JPH11194774A - Automatic playing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a natural accompaniment sound which is free of a feeling of musical physical disorder even when arpeggios are played by holding the sounds of pressed keys as chord decision object sounds in order when the keys are pressed while an operation element for indicating chord decision making is pressed. SOLUTION: A panel switch group 1 generates a switch signal corresponding to each switch operation. A pedal operation element 2 generates an ON signal when stepped on and an OFF signal when released. A keyboard 3 generates playing information such as a key-ON/key-OFF signal corresponding to the key pressing/releasing operation. When keys are pressed while the pedal operation element 2 is turned on, this device holds the sounds of the pressed keys as chord decision object sounds. Further, the chord decision object sounds held corresponding to the ON operation of the pedal operation element 2 are automatically cleared, measure by measure. Or the sounds of the keys are excluded from the chord decision objects according to the envelope state of a key pressed after the pedal operation element 2 is turned on.

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 which enables an automatic accompaniment corresponding to an arpeggio playing method in which chords are played one by one.

[0002]

2. Description of the Related Art Conventionally, in order to assist a chord performance or a bass performance, a root (root note) and a chord type (chord type) of a played chord (chord) are detected, and an accompaniment pattern is determined based on the detected result. There is known an automatic accompaniment apparatus which converts the pitches of each of the sounds constituting the sound, and automatically generates a chord sound and a base sound obtained at a predetermined timing and accompaniment. The accompaniment pattern mentioned here is
It is composed of data representing the pitch, tone generation timing, etc. of each note to be played. For example, a chord backing pattern representing a middle-range accompaniment, or a bass pattern representing an accompaniment of a bass tone in a low-frequency range is used. Often.

[0003]

By the way, in the conventional automatic accompaniment apparatus described above, note names C ₃ , E ₃ , G ₃ (de, mi, so ) Can be easily discriminated as "C major" when the keys are pressed simultaneously, but when the key is pressed as shown in the first bar of FIG. . Hereinafter, this point will be described in detail.

As shown in FIG. 1, there are two methods of chord discrimination: 1) a method in which, when one key is pressed, the key is uniquely regarded as a major of the sound; Conventionally, there is known a method considered as (root), or 3) a method of treating a single note key as no chord and not discriminating the chord.

However, when the key depression operation shown in the first bar of FIG. 25 is performed, based on the chord determination result by the above-mentioned method 2), for example, the pitches of the respective sounds constituting the accompaniment pattern shown in FIG. The chord sound and the bass sound shown in the musical score of FIG. 27 are provided, which results in musically unaccompanied accompaniment. Further, when the method of the above 1) is followed, the accompaniment is completely unacceptable. As described above, the conventional automatic accompaniment apparatus has a problem that a natural accompaniment sound cannot be obtained if an arpeggio playing method is performed in which a key-depressing operation of a chord component sound is performed one by one.

Accordingly, the present invention has been made in view of the above circumstances, and has as its object to provide an automatic accompaniment apparatus capable of obtaining a natural accompaniment sound that is musically uncomfortable even when an arpeggio playing method is performed. I have.

[0007]

In order to achieve the above object, according to the first aspect of the present invention, there is provided an operation device for instructing code determination, and a key is depressed while the operation device is turned on. A code holding unit for sequentially holding a key sound as a code determination target sound; a code erasing unit for deleting a code determination target sound held by the code holding unit in response to an OFF operation of the operation element; and the code holding unit. A chord is determined based on the held chord determination target sound, and the accompaniment sounds obtained by pitch-converting each of the sounds constituting the accompaniment pattern according to the determined chord are pronounced according to the timing specified by the accompaniment pattern. And accompaniment means.

According to the second aspect of the present invention, when the chord determination target sound corresponds to a single key press, the accompaniment means regards the chord determination target sound as a root tone. It is characterized by

According to the third aspect of the invention, the accompaniment means determines that the chord is uncertain as a result of the chord judgment based on the chord judgment target sound. It is characterized in that the target sound is produced as a chord sound according to the timing specified by the accompaniment pattern.

According to the present invention, there is provided a code which includes an operation element for instructing code judgment, and sequentially holds a sound of a key depressed while the operation element is turned on as a sound to be judged. Holding means; a code erasing means for automatically deleting the chord judgment target sound held in the chord holding means every one bar; a chord judgment based on the chord judgment target sound held in the chord holding means; And an accompaniment means for producing an accompaniment sound obtained by pitch-converting each sound constituting the accompaniment pattern according to the chord according to the timing specified by the accompaniment pattern.

According to the present invention, there is provided a cord which includes a manipulator for instructing a chord judgment, and sequentially holds a sound of a key depressed while the manipulator is turned on as a chord judgment target sound. The holding means and the envelope of the sound of the key depressed after the operation element is turned on are released, and when the envelope level reaches a predetermined value, the sound of the key is transferred to the chord holding means. Chord erasing means for erasing from the held chord judgment target sound, performing chord judgment based on the chord judgment target sound held in the chord holding means, and pitching each sound constituting the accompaniment pattern according to the determined chord. An accompaniment means for producing an accompaniment sound obtained by the conversion in accordance with a timing specified by the accompaniment pattern.

In the present invention, when a key is depressed while the operator for instructing chord judgment is turned on, the sound of the depressed key is sequentially held as a chord judgment target sound. Even if an arpeggio playing method is performed in which the key is pressed one note at a time, a natural accompaniment sound without musical discomfort can be obtained.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The automatic accompaniment apparatus according to the present invention includes, for example, various electronic musical instruments having a keyboard such as an electronic piano,
The present invention can be applied to a DTM device using a personal computer. Hereinafter, an automatic accompaniment device according to an embodiment of the present invention will be described as an example with reference to the drawings.

A. First Embodiment (1) Configuration FIG. 1 is a block diagram showing a configuration of a first embodiment according to the present invention. In FIG. 1, reference numeral 1 denotes a panel switch group including various switches disposed on the device panel, such as a start / stop switch for instructing start / stop of an automatic accompaniment, and a switch for selecting an accompaniment pattern. Generates a switch signal according to the operation. Reference numeral 2 denotes a pedal operator that generates an ON signal when depressed and generates an OFF signal when released. What the pedal operator 2 intends will be described later. Reference numeral 3 denotes a keyboard for generating performance information such as a key-on / key-off signal, a key number (or a key code), and a velocity in response to a key press / release operation.

Reference numeral 4 denotes a CPU for controlling each section of the apparatus, and its characteristic operation will be described later. Also, this CPU
Numeral 4 generates a tempo clock that defines the performance tempo during automatic accompaniment, and executes a timer interrupt routine (described later) in synchronization with the tempo clock. Reference numeral 5 denotes a ROM for storing various control programs loaded into the CPU 4 and waveform data of various timbres. Reference numeral 6 denotes a RAM used as a work area of the CPU 4, which temporarily stores various register / flag data.

Reference numeral 7 denotes a data ROM for storing various accompaniment patterns including a chord backing pattern CODE representing a middle-range accompaniment and a base pattern BASE representing an accompaniment of a bass sound in a low-frequency range. Reference numeral 8 denotes a sound source configured by a well-known waveform memory reading method. This sound source 8
Is composed of an accompaniment sound source section and a normal sound source section. In the normal sound source section, based on performance information output from the keyboard 3, the CPU 4
Under the instruction, the waveform data of the designated timbre is read from the ROM 3 to synthesize a musical tone, and this is output as a performance sound. On the other hand, under the direction of the CPU 4, the accompaniment tone generator converts the pitches of the sounds constituting the accompaniment pattern according to the played chord, and outputs the resulting chord sound and bass sound as accompaniment sounds.

