JP2663938B2 - Electronic musical instrument with chord identification function - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic musical instrument, and more particularly to an electronic musical instrument having a keyboard and a chord judgment function.

[0002]

2. Description of the Related Art In an electronic musical instrument having a keyboard, it is possible to play an accompaniment chord along with a melody sound.

[0003] However, it is not always easy for a beginner to always press a chord component sound accurately.

Accordingly, for example, Japanese Patent Publication No. Sho 63-39078
Japanese Patent Laid-Open Publication No. H11-163873 discloses a technique in which the constituent tones of a key depressed for a chord are represented in a 12-note scale configuration, and are generated only when a complete chord is formed by comparing with a stored combination of chords. ing.

[0005] Also, Japanese Patent Application Laid-Open No. 2-173697 discloses that
When a melody sound is played, a melody sound and a melody additional sound that is a fifth lower sound that produces a melody sound and a harmony depending on the rhythm type are generated, or a melody performance sound in a certain section is played before another melody sound. Various automatic accompaniment techniques are disclosed, such as performing a canon performance by repeatedly generating melody performance sounds.

Japanese Patent Application Laid-Open No. 63-193200 classifies chords for accompaniment into major, minor and seventh chords.
By using accompaniment patterns corresponding to each chord type group,
It discloses that the capacity of the accompaniment pattern memory is reduced.

The number of sound channels in a keyboard electronic musical instrument is considerably smaller than the number of keys on a normal keyboard. There are usually enough sounding channels to sound only the sound of the key pressed with both hands, but if you apply sustain or increase the number of notes that make up a chord with automatic accompaniment, it will sound Insufficient channels.

[0008] Japanese Patent Publication No. Hei 1-27437 discloses that when the number of sounds to be produced exceeds the number of sounding channels, the sound in the highest range being produced or the sound with the smallest volume envelope is eliminated. Discloses a truncation process that eliminates the oldest or most attenuated sound.

Japanese Unexamined Patent Application Publication No. Sho 62-135892 discloses that the sounding channel for key depression and the sounding channel for automatic accompaniment are separated, and the sounding channel for automatic accompaniment is also used for key depression when not used for automatic accompaniment. To disclose the technology.

[0010]

In an electronic musical instrument having a chord discriminating function according to the prior art, if a wrong key is pressed for a chord, no chord is generated, and even if a player desires an inventive accompaniment, it is decided. There was only a pattern accompaniment.

An object of the present invention is to provide an electronic musical instrument provided with a chord discriminating function that makes it extremely easy to play an accompaniment chord.

It is another object of the present invention to provide an electronic musical instrument having a chord discriminating function capable of generating an appropriate chord even with a simple or incomplete playing operation.

[0013]

According to the electronic musical instrument of the present invention, a pitch information input means for inputting pitch information, and a root note of a chord is determined by a combination of the pitch information input by the pitch information input means. First chord determining means for determining the type;
Pitch input determining means for determining whether or not new pitch information has been input when the first chord determining means has failed to determine the root and type of the chord; and the pitch input determining means , When it is determined that new pitch information has been input, a second chord for determining the root and type of chord based on a combination of previously input pitch information and newly input pitch information Discriminating means, pattern storing means for storing an accompaniment pattern consisting of time information and an accompaniment sound associated with the time information, and a root note and a type of the chord discriminated by the first or second chord discriminating means. Output means for converting the accompaniment sound of the accompaniment pattern into accompaniment sound information based on the accompaniment pattern information, and outputting the accompaniment sound information converted based on the time information of the accompaniment pattern.

[0014]

By giving priority to the input pitch information that satisfies a predetermined relationship, it is possible to preferentially extract important pitch information in the chord configuration.

Further, when necessary pitch information is insufficient in the chord configuration, the chord can be determined by supplementing the insufficient pitch information even if incomplete pitch information is input.

In this way, a desired chord can be generated by inputting important pitch information in the chord configuration.

Normally, chords are determined in relation to each other in the flow of music.

The pitch information of the chord component sound being generated is stored,
If the newly input pitch information does not form a complete chord, a new chord is determined by discriminating the pitch information of the new input with the pitch information of the stored chord constituent tones. By doing so, it is possible to obtain a chord that matches the flow of the music. However, when the newly input pitch information constitutes a complete chord, it is not necessary to consider the previous input, so that a chord based on the new input may be generated.

By using this function, for example, when a chord is changed, only the chord component to be changed need be inputted.

