and
At 2, it is determined whether or not a key is pressed. If there is a key depression, in step 603, the pitch of each key being depressed and the touch of each key are stored in the ARPBUF 313. At this time, if the ARPBUF 313 retains the data of the arpeggio performance that has been continued after all keys have been released with the hold function turned on, after deleting them, each pitch currently pressed and each The key touch is stored in the ARPBUF 313.

Next, at step 604, it is determined whether or not the arpeggio mode is on. When the arpeggio mode is on, in step 605, an arpeggio event is developed according to the currently selected tone color and arpeggio type (arpeggio pattern) based on the pitch information (KC) of the key depression sound stored in the ARPBUF 313. I do. The development of an arpeggio event is to create event information for each arpeggio sound generation in order to generate a dispersed arpeggio sound according to a specified arpeggio pattern. Next, in step 606, an arpeggio event touch control process is performed. This processing will be described in detail with reference to FIG. Next, in step 607, an arpeggio event is generated as a performance event in accordance with the specified arpeggio tempo and arpeggio subdivide. Then, at step 608, a performance event tone generation process is performed, and the routine returns. If the arpeggio mode has been turned off in step 604, the process directly proceeds to step 608. The performance event tone generation process of step 608 is a process of generating a tone by the tone generator 107 according to the generated performance event. After step 607, step 6
When the number reaches 08, a performance event of each arpeggio sound is generated, and the arpeggio sounds are dispersed and emitted. When the step directly reaches step 608 from step 604, the key press event becomes a performance event as it is. Sound is emitted directly from the keyboard in response to a key press.

If no key is depressed in step 602,
In step 609, it is determined whether or not all keys have been released (including MIDI input key-off). If all keys have been released, it is determined in step 610 whether or not the hold function has been turned on. When the hold function is ON, it means that the arpeggio performance should be continued based on the information currently stored in the ARPBUF 313. Therefore, the process proceeds to step 604 to continue the arpeggio performance.
When the hold function is turned off in step 610,
Alternatively, if all keys have not been released in step 609, it is determined in step 611 whether or not there has been a key release event (partial release of a key that has been pressed). If there is a key release, the entry of the ARPBUF 313 corresponding to the key code KC of the key released in step 612 is deleted, and the flow advances to step 604 to continue the arpeggio performance according to the current key depression state. . If there is no key release in step 611, the flow directly proceeds to step 604.

FIG. 7 shows details of the arpeggio event touch control processing in step 606 of FIG. In the arpeggio event touch control process, in step 701, touch information of each arpeggio sound is generated in a mode corresponding to the currently selected arpeggio touch mode. The generated touch information of each arpeggio sound is reflected as touch information of each arpeggio sound when a performance event of each arpeggio sound is generated in step 607 of FIG. Thus, touch control for each arpeggio sound is realized in a mode according to the currently selected arpeggio touch mode.

Specifically, the selection or combination of the modes shown in blocks 711 to 718 is performed in accordance with the currently selected arpeggio touch mode, and touch information of each arpeggio sound is generated. The mode of the block 711 is that the touch of each arpeggio sound is the corresponding key press touch of each key code (the key code corresponding to each arpeggio sound). Numeral 712 indicates that the touches of all the arpeggio sounds are the average value of the touches of each key press. Reference numeral 713 indicates that the touch of all arpeggio sounds is the maximum value (or the minimum value) of the touch of each key press. Reference numeral 714 denotes that the touch of all arpeggio sounds is the touch of the last (or first) key press. Reference numeral 715 is for setting the touch of all arpeggio sounds to a predetermined constant value. Numeral 716 indicates that the touch of each arpeggio sound is a value set according to the pitch (key code). Numeral 717 is for setting the touch of each arpeggio sound to a set value according to the arpeggio pattern. Numeral 718 indicates that the touches of all the arpeggio sounds are set values according to the number of key presses at that time.

For beginners, the touch of each key pressed during an arpeggio performance tends to vary. Therefore, it is preferable to select the touch mode so that the touch information is generated in a form like blocks 715 to 718 or 712.
Experts can control touch by fingering,
It is preferable to select the touch mode so as to have an aspect like blocks 711, 713, and 714.

In the embodiment of the invention described above, when the arpeggio mode is on, the entire key range is used for arpeggio performance. However, the key range is divided into a key range for arpeggio performance and a key range for normal performance ( (In this key range, even when the arpeggio mode is on, a key press is sounded as it is). Further, in the above-described embodiment of the present invention, the touch is basically an initial touch (velocity),
It is also possible to apply to after touch. further,
It may be applied to performance information other than touch information.

[0034]

As described above, according to the present invention, performance information (for example, touch information) is generated in different modes according to the mode selected by the user, and is reflected in each arpeggio sound. It is possible to reflect the performance information on the arpeggio sound in a manner according to the respective skill levels from to a skilled person. Therefore, it is possible to easily perform various arpeggio performances in a manner of controlling the performance information according to the user's preference.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an automatic accompaniment apparatus according to the present invention.