(2) Operation Next, the operation of the first embodiment having the above configuration will be described with reference to FIGS. In the following, the "operation of the main routine" will be described first as a schematic operation, followed by the "operation of the timer interrupt processing routine" which is executed by interruption, and the various processing routines called from the main routine will be sequentially described. And

A. Operation of Main Routine (Schematic Operation) First, when the power is turned on in the present embodiment, the CPU 4
A predetermined control program is loaded from M5, the main routine shown in FIG. 2 is executed, and the process proceeds to step Sa1. At step Sa1, various registers and flags provided in the RAM 6 are reset, and initialization for instructing the sound source 8 to initialize various registers and flags is executed.

After the initialization is completed, the CPU 4
Proceeds to the next step Sa2, and executes a process corresponding to the operation of the panel switch group 1 arranged on the console, that is, a switch process for instructing start / stop of the automatic accompaniment. When the start of the automatic accompaniment is instructed by the switch processing, the interrupt mask is released and the execution of the timer interrupt processing routine described later is instructed. On the other hand, when the stop of the automatic accompaniment is instructed, the interrupt mask is applied. Thus, the execution of the timer interrupt processing routine is prohibited.

Next, in step Sa3, preparations for regarding a key which is depressed / released while the pedal operator 2 is being turned on as a performance code for performing automatic accompaniment ( A pedal processing routine for adjusting the register setting is executed. Then, in the next step Sa4, a keyboard process for scanning each key of the keyboard 3 and detecting the key number (or key code) of the changed key is executed. In the following step Sa5, a code detection processing routine for detecting a performance code based on the pressed key of the keys pressed and released while the pedal operator 2 is turned on is executed.

When the performance chord is detected in this way, the CP
U4 advances the process to step Sa6 to generate a chord sound and a bass sound to be automatically accompanied based on the detected performance chord, and to generate a key depression sound corresponding to a key depressed when the chord is indeterminate. Execute the data generation routine. The accompaniment data generation processing routine includes a chord sound data generation processing routine, a base sound data generation processing routine, and a key depression sound generation processing routine. Data R based on route and type
Code backing pattern CO stored in OM7
The DE and the base pattern BASE are each subjected to pitch conversion to generate a chord sound / bass sound. And step S
In a7, a sound generation process for generating the generated chord sound / bass sound at the designated tempo is executed, and thereafter, the above-described steps Sa2 to Sa7 are repeated until the apparatus power is turned off.

B. Operation of Timer Interrupt Processing Routine Next, the operation of the timer interrupt processing routine will be described with reference to FIG. As described above, when the automatic accompaniment is started in the switch processing routine, the timer interrupt prohibition is released accordingly, and the CPU 4 executes step Sb1 shown in FIG. The stored tempo clock accumulated value is incremented by one to advance. Then, in the next step Sb2, it is determined whether or not the accumulated tempo clock accumulated value has reached "49", that is, whether or not one bar has elapsed when the quarter note length is 12 clocks. Here, when one bar has elapsed, the judgment result is "YES", and the routine proceeds to the next step Sb3, where the accumulated value of the tempo clock stored in the register T is reset to zero and the present routine is completed. Before the elapse, the judgment result is “N
O ", and this routine is immediately completed.

C. Operation of Switch Processing Routine FIG.
When the switch processing routine shown in FIG.
Advances the process to step Sc1 shown in FIG.
It is determined whether or not a stop switch (not shown) has been turned on. Here, if the switch is not turned on, the determination result is “NO”, the process proceeds to step Sc9, and the process corresponding to the operation of the other switches provided in the panel switch group 1 is executed. To complete this routine.

On the other hand, when the start / stop switch is turned on, the result of the determination in step Sc1 is "YES", and the process proceeds to the next step Sc2.
The start flag STF stored in the register STF is inverted. Next, in step Sc3, it is determined whether or not the inverted start flag STF is “1”, that is, whether or not the automatic start of the accompaniment is instructed.

In the case of automatic accompaniment start If the flag STF is set to "1" and the start of automatic accompaniment is instructed, the judgment result is "YES" and C
PU4 proceeds to the next step Sc4, and sets
Is reset to zero, and in subsequent steps Sc5 and Sc6, the code read address C stored in the registers CAD and BAD, respectively.
The AD and the base read address BAD are set to “zero (head address value)”. The code read address CAD and the base read address BA
D is an address for reading out the chord backing pattern CODE and the base pattern BASE from the accompaniment pattern data stored in the data ROM 7.

When the tempo clock accumulated value and each read address are reset to zero in this way, the CPU 4 proceeds to step Sc7 to release the inhibition of the timer interrupt processing. As a result, the above-described timer interrupt processing routine is executed, and the accumulation of the tempo clock is started.
Then, after the above-described step Sc9 and other switch processes, the present routine is completed.

In the case of automatic accompaniment stop On the other hand, if the start flag STF is "0" and the stop of automatic accompaniment is instructed, the above-described step Sc is performed.
3 is "NO", and the CPU 4 determines in step S
The process proceeds to c8, an interrupt mask is applied, timer interrupt processing is prohibited, and after step Sc9, other switch processing is performed, and then this routine is completed.

D. Operation of Pedal Processing Routine When the processing proceeds to step Sa3 of the main routine described above with the automatic accompaniment started via the above-described switch processing routine, CPU 4 executes step Sd1 of the pedal processing routine shown in FIG. I do. In step Sd1, first, it is determined whether or not the operation event generated by the pedal operator 2 has changed. Here, when the pedal operation is not performed and there is no event change, this routine is completed without performing any processing. However, when the pedal operator 2 is released, the state changes from the on state to the off state, and the off event occurs, Proceeding to step Sd2, the pedal flag PF
Is set to “0”. On the other hand, when the pedal operator 2 is depressed and the state changes from the OFF state to the ON state and an ON event occurs, the process proceeds to step Sd3, and the pedal flag PF
Is set to “1”.

Next, when the process proceeds to step Sd4, the CPU
Reference numeral 4 sets "1" in the register N and clears the registers ROOT and TYPE. Here, register N
Is a register in which a value indicating a sounding channel of the accompaniment tone generator is set, and registers ROOT and TYPE are registers in which values indicating a root tone and a chord type of a performance chord obtained by chord discrimination described later are set.

Subsequently, in step Sd5, the register C
H (N), ONKAI (N), and ONTIME (N) are reset to zero. Note that this register CH
(N) is a register for holding data indicating whether or not sound can be generated for the sound channel specified by the register N;
When "1", pronunciation is instructed, and when "0", pronunciation inhibition is instructed. The register ONKAI (N) is
A register ONTIME (N) is a register for storing data indicating a tone pitch of a musical tone to be emitted for the tone generation channel specified by the register N.