In this way, for example, a chord can be generated by inputting at least one important pitch information in the chord configuration, and any chord can be determined from any input.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to embodiments.

FIG. 1 is a block diagram showing the system configuration of the electronic musical instrument. The keyboard 11 has a plurality of keys, and can generate a desired pitch signal by pressing a desired key. The keyboard 11 has a keyboard interface 12
Is connected to the bus 13 via the bus, and supplies a pitch signal by a key press to the bus 13.

The electronic musical instrument has an accompaniment sound correction and an additional function. The electronic musical instrument also has an operator 14 such as a foot pedal for selectively activating the function. The start signal by the operation of the operator is supplied to the bus 13 via the operator interface 15.

The electronic musical instrument has an operation panel 21 for determining a performance environment. The operation panel 21
For example, it has a configuration as shown in FIG. In the illustrated configuration, on the operation panel 21, a power switch 31, tempo conversion keys 32a and 32b, an ABC key 33, a harmonic key 34, and the like are arranged. The power key 31 is a switch that alternately turns on and off when pressed, for example. The tempo conversion keys 32a and 32b are keys for increasing or decreasing the tempo of the accompaniment sound. The ABC key 33 is a key for selecting a mode for adding a chord as an accompaniment sound, and alternately turns on and off when pressed. The harmonic key 34 is used as an accompaniment sound.
This key is used to select a mode for generating a melody sound 3rd below, 5th below the melody sound, and an additional sound with good harmony, and alternately turns on and off when pressed.

To the right of these keys are three indicators 3
5a, 35b and 35c are provided. Display 35a
Indicates the ABC mode, and the display 35b
Indicates the harmony mode.
c indicates the tone color. These indicators 35
a, 35b, and 35c are composed of two 7-segment display elements, and can display numbers from 00 to 99.
In the illustrated configuration, the indicators 35a and 35b are ABC
Indicates that no mode or harmony mode is selected. The display 35c indicates that the 00th tone color is selected as the tone color.

In the figure, a plurality of tone color selection switches 37, a plurality of ABC style selection switches 38, and a plurality of harmony mode selection switches 39 are provided below. For example, when the ABC mode is turned on and a desired switch of the ABC style selection switch group 38 is pressed,
The ABC switch is selected and displayed on the display 35a. Similarly, when the harmony mode is selected by the harmony key 34 and a desired switch of the harmony mode selection switch group 39 is pressed, the harmony mode is selected and displayed on the display 35b. Tone selection switch group 3
When a timbre is selected in step 7, the timbre is displayed on the display 35c.

Returning to FIG. 1, the performance environment selection signal selected on the operation panel 21 is supplied to the bus 13 via the panel interface 22.

The bus 13 has a CP having an arithmetic processing function.
U16, a ROM 17 for storing programs such as a tone signal generating program, etc., a RAM 18 having flags, registers, buffer memories, etc. for storing temporary parameters and the like generated when the CPU 16 performs arithmetic processing according to the programs stored in the ROM 17. Is also connected. According to the pitch signal and control signal supplied from the bus 13, C
PU 16 generates musical tone forming parameters,
The tone generator 23 supplies a tone signal based on these signals and supplies the tone signal to the sound system 24. The sound system 24 converts, for example, a digital musical tone signal into an analog musical tone signal, assigns a volume, a localization, and the like, and supplies a musical tone signal to each of the left channel speaker 25 and the right channel speaker 26 to generate a desired musical tone.

The CPU 16 controls the operation of the entire electronic musical instrument in addition to the above. Also, a hardware function such as a timer interrupt is realized. Since the tone generator 23 generates a plurality of sounds, the tone generator 23 may operate in a time-division manner or may include a multiple tone generator.

Before describing the tone processing, flags, registers, buffer memories and the like provided in a RAM and the like will be briefly described.

I, k, l, m, n, p: Store a sequentially increasing coefficient or the like which is a general-purpose counter.

KON: a key-on flag indicating key depression.

KOFF: a key-off flag indicating a key release.

FS: A flag indicating that the operator is being operated.

ABC: A flag indicating the accompaniment mode.

RUN: A flag indicating that the ABC mode is running.

TCNT: A register for storing the count number of the timer.

KC: A register for storing a key code indicating a pitch.

KN: A register for storing the number of keys depressed in the chord playing chord area (key number).

KB: a buffer memory for storing a pitch signal generated by depressing a key in a code area.

I: A pointer indicating the root position.

"Perfect": key press KB in the code area
Is a flag indicating that a complete chord is formed.