FIG. 2 is an external view of a setting / display unit.

FIG. 3 is a diagram showing a memory map of a RAM;

FIG. 4 is a flowchart of a main program.

FIG. 5 is a flowchart of an arpeggio-related setting processing routine;

FIG. 6 is a flowchart of a sound generation processing routine.

FIG. 7 is a flowchart of an arpeggio event touch control processing routine;

FIG. 8 is a diagram illustrating an example of an arpeggio pattern corresponding to a key press;

[Explanation of symbols]

101: Central processing unit (CPU), 102: Read-on memory (ROM), 103: Random access memory (RAM), 104: Backup battery, 105: Performance operator, 106: Setting / display unit, 107: Tone generator ,
108 bus line, 201 arpeggio on / off switch, 202 arpeggio hold switch, 208
... Arpeggio touch mode switch, 209 ... Liquid crystal display (LCD).

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-7-49689 (JP, A) JP-A-1-4795 (JP, A) JP-A-7-49689 (JP, A) 29272 (JP, B1) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10H 1/00-7/00 541

Claims (7)

(57) [Claims] 1. Pitch information input means for inputting a plurality of pitch information and touch information corresponding to each piece of pitch information, and in accordance with the plurality of pitch information input by the pitch information input means, and Arpeggio sound forming means for forming a dispersed arpeggio sound according to a predetermined arpeggio pattern; and a touch for setting a touch of the arpeggio sound to a touch of an average value of the plurality of pieces of touch information input by the pitch information input means. Control means for controlling each arpeggio sound formed by the arpeggio sound forming means using the touch information of each arpeggio sound controlled by the touch control means; Accompaniment device. 2. Pitch information input means for inputting a plurality of pitch information and touch information corresponding to each of the pitch information, and in accordance with the plurality of pitch information input by the pitch information input means, and Arpeggio sound forming means for forming a dispersed arpeggio sound in accordance with a predetermined arpeggio pattern; and a maximum or a minimum value of the plurality of pieces of touch information input by the pitch information input means. Touch control means to be a touch, and control means for executing control on each arpeggio sound formed by the arpeggio sound forming means using the touch information of each arpeggio sound controlled by the touch control means. A featured automatic accompaniment device. 3. Pitch information input means for inputting a plurality of pitch information and touch information corresponding to each piece of pitch information, and in accordance with the plurality of pitch information input by said pitch information input means, and Arpeggio sound forming means for forming a dispersed arpeggio sound according to a predetermined arpeggio pattern, and a touch of the arpeggio sound as a last or first touch of the plurality of pieces of touch information inputted by the pitch information input means. Touch control means, and control means for executing control on each arpeggio sound formed by the arpeggio sound forming means, using the performance information of each arpeggio sound controlled by the touch control means. Automatic accompaniment device. 4. A plurality of pitch information input means for inputting a plurality of pitch information, and distributed in accordance with a plurality of pitch information input by said pitch information input means and according to a predetermined arpeggio pattern. Arpeggio sound forming means for forming an arpeggio sound, touch control means for setting a touch of the arpeggio sound to a set value corresponding to the number of inputs inputted by the pitch information input means, and each arpeggio controlled by the touch control means An automatic accompaniment apparatus comprising: control means for executing control on each arpeggio sound formed by the arpeggio sound forming means using sound touch information. 5. A pitch information inputting means for inputting a plurality of pitch information and performance information corresponding to each pitch information, an arpeggio pattern specifying means for specifying an arpeggio pattern, and an arpeggio sound. A mode designating unit for designating a mode that defines a mode of control, and a distribution according to a plurality of pitch information input by the pitch information input unit, and according to the arpeggio pattern specified by the arpeggio pattern designating unit. Arpeggio sound forming means for forming the selected arpeggio sound, and the arpeggio sound forming means based on only the performance information input by the pitch information input means and in a mode corresponding to the mode specified by the mode specifying means. Performance information generating means for generating performance information of each arpeggio sound formed by the arpeggio generated by the performance information generating means; Using performance information of the sound, the automatic accompaniment apparatus characterized by comprising a control means for performing control for each arpeggio notes formed in arpeggio forming means. Wherein said performance information is a touch information claim 5
4. The automatic accompaniment device according to 2. 7. A pitch information input means for inputting a plurality of pitch information and touch information corresponding to each pitch information, an arpeggio pattern specifying means for specifying an arpeggio pattern, and a touch mode for specifying an arpeggio touch mode. Specifying means, and arpeggio sound forming means for forming a dispersed arpeggio sound according to a plurality of pitch information input by the pitch information input means and according to the arpeggio pattern specified by the arpeggio pattern specifying means And touch information of each arpeggio sound formed by the arpeggio sound forming means based on the touch information input by the pitch information input means and in a manner corresponding to the arpeggio touch mode specified by the touch mode specifying means. Touch information generating means for generating the arpeggios generated by the touch information generating means By using the touch information of the sound, the automatic accompaniment apparatus characterized by comprising a control means for performing a touch control for each arpeggio notes formed in the arpeggio forming means.