Next, in step Sd6, the register N
Is incremented by one, and the next step Sd
At 7, it is determined whether or not the value of the incremented register N exceeds N1. That is, in the case of the present embodiment, the accompaniment tone generator section has N1 sound channels for convenience.
The chord sounds are assigned to the (N1-1) th sounding channels, and the base sound is assigned to the N1st sounding channel. Therefore, in step Sd7, it is determined whether or not each register to which a chord tone is assigned has been initialized. If the initialization has not been completed, the determination result is “NO”, the process returns to step Sd5, and steps Sd5 to Sd6 are repeated. If completed, the determination result in step Sd7 becomes “YES”, The process proceeds to the next step Sd8.

Next, in step Sd8, the initial value "1" is stored in the register M, and in the following step Sd9, an invalid value FF (hexadecimal notation) indicating an empty state is stored in the key buffer register NOTE (M). Then, in step Sd10, the register M is incremented and incremented. In the following step Sd11, it is determined whether the incremented value of the register M exceeds the maximum value, that is, the key buffer register NOTE (M )
It is determined whether or not has been completed, and if the setting has not been completed, the process returns to step Sd9. At the time when the setting has been completed, the determination result is "YES", and this routine is completed.

E. Operation of Keyboard Processing Routine Next, the operation of the keyboard processing routine will be described with reference to FIG. When this routine is executed via step Sa4 of the main routine described above, the CPU 4 advances the processing to step Se1 shown in FIG. 6, and the key number (key number)
Is reset to zero. In step Se2, a change in the key event is determined for the key of the key number stored in the register N. When a key-on event has occurred, the process proceeds to the next step Se3, where "1" is set in the register KEY (N), and then the process proceeds to step Se5. On the other hand, when a key-off event has occurred, the process proceeds to step Se4, where "0" is set in the register KEY (N), and then the process proceeds to step Se4.
The process proceeds to e5.

When no key event occurs,
The process proceeds to Step Se5, and the register N is scanned to perform key scanning.
Is incremented by one and the key number is incremented. Thereafter, the process proceeds to step Se6, where it is determined whether the stepped key number has exceeded the number of keys, that is, whether key scanning has been completed for all keys. If key scanning is in progress, the determination result is "NO", and the above-described processing from step Se2 is repeated. If key scanning is completed, the determination result is "YES", and this routine is completed.

F. Operation of Code Detection Processing Routine Next, the operation of the code detection processing routine will be described with reference to FIGS. When this routine is executed through step Sa4 of the main routine, first, C
The PU 4 advances the process to step Sf1 shown in FIG. 7, resets the register N to zero, and in the subsequent step Sf2, the register KEY (N) corresponding to the register N is set to “1”.
That is, it is determined whether or not the key of the key number corresponding to the value of the register N is in a key pressed state. Hereinafter, the processing in the case of the key pressed state and the processing in the case of the key released state will be described separately.

Processing when key is depressed In this case, the determination result is "YES", the flow advances to the next step Sf3, the register M is reset to zero, and in the next step Sf4, the key corresponding to the value of this register M is It is determined whether or not the value of the buffer register NOTE (M) is FF (hexadecimal notation), that is, whether or not the buffer register is empty. Here, when it is not empty, the determination result is "NO", and the process proceeds to the next step Sf5 to increment the value of the register M. Then, in a step Sf6, it is determined whether or not the value of the stepped register M exceeds the maximum number.

If the maximum number is not exceeded, the result of the determination in step Sf6 is "NO", and the process returns to step Sf4 to determine again whether the key buffer register NOTE (M) is empty. to decide. Here, if it is empty, the determination result is “YES”.
And the CPU 4 advances the process to the next step Sf7,
The key number of the key being pressed is set in this register NOTE (M). Thereafter, the process proceeds to step Sf8, in which the value of the register N is incremented by one and is incremented, and in a succeeding step Sf9, it is determined whether or not the incremented value of the register N exceeds the number of keys. When the value of the register N does not exceed the number of keys, the determination result is “NO”, and the above-described steps Sf2 to Sf2 are performed until all keys are determined.
Repeat f9. Thus, steps Sf2 to Sf9
Then, the key number of the key in the depressed state is extracted and stored in the key buffer register NOTE (M) in the empty state.

Processing in Key Release State On the other hand, in the key release state, the result of the determination in step Sf2 is "NO", and the CPU 4 advances the processing to step Sf10 shown in FIG. In step Sf10, it is determined whether or not the pedal flag PF is “0”, that is, whether or not the pedal operator 2 has been turned off. Here, if the pedal operator 2 has been turned on, the determination result is "NO", and the process returns to step Sf8 shown in FIG. 7, but if the pedal operator 2 has been turned off, the determination result is "YES". Becomes
The process proceeds to step Sf11.

In step Sf11, the register M is reset to zero. In the following step Sf12, the key buffer register NOTE (M) corresponding to the value of the register M is set.
It is determined whether or not the key number value N is stored in the first step. Here, in the key buffer register NOTE (M),
When the key number value N is not stored, the determination result is “NO”, and the process proceeds to the next step Sf13 to increase the value of the register M. Then, the process proceeds to step Sf14, and the process proceeds to step Sf14. It is determined whether or not the value of the register M exceeds the maximum number, that is, whether or not the storage of the key number value N has been completed for all the key buffer registers.

If the determination is in the middle, the result of the determination in step Sf14 is "NO", and the process returns to step Sf12 to return to the key buffer register NO.
It is determined whether or not the key number N is stored for TE (M). Here, if the key number value N is stored, the determination result is “YES”, and the CPU 4 proceeds to the next step S
The process proceeds to f15, and after setting an invalid value FF indicating an empty state in this register NOTE (M), the process returns to step Sf8 in FIG. Note that the above step Sf14
In step Sf8, the process returns to step Sf8 also when it is determined that the key number N is stored in all the key buffer registers. Thus, step Sf10
In steps Sf15 to Sf15, when the pedal operator 2 is turned off, the key buffer register NOTE (M) storing the key number of the key in the released state is searched out, and if found, the register NOTE (M) is searched. ) Is set to empty.

As described above, if the pedal operator 2 is in the ON state, the key numbers of the depressed keys are sequentially stored in the empty key buffer register NOTE (M). On the other hand, if the pedal operator 2 is in the off state, the key buffer register NOTE (M) storing the key number of the key in the released state is set to an empty state.
U4 proceeds with the process after step Sf16 in FIG. 7 to detect a code. In this code detection, first, in step Sf16
Through Sf18, the number of key presses is determined. Hereinafter, processing for each keystroke count will be described.

(A) When the number of key presses is "0" or the number of key presses is "2" When the number of key presses is "0", the result of determination in step Sf16 is "NO", and in this case, code detection is performed. Because you can not do
This routine is completed without performing any processing. In the case of a two-key press, it is possible to regard the chord as an abbreviated performance chord. The decision result in the step Sf18 becomes "NO", and this routine is completed without performing any processing.

(B) When the Number of Key Presses is "1" When the number of key presses is "1", the decision result in the step Sf17 is "NO", and the process proceeds to a step Sf19 shown in FIG. In step Sf19, it is determined whether or not the route flag ROOTF stored in the register ROOTF is “0”. Here, when the key is the first key depressed after the start of this routine, that is, in a situation where the chord judgment by the single key depressed is not performed, the route flag ROOTF is set to “0”. YES ", and it proceeds to the next step Sf20. In step Sf20, the key number pressed by one note is uniquely regarded as the "root note" and set in the register ROOT. Then, step Sf
At 21, the root flag ROOTF is set to “1” to indicate that the chord is determined by a single key press, and this routine is completed.