ROOT: A register for storing a root sound of a chord.

TYPE: A register for storing types of chords such as major, minor and seventh.

BASE: A register for storing a key code of a bass sound.

PTN: A register for storing the pitch of a key press corresponding to 12 bits of 1 to 12.

CT: A register for storing a code type.

KEY: Key code (twelve-tone notation) adopted as a chord component sound candidate.

KEY3: Third degree sound register.

KEY7: Register for seventh-degree sound.

KEY2: Register for second-degree sound.

KEY5: 5th-system sound register.

KO: A register for storing the previous value of KEY.

Find: A flag indicating that detection has been performed.

CMPLT: A flag indicating chord formation.

SK: A register for storing a key to be searched.

CPTN: A register for storing a code pattern.

TC: a register for storing a tone color.

MD: A register for storing the harmony mode.

MDnKON: Register for storing key-on processing for each mode.

OFS: A register for storing offset information.

TH: a register for storing a threshold value.

PAN: A register for storing localization (panning).

IT: A register for storing initial touches.

STYL: A register for storing the ABC mode style.

HMNY: Harmony mode flag.

In addition, various flags are provided for the entire electronic musical instrument.
Although a register, a buffer memory, and the like are provided, description thereof will be omitted.

FIG. 3 shows a flowchart of the main processing routine.

When the main processing starts, initialization is first performed in step S11. The contents of the initialization include the initial setting of the tone generator circuit and each interface, the clearing of each register, and the like. For example, ABC =
0, RUN = 0, HMNY = 0, STYL = 0, MD =
1, TC (0) = 0, FS = 0, etc.

Subsequently, the key release on the keyboard is checked by key scanning in step S12. As a result of the key scan, it is determined in step S13 whether or not there is an event such as a key press or a key release. If there is a key press or key release event,
In step S14, key processing is performed. If there is no event such as key press or key release, step S14 is bypassed.

Subsequently, in order to check whether or not an operation has been performed on the panel, the panel is scanned in step S15, and it is determined in step S16 whether or not an operation has been performed on the panel. If there is an operation on the panel, panel processing is performed in step S17. If there is no operation on the panel, step S17 is bypassed.

Subsequently, in order to check whether or not the operating element has been operated, the operating element is scanned in step S18, and it is determined in step S19 whether or not the operating element has been operated. If there is an operation of the operator, an operator process is performed in step S20. If there is no operation of the operator, step S20 is bypassed. Thereafter, the process returns to the key scan in step S12.

Hereinafter, the key processing in step S14, the panel processing in step S17, and the operator processing in step S20 will be described in more detail.

The key processing is shown in FIG. 4, and includes the ABC / KON processing shown in FIG. 5, the KON processing shown in FIG. 14, and the KOFF processing shown in FIG.

The ABC / KON processing shown in FIG. 5 further includes a code detection processing shown in FIG. 6 and a new key press addition processing shown in FIG. 13. The code detection processing in FIG.
The predetermined sound search process and the chord establishment process of FIG. The KON processing of FIG. 14 is further performed by the MD1KO of FIG.
N processing is included. Also, the KOFF processing of FIG.
7 MD1KOFF processing and the like.

The panel processing is shown in FIG. 18, and further includes the ABC processing in FIG. 19, the harmony processing in FIG. 20, and the like. The control processing is shown in FIG. In addition, the interrupt processing shown in FIGS. 21 and 22 is performed separately from the processing included in these main flows.

The key processing shown in step S14 of FIG. 3 will be described in detail below.

FIG. 4 shows a flowchart of the key processing routine. When the key processing starts, it is determined in a step S21 whether or not there is a key pressing processing. K by key press
If there is an ON signal, the process proceeds to step S22, and it is determined whether the accompaniment flag ABC indicating the accompaniment mode is 1 or not.

If the ABC flag is 1, step S2
Proceed to 3 to perform ABC / KON processing. If there is a key-on and the ABC mode is not selected, the process proceeds to step S25 to perform a KON process.

If there is no key-on, step S26 is completed, and KOFF processing is performed.

Thereafter, a synchronous start is performed.
That is, it is determined whether or not the traveling flag RUN is -1 in step S24. The running flag RUN takes a value of 1 during running, -1 during standby, and 0 during stopping. If the running flag RUN is -1, the process proceeds to step S27, in which "1" indicating that the vehicle is running is set to RUN, and the process proceeds to step S28, where the timer counter TCNT is set to 0 and time counting is started. Then, return. If the running flag RUN is not -1 in step S24, the routine immediately returns.