(C) When the number of keys pressed exceeds "2" When two or more keys are pressed, the result of determination in step Sf18 is "YES", and the CPU 4 advances the process to step Sf22 shown in FIG. , A code determination based on a well-known rotation scheme or memory scheme is performed. Next, in step Sf23, it is determined whether or not the code is determined as a result of the code determination. If the code is not determined, the determination result is "NO", and this routine is completed. When the code is determined, the process proceeds to the next step Sf24, where the root tone is set in the register ROOT and the code type is set in the register TYPE.

In step Sf25, the code flag CODEF stored in the register CODEF is set to "1" when the code is determined.
From 6 to Sf29, with the new code decision,
Reset the register CH (L) to zero. That is, in step Sf26, the initial value “1” is set in the register L, and in step Sf27, the register CH (L) corresponding to the value of the register L is reset to zero. Next, in step Sf28, the value of the register L is incremented. In the following step Sf29, the value of the incremented register L has exceeded N1, that is, the register CH (L) corresponding to the N1 tone generation channels. It is determined whether or not zero reset has been completed for all of. If it is in the middle, the determination result is “NO”, and step S
The process returns to f27, but when the zero reset is completed, the determination result is “YES”, and this routine is completed.

G. Operation of Accompaniment Data Generation Processing Routine When the chord detection processing routine is completed,
U4 executes the accompaniment data generation processing routine shown in FIG. 10 through step Sa6 of the main routine described above,
The process proceeds to step Sg1. In step Sg1,
It is determined whether or not the code flag CODEF is “1”, that is, whether or not the code is in a code determined state. If the chord is in the determined state, the result of this determination is "YES", the flow proceeds to the next step Sg2, and a chord tone data generation process for pitch-converting each tone forming the chord backing pattern CODE according to the determined chord. Execute a routine (described later).

Then, in the next step Sg3, a base sound data generation routine (to be described later) for converting the pitch of the base pattern BASE according to the determined chord is executed, and then this routine is completed. On the other hand, if the code is not fixed, the result of the determination in step Sg1 is “N
O ", and in this case, the key press sound data generation processing routine (described later) for generating the sound of the pressed key is executed, and then this routine is completed.

H. Operation of Chord Sound Generation Processing Routine Next, the operation of the chord sound data generation processing routine will be described with reference to FIG. When this routine is executed via step Sg2 of the above-described accompaniment data generation processing routine, the CPU 4 advances the processing to step Sh1, and among the accompaniment patterns stored in the data ROM 7, the code read address CAD is read from the code backing pattern CODE. And reads the timing data and stores it in the register TIMING. And step S
At h2, it is determined whether or not the tempo clock accumulated value stored in the register T matches the timing data stored in the register TIMING, that is, whether or not the timing is under the chord tone generation timing.

If the timing is not under the chord sound generation timing, the judgment result is "NO", and this routine is completed once. If the timing is under the chord sound generation timing, first, steps Sh3 to Sh6 are performed. To search for an empty pronunciation channel. That is, step S
At h3, the initial value "1" is set to the register N, and at the next step Sh4, it is determined whether or not the register CH (N) corresponding to the value of the register N is "0".

If the sounding channel indicated by the register CH (N) is sounding, the judgment result is "N
O ", the process proceeds to the next step Sh5, and the value of the register N is incremented. Next, in step Sh6, it is determined whether or not the value of the incremented register N has reached N1, that is, whether or not the checking of the sound channel to which the chord sound can be assigned is completed.

When the check is in progress, the judgment result is "NO", and the process returns to step Sh4 to check whether there is a free space. When the check is completed, the judgment result becomes "NO". YES "
The process returns to step Sh3 again, and the check about the presence or absence of a vacancy is performed again from the beginning. If an empty sounding channel is found in such a check process, the result of the determination in step Sh4 is "YES", and the process proceeds to step Sh7.

In step Sh7, the code read address CAD is incremented, and based on this, the pitch data is read from the code backing pattern CODE and stored in the register ONKAI (N). Next, at step Sh8, the pitch data stored in the register ONKAI (N) is converted according to the root note and the chord type stored in the registers ROOT and TYPE.

Next, at step Sh9, the code read address CAD is incremented, and based on this, the timing data (event interval) is read from the code backing pattern CODE and the register ONTI
Store in ME (N). Subsequently, in step Sh10, the tempo clock accumulated value T is added to the timing data (event interval) stored in the register ONTIME (N) to form a note length, which is then stored in the register ONT (N).
Store back to IME (N).

The pitch data converted in accordance with the root note and the chord type is stored in the register ONKAI (N).
And the note length (pronunciation time length) is stored in the register ONTIME.
When the chord sounds are prepared and stored in (N), the CPU 4 proceeds to step Sh11 and sets "1" in the register CH (N) to occupy the sound channel to which this chord sound is assigned. .

Thereafter, the process proceeds to step Sh12,
The code read address CAD is incremented and incremented, and in a succeeding step Sh13, it is determined whether or not the incremented code read address CAD has exceeded the pattern end END. If it does not exceed the pattern end END, the determination result is “NO”, and the processing from step Sh1 is repeated. On the other hand, the pattern end EN
D, that is, when all the patterns have been read, the determination result is “YES”, and step S
At h14, the code read address CAD is returned to the head address, and the above-described processing is repeated again.

As described above, in the chord tone data generation processing routine, each chord backing pattern CODE constituting the chord backing pattern and chord type (chord type) determined in the chord detection processing routine is determined. It is instructed to produce a chord tone obtained by converting the pitch of the tone at a tone generation timing specified by the pattern.

I. Operation of Base Sound Generation Processing Routine Next, the operation of the base sound data generation processing routine will be described with reference to FIG. When this routine is executed via step Sg3 of the above-described accompaniment data generation processing routine, the CPU 4 advances the process to step Si1, and among the accompaniment patterns stored in the data ROM 7, from the base pattern BASE to the base read address BAD. Read the timing data and store it in the register TI
Store in MING. Then, in step Si2,
It is determined whether the tempo clock accumulated value stored in the register T matches the timing data stored in the register TIMING, that is, whether or not the timing is under the bass sound generation timing.

If it is not under the bass sound generation timing, the determination result is "NO", and the routine is completed once. If it is under the base sound generation timing, the determination result is "YES", The process proceeds to step Si3, in which the base read address BAD is incremented by one and is advanced. Next, in step Si4,
The pitch data is read from the base pattern BASE in accordance with the stepped base read address BAD, and stored in the register ONKAI (N1) corresponding to the sound channel to which the base sound is assigned. Then, at step Si5, the pitch data stored in the register ONKAI (N1) is converted into a pitch according to the root stored in the register ROOT.

Then, when proceeding to step Si6, the CPU
4 increments the base read address BAD, and in the subsequent step Si7, the base pattern BASE is set in accordance with the increased base read address BAD.
, The timing data (event interval) is read out and stored in the register ONTIME (N1). Subsequently, at step Si8, this register ONTIME (N1)
Is added to the timing data (event interval) stored in the register to form a note length, and this is stored back in the register ONTIME (N1).

In this way, the pitch data of the bass tone converted in accordance with the root note is stored in the register ONKAI (N1).
The note length (toning time length) is stored in the register ONTIME (N
When the preparation is completed, the CPU 4 proceeds to step Si9, increments the base read address BAD to correspond to the next event, and proceeds to step Si10 to increment the base read address. It is determined whether the BAD has exceeded the pattern end END. If the pattern end END has not been exceeded, the determination result is "NO", and the above-described processing after step Si1 is repeated. On the other hand, when the pattern end END is exceeded, that is, when the pattern is completely read, the determination result is “YES”,
Proceeding to step Si11, the base read address BA
After returning D to the start address, the above-described processing after step Si1 is repeated again.

As described above, in the bass sound data generation processing routine, the base pattern BA is determined based on the root (root note) determined in the chord detection processing routine.
A tone is instructed at a tone generation timing designated by the pattern BASE for a bass tone obtained by pitch-converting each tone constituting the SE.

J. Operation of Key Press Sound Data Generation Processing Routine Next, the operation of the key press sound data generation processing routine will be described with reference to FIGS. When this routine is executed via step Sg4 of the above-mentioned accompaniment data generation processing routine, the CPU 4 advances the processing to step Sj1 and, among the accompaniment patterns stored in the data ROM 7, the code reading address CAD from the code backing pattern CODE. And reads the timing data and stores it in the register TIMING. Then, in a step Sj2, it is determined whether or not the tempo clock accumulated value stored in the register T matches the timing data stored in the register TIMING, that is, whether or not the timing is below the chord sound generation timing.

If it is not under the chord sound generation timing, the judgment result is "NO", and this routine is completed once. However, if it is under the chord sound generation timing, the judgment result is "YES". The process proceeds to step Sj3 to determine whether or not there is a key number stored in the key buffer register NOTE. Here, when the key number is not stored in the key buffer register NOTE, that is, when the key is not depressed, the determination result is "YES", and this routine is completed.

On the other hand, when the key number is stored in the key buffer register NOTE, that is, when a depressed key exists, the determination result is "NO",
The process proceeds to the next step Sj4. In step Sj4, it is determined whether or not the key is a one-tone key press. If the key is a one-tone key press, the determination result is "YES", and the flow advances to the next Step Sj5 to execute the key buffer register NOTE
The key number stored in (1) is converted into pitch data and stored in the register ONKAI (1), and the subsequent step S
At j6, "1" is set in the register CH (1).
This completes the sound assignment of the key that has been depressed one note.

Thereafter, steps Sj7 to Sj10
Then, the sound assignment for the remaining sound channels is cleared. That is, in step Sj7, the register N
Is set to "2", and in the subsequent step Sj8, "0" is stored in the register CH (N) corresponding to the value of the register N. Next, in step Sj9, the register N
Is incremented by one and the value is incremented by one, and the next step S
At j10, it is determined whether or not the value of the incremented register N is equal to or more than N1, that is, whether or not the remaining tone generation channels to which the chord sounds are assigned have been cleared. If it is in the middle, the result of the determination is “NO”.
The process returns to step Sj7. If the sounding assignment has been cleared for the remaining sounding channels to which the chord sound is assigned, the determination result is “YES”.
And this routine is completed.

On the other hand, if two or more keys are pressed, the result of the determination at step Sj4 is "NO", and the CPU 4 advances the process to step Sj11 shown in FIG. In step Sj11, "1" is set in the register M.
In the subsequent step Sj12, “1” is set in the register N. Next, in step Sj13, the key buffer register NOTE corresponding to the value of the register N
It is determined whether the root note is held in (N).

Here, the key buffer register NOTE
If the root note is stored in (N), step Sj13
Is "NO", the process proceeds to step Sj14, and the value of the register N is incremented by one to advance. Next, in step Sj15, it is determined whether or not the value of the incremented register N has exceeded the maximum value (the number of key buffer registers). Then, when the value of the register N does not exceed the maximum value, the process returns to the step Sj13 in order to make the same judgment for the next key buffer register NOTE (N + 1).

On the other hand, when the value of the register N exceeds the maximum value, the decision result in the above step Sj15 becomes "YES", and the process proceeds to step Sj16. Step Sj
At step S16, the CPU 4 sets "0" to the register CH (M) corresponding to the value of the register M, and then proceeds to step S
At j17, the value of the register M is incremented by 1 to advance. Next, in step Sj18, it is determined whether or not the value of the stepped register M is equal to or larger than N1.
It is determined whether or not an instruction to stop sound generation has been issued for all sound channels to which a chord sound is assigned. here,
If the instruction has not been completed, the determination result is "NO", and the process returns to step Sj16. If the instruction has been completed, the determination result becomes "YES", and the process proceeds to step Sj26 described later.

Now, the key buffer register NOTE
If the root note is not stored in (N), the result of the determination in step Sj13 is "YES", the process proceeds to step Sj19, and the key buffer register NOT
It is determined whether E (N) is empty. here,
If it is empty, the determination result is "NO", and the process proceeds to step Sj14 described above. If it is not empty, that is, if the key number of the pressed key is stored, the determination result is "YES". ”And step S
The process proceeds to j20.

In step Sj20, the register NOTE
(N) is converted to pitch data and stored in the register ONKAI (N), and the subsequent step S
At j21, a predetermined value is stored in the register ONTIME (N). Then, in step Sj22, the note length obtained by adding the tempo clock accumulated value T to the value of the register ONTIME (N) is stored in the register ONTIME (N). Thereafter, the process proceeds to step Sj23 where the register M
Is set to “1” in the register CH (M) corresponding to the value of “1”, and in the succeeding step Sj24, the value of the register M is incremented by 1 to advance.

Next, in step Sj25, it is determined whether or not the value of the stepped register M is equal to or greater than N1, that is, whether or not the sounding instruction has been instructed for all sounding channels to which the chord sound is assigned. here,
If the instruction has not been completed, the determination result is "NO", and the process returns to step Sj14. However, if the instruction has been completed, the determination result becomes "YES", and step Sj
The process proceeds to 26. Then, in step Sj26,
In order to extract the next tone generation timing from the code backing pattern CODE, the code read address CAD is set to 3
Skip and step forward.

Subsequently, in step Sj27, it is determined whether or not the code read address CAD after the three address increments has exceeded the pattern end END. If not, the process returns to step Sj1. After the next step Sj28, the code read address CAD is reset to the start address, and the process returns to the above-described step Sj1.

K. Operation of Sounding Processing Routine Next, the operation of the sounding processing routine will be described with reference to FIG. Step Sa7 of the main routine described above
When this routine is executed through (see FIG. 2), the CP
U4 advances the process to step Sk1 shown in FIG. 15, sets initial value "1" in register N, and proceeds to step Sk2.
Then, it is determined whether or not the register CH (N) corresponding to the register N is "1", that is, whether or not the sound channel indicated by the register CH (N) is in the sound instructing state.
If it is not in the sounding instruction state, the determination result is “NO”.
Then, the process proceeds to step Sk7, which will be described later. If it is in the sounding instruction state, the determination result is “YES”, and the process proceeds to the next step Sk3.

At step Sk3, the register ONTIM
It is determined whether or not the value of E (N) matches the tempo clock accumulated value T. If they do not match, step Sk
The process proceeds to step 4 to instruct the sound source 8 to generate a sound based on the pitch data stored in the register ONKAI (N). on the other hand,
When they match, the process proceeds to step Sk5, and instructs the sound source 8 to mute the tone generated based on the pitch data stored in the register ONKAI (N). After this, the CPU
In step Sk6, the register CH (N) is reset to zero, and in step Sk7, the value of the register N is incremented by one to advance. In step Sk8, it is determined whether or not the value of the incremented register N has exceeded N1, and if it has, the present routine is completed, and if not, the process returns to step Sk2.