ABC / KO shown in step S23 of FIG.
The N processing will be described in detail below.

FIG. 5 shows a flowchart of the ABC / KON processing. When the process is started, in step T1, it is determined whether or not the key code of the key pressed on the keyboard is in the chord designation area. That is, in this embodiment, KC
= 53 (F2) is a melody designation area, and below it is a chord designation area. If there is a key press in the chord designation area of KC = 53 or less, step T2 follows the arrow of YES.
And the number of keys depressed in the code area (key number) is stored in the register KN. For example, the key number KN = 3 when the key of the domeso chord is pressed.

Subsequently, in step T3, the key codes of the keys that are KON in the chord designation area, which is the chord designation area, are stored in the key buffer KB array in the order of the smallest key code in KB (1) to KB (KN). I do. That is, KB
(1) indicates the lowest sound, and KB (KN) indicates the highest sound.
Subsequently, 1 is stored in the flag "perfect" in step T4, and the process proceeds to code detection in step T5.

The flag "perfect" takes "1" when the key code of the array KB forms a complete chord, and takes "0" when the key code does not form a complete chord. If no chord is detected in the chord detection step T5, the flag per
fact is changed to 0.

In the chord detection, as described later, a chord component sound is extracted from an array KB for storing a key code of a depressed key, and when it is insufficient, a predetermined chord component sound is added, and a predetermined pattern is added. Code detection is performed by comparison. Unnecessary key presses are ignored in the chord configuration, and necessary keys are added in the chord configuration. When a chord is composed only of actual key depression, the flag "perfect" is set to 1
Is kept. When any constituent sound is added to form a chord, the flag "perfect" is set to "0".

At step T6, it is determined whether the flag "perfect" is "1". If the flag "perfect" is not "1", the process proceeds to step T7 to perform a new key press addition process. Thereafter, code detection is performed in step T8. Step T
When a chord is generated in both 5 and T8, priority is given to the subsequent step T8. If the number of key presses is 2 or less, step T5,
The process may bypass T6 and enter step T7.

If the flag "perfect" is "1", the processes of steps T7 and T8 for adding a new key press and detecting a code are bypassed. In the new key press addition, the new key press and the previous key press (the currently sounding sound) are examined together to determine a new chord.

The chord is determined by these steps. The root sound of the chord is the sound shown in KB (i).

In step T9, the key buffer KB (i) indicating the root tone is modulo-processed by 12 to indicate a 12-note scale, and stored in the chord root register ROOT. In addition, the chord is greatly increased to three major, minor, and seventh
And classify these code types into CT
= 0, CT = 1, CT = 2. This code type CT (k) is stored in the register TYPE.

Subsequently, in step T10, it is determined whether or not the lowest sound KB (1) in the key buffer KB is a constituent sound of the chord. If the lowest sound KB (1) during key depression is a constituent sound of the chord, this sound is stored in the register BASE as a base sound in step T11. When KB (1) is not a constituent sound of a chord, KB is a root sound of a chord.
(I) is bass-sounded and stored in the register BASE.

Thereafter, in step T12, the K determined as a chord component sound in the chord detection step
EY (1) to KEY (j) are stored in the previous chord sound register KO.
(1) Copy to KO (j) and prepare for the next processing. In the next process, when a new key press addition process is required, the key code KC of the register KO is used.

FIG. 6 shows a flowchart of the code detection process in steps T5 and T8 in FIG.

First, in step X1, the register i is set to 0. This i is used to designate a key being depressed as the root sound, and serves as a pointer to the root sound. It is used to search for chords, assuming the root note in order from the lowest note.

Subsequently, the value of the register i is incremented (step X2). That is, i starts from 1 (the lowest note). In the next step X3, 1 is set to the register j. This j is a count indicating the number of pitch signals KEY to be adopted as chord constituent tones.

In step X4, array KB (i)
(Where i = 1) is stored as an array KEY (j) = KEY (1). That is, initially, the key code of the lowest tone during key depression is adopted as a reference tone for chord detection. At the same time, j is advanced. Thereafter, a key code which is another chord component sound is searched for the reference sound.

At step X5, the reference sound KEY
A search is made to see if there is a third-degree key press for (1). The key press search process will be described later in detail. Subsequently, in step X6, it is determined whether or not a third key press is detected. When it is detected, the flag find is set to 1, and the result of the three-time key press search is determined by determining whether this flag is 1 or not. If a third key press is detected, the flow advances to step X7 according to the arrow of YES, and the KEY detected as the third key press is detected.
3 is stored in the register KEY (j) = KEY (2), and j
Step up the value of.