(3) Specific Operation Next, a specific operation of the first embodiment based on each processing described above will be described with reference to FIG. In the following, the chord backing pattern CO shown in FIG.
When the DE and the base pattern BASE are stored in the data ROM 7, it is assumed that the key pressing operation and the pedal operation shown in FIG. 16 are performed. First, when the first note of the first measure (C ₃ sound) is depressed, this is regarded because it is one sound depressed, the first sound and root (C), according to which Thus, the first beat of the chord backing pattern CODE and the base pattern BASE are sounded without pitch conversion.

[0076] Then, the second sound eyes while the pedal operating element 2 is turned on when the (G ₃ sound) is depressed, the second sound-th (G ₃ sound) and above the first note of the (C ₃ sound) but are subject to the code determination, in this case, since the two sound depressed becomes code uncertain, second sound th is depressed (G ₃ sound) is directly in accordance with the tone generation timing of the chord backing pattern cODE while the sound, bass sounds maintains C ₂ sound.
When further the pedal operating element 2 a third sound eye while being turned on (E ₃ sound) is depressed, the third sound eyes (E ₃
Sound) and the first and second sounds (C ₃ , G
₃ ) is a chord judgment target sound. In this case, the chord is judged to be C major. As a result, the third of the code backing pattern CODE and the base pattern BASE
The beats are pronounced without pitch conversion.

Similarly, when the fourth note (G ₃ note) and the first note (C ₃ note) of the second bar are depressed, the chord judgment target sounds are still “C, E, G”. So, CM (C
Major), and the chord backing pattern CODE and the base pattern BASE are sounded as they are. When the second sound-th second measure (A ₃ sound) is depressed, since the pedal operating element 2 is in the on state in this case, the code determination target sound "C, E, G, A "
In this case, the chord is determined as C ₆ (C major / six), and the chord backing pattern CODE is pitch-converted into E ₃ and A ₃ to be sounded accordingly.

As described above, according to the first embodiment, when a key is depressed while the pedal operator 2 is turned on, the sound of the depressed key is sequentially held as a chord judgment target sound. Therefore, even if an arpeggio playing method is performed in which chord components are depressed one by one, a natural accompaniment sound can be obtained without any musical discomfort.

B. 2. Second Embodiment (1) Differences from First Embodiment In the first embodiment described above, if the pedal operator 2 is in the ON state, the sound of a depressed key is always held as a chord determination target. In contrast to the case where the pedal operator 2 is cleared when the pedal operator 2 is turned off, the second embodiment clears the held chord determination target sound for each bar, thereby turning the pedal operator 2 off. The difference is that erroneous code determination can be avoided without any operation. Hereinafter, a timer interrupt processing routine according to the second embodiment will be described as an operation for implementing such a difference. The processes other than the present routine are the same as those in the first embodiment, and a description thereof will be omitted.

(2) Operation of the Second Embodiment (Operation of the Timer Interrupt Processing Routine) As in the first embodiment, when the automatic accompaniment is started through the switch processing routine, the timer interrupt is prohibited. Is released, and the CPU 4 executes a timer interrupt processing routine shown in FIG. 17 and proceeds to step Sm1. In step Sm1, the accumulated value of the tempo clock stored in the register T is incremented by one and incremented, and in step Sm2, it is determined whether or not the incremented tempo clock accumulated value has reached "49", that is, When the quarter note length is 12 clocks, it is determined whether one bar has elapsed.

Here, when one bar has not elapsed, the judgment result is "NO", and this routine is completed.
When the measure has elapsed, the determination result is "YES", and the process proceeds to the next step Sm3. In step Sm3,
The tempo clock accumulated value stored in the register T is reset to zero. In the following step Sm4, the initial value “1” is set in the register N, while the registers ROOT and TY are set.
Clear the root note and chord type stored in the PE.

Then, in steps Sm5 to Sm7, the tone generation assignment is cleared. That is, in step Sm5, "0" is stored in the register CH (N) corresponding to the value of the register N, and in the following step Sm6, the value of the register N is incremented by one, and stepped up, and in the next step Sm7, It is determined whether the value of the incremented register N is equal to or greater than N1, that is, whether or not the tone assignment has been cleared for each tone channel to which the chord tone is assigned.

Here, if the clearing is in progress, the determination result is "NO", and the process returns to step Sm5. If the sounding assignment is cleared, the determination result is "YES". , Steps Sm8 to Sm11
To set the key buffer register NOTE to an empty state. That is, in step Sm8, the initial value "1" is set in the register M, and in the subsequent step Sm9, the key buffer register NOTE corresponding to the value of the register M is set.
An invalid value FF indicating an empty state is stored in (M). Then, in step Sm10, the value of the register M is incremented by one and the step is advanced. In the following step Sm11, it is determined whether or not all the key buffer registers NOTE have been set to the empty state. When the processing is returned and the setting is completed, this routine is completed.

(3) Specific Operation Next, a specific operation of the second embodiment based on the timer interrupt processing routine described in detail above will be described with reference to FIG. In this operation description,
As in the first embodiment, the accompaniment pattern shown in FIG.
It is assumed that the code backing pattern CODE and the base pattern BASE shown in FIG. 6 are stored in the data ROM 7, and the key pressing operation and the pedal operation shown in FIG. 18 have been performed.

In this case, the key presses of the first to fourth notes in the first bar are performed in the same manner as in the first embodiment.
On the first beat, the chord backing pattern CODE and the base pattern BASE are pronounced without pitch conversion, and on the second beat, the chord backing pattern COD
While the sound of the depressed key in accordance with the tone generation timing of the E is directly sound, bass sound C ₂ maintains the sound continues third and chord backing pattern CO in the fourth beat
Both the DE and the base pattern BASE are sounded without pitch conversion.

When the first bar ends, the held chord judgment target sound is temporarily cleared by the above-described timer interrupt processing routine, and the first bar of the second bar is newly added.
The sound th (C ₃ sound) is depressed, because it is one sound depressed,
This first sound is regarded as the root tone (C), and the first beat of the chord backing pattern CODE and the base pattern BASE is generated without pitch conversion. Then, the second sound-th second measure (A ₃ sound) is depressed, since the pedal operating element 2 is in the on state, the target code determination is "C, A"
Although the, in this case, since the two sound depressed becomes code uncertainty, while the second sound th is depressed (A ₃ sound) is pronounced as it is in accordance with the tone generation timing of the chord backing pattern CODE, bass sound to maintain a C ₂ sound.

As described above, in the second embodiment, the chord judgment target sound held in response to the on operation of the pedal operator 2 is automatically cleared for each bar, so that the pedal operator 2 is turned off for each bar. An erroneous chord determination can be avoided without operation, and even if an arpeggio playing method is performed in which chord components are depressed one by one, a natural accompaniment sound can be obtained without any musical discomfort.

C. Third Embodiment (1) Difference from First Embodiment In the above-described first embodiment, if the pedal operator 2 is in the ON state, the sound of the key that has been depressed is always held as the chord determination target sound. , When the pedal operator 2 is turned off, this is cleared, whereas in the third embodiment, the pedal operator 2 is cleared.
Even when is turned on, when the sound of the depressed key is released in response to the key release (or when the envelope level reaches a predetermined value), the sound of the released key is recorded as a chord. The difference is that, by excluding from the judgment target sound, only the sounding sound can be set as a chord judgment target and erroneous chord judgment can be avoided.

(2) Operation of the Third Embodiment A main routine, a pedal processing routine, a keyboard processing routine, and an envelope monitoring processing routine according to the third embodiment that embody such differences will be described below. The processing other than these routines is the same as in the first embodiment described above, and a description thereof will not be repeated.

Operation of Main Routine The main routine according to the third embodiment is different from the first embodiment described above in that, as shown in FIG.
8 is provided with an envelope monitoring processing routine, and other steps Sa1 to Sa7 are the same as those in the first embodiment (see FIG. 2). The contents of the envelope monitoring processing routine will be described later.

Operation of Pedal Processing Routine Next, the operation of the pedal processing routine according to the third embodiment will be described with reference to FIG. As in the first embodiment,
With the automatic accompaniment started via the switch processing routine, step Sa of the main routine (see FIG. 19)
When the process proceeds to 3, the CPU 4 executes this routine and proceeds to step Sd1 shown in FIG. 20, and determines whether or not the operation event generated by the pedal operator 2 has changed. Here, when no pedal operation is performed and there is no event change, this routine is completed without performing any processing.
When the pedal operator 2 is released and transitions from the ON state to the OFF state and an OFF event occurs, the process proceeds to step Sd2, and the pedal flag PF is set to “0”.

On the other hand, when the pedal operator 2 is depressed and the state changes from the OFF state to the ON state and an ON event occurs, the process proceeds to step Sd3, where the pedal flag PF is set to "1" and the ON flag ONF is set to "1". Set to "0". The ON flag ONF is a flag that is set to “1” when a key is pressed in a keyboard processing routine described later, and indicates a key pressed state. Therefore, in this step Sd3, the key in the key pressed state is forcibly regarded as the key released state in accordance with the ON operation of the pedal operator 2. Then, in the next step Sd4,
The flag FIRST is set to "1".

Next, when the process proceeds to step Sd5, the CPU
4 clears the registers ROOT and TYPE, and sets "1" to the register N in the following step Sd6. Here, a register N is a register in which a value indicating a sound channel of the accompaniment tone generator is set.
TYPE is a register in which values representing the root note and chord type of a performance chord obtained by chord discrimination described later are set. Therefore, this step Sd
5 and Sd6, the registers N, ROOT, and TYPE are initialized.

Thereafter, in steps Sd7 to Sd13,
The sounding channel register CH (N) is all reset to zero to clear the sounding assignment, and furthermore, an invalid value FF (hexadecimal notation) indicating an empty state is stored and initialized in all of the key buffer registers NOTE (N). . As described above, the feature of the pedal processing routine according to the third embodiment is that, as shown in steps Sd3 and Sd4, the on-flag ONF is set to “0” in accordance with the on-operation of the pedal operator 2, while the flag is set to “0”. FIRSTF is set to "1". The purpose of the flag FIRSTF will be described later.

Keyboard Processing Routine The keyboard processing routine according to the third embodiment differs from the first embodiment in that the on-flag ONF is set to "1" in accordance with the key depression, as shown in FIG.
7 is newly added, and other steps Se1 to Se6 are the same as those in the first embodiment (see FIG. 6).

Next, the operation of the envelope monitoring processing routine will be described with reference to FIG. When this routine is executed via step Sa8 of the main routine (see FIG. 19) according to the third embodiment described above, the CPU 4 proceeds to step Sn1.
The flag FIRSTF is set to "1", that is,
It is determined whether or not the pedal operator 2 is on. Here, if the pedal operator 2 is in the off state, there is no need to monitor the envelope, so the determination result is “NO” and this routine is completed. Otherwise, the determination result is “YES” and the next The process proceeds to Step Sn2.

In step Sn2, it is determined whether or not the ON flag ONF is "1". This ON flag ONF
Is forcibly set to "0" when the pedal operator 2 is turned on, as described in the pedal processing routine according to the third embodiment described above. This is to exclude a case where the key is pressed at the same time as the child 2 is turned on. Therefore, in step Sn2, it is determined whether or not a key is pressed after the pedal operator 2 is turned on. If the key is not pressed, the determination result is "NO", and this routine is completed. If the key is depressed after the pedal operator 2 is turned on, the determination result is "YES" and the process proceeds to the next step Sn3.

In step Sn3, the sound of the key depressed after the pedal 2 is turned on, that is, the envelope level corresponding to the pitch data stored in the register ONKAI (1) is stored in the register ENVELOPE. Store. Next, in step Sn4, it is determined whether or not the current envelope is in the released state. Here, if the corresponding key is not released, the release state is not reached, the determination result is "NO", and this routine is completed. However, if the release state is entered in response to the key release, the determination result is "YES". ", And the process proceeds to the next step Sn5.

At step Sn5, the register ENVEL
It is determined whether the envelope level stored in the OPE has fallen below a predetermined value. Here, if the envelope level is higher than the predetermined value, the determination result is “NO”, and this routine is completed once. If the envelope level is lower than the predetermined value, the determination result is “YES”, and the next step S
The process proceeds to n6, where the flag FIRSTF is reset to zero.
Clear T and TYPE. And hereafter, step S
n8-Sn15 and sound channel register CH
(N) is all reset to zero to clear the tone assignment, and furthermore, an invalid value FF (hexadecimal notation) indicating an empty state is stored and initialized in all of the key buffer registers NOTE (N).

As described above, in the envelope monitoring routine, when the envelope of the key pressed after the pedal operator 2 is turned on is released, it is determined whether the envelope level falls below a predetermined value. It is determined that the sound of the released key is excluded from the chord judgment target sound if the sound is below the threshold.

(3) Specific Operation Next, specific operations according to the third embodiment will be described with reference to FIGS.
This will be described with reference to FIG. In this operation description, as in the first embodiment, the chord backing pattern COD shown in FIG.
It is assumed that E and the base pattern BASE are stored in the data ROM 7.

First, an example shown in FIG. 23 shows a mode in which when the envelope level reaches zero, it is excluded from the chord judgment target sound. In this case, the first measure of the first sound (C ₃ sound) is depressed to the ON operation after pedal operating element 2, and the envelope level reaches zero at the third beat. Then, the same chord sounds and bass sounds as in the first embodiment described above are produced from the first sound (C ₃ sound) to the third sound (E ₃ sound). Then, when the envelope level of the first sound (C ₃ sound) reaches zero, the first sound (C
₃ sound) is excluded from the code determination target sound, followed code judging target sound in response to key depression of the fourth sound th (G ₃ sound) becomes G ₃ sounds and E ₃ sound, E is bass of the lowest note , EG are pronounced as chord sounds.

Next, an example shown in FIG. 24 shows a mode in which the first sound envelope is excluded from the chord judgment target sound when it reaches the 5% level. Also in this case, the first
When the envelope level of the sound (C ₃ sound) has reached the 5% level, the first sound (C ₃ sound) is excluded from the code determination target sound, followed third sound th and the (E ₃ sound) Press code determination target sound becomes G ₃ sounds and E ₃ sounds according to the key, the result, E of the lowest sound is pronounced as bass, EG code tone.

As described above, according to the present invention, when a key is depressed while the pedal operator 2 is turned on, the sound of the depressed key is sequentially held as a chord judgment target sound. Therefore, even if an arpeggio playing method is performed in which a chord component is depressed one by one, a natural accompaniment sound can be obtained without any musical discomfort. Further, the chord judgment target sound held in response to the on operation of the pedal operator 2 is automatically cleared for each bar, or the envelope state of a key pressed after the on operation of the pedal operator 2 is changed. Since the sound of the key is excluded from the chord judgment target, chord judgment based on each sound that is actually sounding can be performed, and erroneous chord judgment can be avoided.

In each of the above embodiments, the sound of the key pressed in response to the ON operation of the pedal operator 2 is sequentially held and regarded as the chord judgment target sound. The mode is not limited to the operation unit 2, and for example, a mode using a switch disposed on a panel may be used, or a key is pressed only during a period in which a specific key (for example, a bass key range) is pressed in the automatic accompaniment mode. It is also possible to adopt a mode in which the sounds of keyed keys are sequentially held and regarded as a chord judgment target sound.

[0106]

According to the first aspect of the present invention, the sounds of the keys depressed while the operator for instructing the chord judgment is turned on are sequentially held as chord judgment target sounds. Since the chord judgment is performed based on the chord judgment target sound to be performed, a natural accompaniment sound can be obtained without any musical discomfort even if the arpeggio playing method of depressing the chord constituent sounds one by one is performed. According to the second aspect of the present invention, when the chord judgment target sound corresponds to a single key press, the chord judgment target sound is regarded as the root note, so that a natural accompaniment sound can be obtained without any musical discomfort. be able to. According to the third aspect of the invention, as a result of the chord judgment based on the chord judgment target sound, if the chord is uncertain, the chord judgment target sound is emitted as a chord sound in accordance with the timing specified by the accompaniment pattern. Therefore, it is possible to obtain a natural accompaniment sound without any musical discomfort. According to the fourth aspect of the present invention, the chord judgment target sound held in response to the on operation of the manipulator for instructing chord judgment is automatically cleared for each bar, so that the manipulator is turned off for each bar. Without this, erroneous chord judgment can be avoided, and even if an arpeggio playing method is performed in which chord components are depressed one by one, a natural accompaniment sound can be obtained without any musical discomfort. According to the invention described in claim 5, the envelope of the sound of the depressed key is released after the operator for instructing the chord is turned on, and the envelope level reaches a predetermined value. Since the sound of the key is erased from the target sound for chord judgment, chord judgment based on each sound that is actually sounding can be performed, and erroneous chord judgment can be avoided. Even if the arpeggio playing method to be operated is performed, a natural accompaniment sound can be obtained without any musical discomfort.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment according to the present invention.

FIG. 2 is a flowchart showing an operation of a main routine.

FIG. 3 is a flowchart showing the operation of a timer interrupt processing routine.

FIG. 4 is a flowchart showing an operation of a switch processing routine.

FIG. 5 is a flowchart showing an operation of a pedal processing routine.

FIG. 6 is a flowchart showing the operation of a keyboard processing routine.

FIG. 7 is a flowchart showing an operation of a code detection processing routine.

FIG. 8 is a flowchart showing an operation of a code detection processing routine.

FIG. 9 is a flowchart showing the operation of a code detection processing routine.

FIG. 10 is a flowchart showing an operation of an accompaniment data generation processing routine.

FIG. 11 is a flowchart showing an operation of a chord sound data generation processing routine.

FIG. 12 is a flowchart illustrating an operation of a bass sound data generation processing routine.

FIG. 13 is a flowchart illustrating an operation of a key press sound data generation processing routine.

FIG. 14 is a flowchart showing an operation of a key press sound data generation processing routine.

FIG. 15 is a flowchart showing an operation of a sound generation processing routine.

FIG. 16 is a diagram for explaining a specific operation of the first embodiment.

FIG. 17 is a flowchart showing the operation of a timer interrupt processing routine according to the second embodiment.

FIG. 18 is a diagram for explaining a specific operation of the second embodiment.

FIG. 19 is a flowchart showing an operation of a main routine according to the third embodiment.

FIG. 20 is a flowchart showing the operation of a pedal processing routine according to the third embodiment.

FIG. 21 is a flowchart showing the operation of a keyboard processing routine according to the third embodiment.

FIG. 22 is a flowchart illustrating an operation of an envelope monitoring processing routine according to the third embodiment.

FIG. 23 is a diagram for explaining a specific operation of the third embodiment.

FIG. 24 is a diagram for explaining a specific operation of the third embodiment.

FIG. 25 is a diagram for explaining a conventional example.

FIG. 26 is a diagram showing an example of an accompaniment pattern.

FIG. 27 is a diagram for explaining a conventional example.

[Explanation of symbols]

Reference Signs List 1 panel switch group 2 pedal operator (chord holding means, chord erasing means) 3 keyboard 4 CPU (chord holding means, chord erasing means, accompaniment means) 5 ROM 6 RAM (chord holding means, accompaniment means) 7 data ROM (accompaniment means) Means) 8 sound sources (accompaniment means)

Claims (5)

[Claims] 1. A code holding means, comprising: a manipulator for instructing chord judgment; a chord holding unit for sequentially holding a sound of a key depressed while the manipulator is turned on as a chord judgment target sound; A code erasing unit for erasing a code determination target sound held by the code holding unit in response to the OFF operation of the code, and performing a code determination based on the code determination target sound held by the code holding unit, and according to the determined code. An automatic accompaniment apparatus comprising: an accompaniment unit that produces an accompaniment sound obtained by pitch-converting each of the sounds constituting the accompaniment pattern in accordance with timing specified by the accompaniment pattern. 2. The automatic accompaniment device according to claim 1, wherein the accompaniment means regards the chord judgment target sound as a root note when the chord judgment target sound corresponds to a single key press. 3. When the chord is determined based on the chord judgment target sound and the chord is uncertain as a result, the chord judgment target sound is produced as a chord sound in accordance with the timing specified by the accompaniment pattern. The automatic accompaniment device according to claim 1, wherein: 4. A code holding means comprising an operation element for instructing chord judgment, and sequentially holding a sound of a key depressed while the operation element is turned on as a code judgment target sound; The chord judgment target sound held by the means is 1
Chord erasing means for automatically erasing each bar, performing chord judgment based on the chord judgment target sound held in the chord holding means, and pitch-converting each sound constituting the accompaniment pattern according to the determined chord. An automatic accompaniment device comprising: an accompaniment means for producing an accompaniment sound according to a timing specified by the accompaniment pattern. 5. A code holding means, comprising: a manipulator for instructing chord judgment; a chord holding means for sequentially holding a key sound depressed while the manipulator is turned on as a chord judgment target sound; Is turned on, the envelope of the sound of the key pressed is released, and when the envelope level reaches a predetermined value, the sound of the key is held in the chord holding means. A chord erasing means for erasing from a chord judgment means based on a chord judgment target sound held in the chord holding means, and an accompaniment sound obtained by pitch-converting each sound constituting an accompaniment pattern according to the determined chord. And an accompaniment means for producing a sound in accordance with the timing specified by the accompaniment pattern.