If the third key depression is not detected, the flow advances to step X8 to set a third (long third) pitch signal of the reference tone KEY (1) +4, and the register KEY (j) = KE
Stored in Y (2). In this case, since the pitch signal other than the key depression is added, 0 is stored in the flag "perfect". Also, the value of j is incremented.

That is, if a third-tone tone is depressed, it is adopted, but a standard third-tone pitch signal is added even if the key is not depressed.

Then, the flow advances to a step X9 to search for a 7-degree key press. In this case, when a 7th key press is detected,
The pitch signal is temporarily stored in the register KEY7, and the flag find is set to 1. In step X10, flag find
Is 1 or not. When it is 1, since a 7-degree key depression has been detected, follow the arrow of YES and follow step X1.
Proceeding to 1, the pitch signal stored in the register KEY7 is stored in the register KEY (j) as the next chord component sound candidate, and the value of j is incremented. In the seventh key press search, when a seventh musical tone is not detected, a pitch signal is not added, and the process directly proceeds to the next step.

Next, in step X12, a second key press is searched. The second-degree key depression search is performed in the same manner as the seventh-degree key depression search. When a key depression is detected, the flag fin is set.
1 is set to d, and the pitch signal is stored in the register KEY2. In step X13, the flag fin
It is determined whether d is 1 or not, and if find is 1, YES
The process proceeds to step X14 according to the arrow of
The register KEY uses the pitch signal of Y2 as a chord candidate.
(J) and increment the value of j. Note that the flag f
When ind is 0, step X14 is bypassed.

Next, at step X15, a five-degree key press search is performed. In step X16, it is determined whether or not the fifth-degree key depression is detected based on whether or not the flag "find" is "1". If detected, step X
To the register K for storing the detected pitch signal
The value of EY5 is stored in the chord constituent tone candidate register KEY (j). If no fifth-degree key depression is detected, the flow advances to step X18 according to the NO arrow, and the standard fifth-degree musical tone KEY (1) +7 (complete fifth) is registered as a chord candidate tone register KEY (j). And the flag "perfect" is set to 0 to indicate that the pitch signal has been added.

As described above, the key presses of the third, seventh, second, and fifth systems are searched, and when there is a key press, the pitch signal thereof is extracted. For the third and fifth pitch signals, a standard pitch signal is added when no key is depressed. Key press signals that do not satisfy these relationships are ignored.

On the basis of the signal thus obtained,
At step X19, it is determined that the code is established. That is, it is determined whether or not the sound {KEY (j)} adopted as the chord component sound matches a predetermined chord. If a chord is formed, 1 is set to the flag CMPLT.

Next, at step X20, the flag CM
It is determined whether PLT is 1 or not. If a chord holds,
CMPLT is 1 and returns according to the YES arrow. When the chord is not established, the process returns to step X2 according to the arrow of NO. At this time, of the key presses, the reference sound is shifted to the next higher sound, j is returned to 1, and the same procedure is repeated. As described above, when a chord is not established, a root note candidate is changed to search for a chord.

Even if no chord is detected, the process may proceed to the next step after some chord search.

Next, referring to FIG. 7 to FIG.
The key press search processing of the degree system, the seventh degree system, and the fifth degree system will be described.

FIG. 7 shows a flowchart of the three-time key depression search process. When the process starts, step X21
, The flag find is set to 0. Next, in step X22, a key code on the fourth semitone (up to the third major) of the pitch KEY (1), which is a reference assumed for the root sound of the chord, is created and stored in the register SK indicating the key to be searched. In step X23, a search is performed by a key buffer search to determine whether there is a key press that matches this register SK. The key buffer search will be described later.
When the corresponding key press is detected, the flag find is set to 1. Next, in step X24, the flag f
It is determined whether ind is 1 or not. If the flag "find" is 1, since the third key depression has already been detected, the pitch stored in the register SK is adopted as the chord component KEY3 according to the arrow of YES (step X31).
If not detected, step X follows the arrow of NO.
Proceeding to step S25, the data is stored in the register SK indicating a key for searching for a sound of the reference sound KEY (1) +3 (minor third of the reference sound). For this SK, a key buffer search is performed in step X26. When the corresponding sound is detected, the flag fi
nd is set to 1. Subsequently, at step X27, the flag f
It is determined whether ind is 1 or not. When find is 1,
Since the corresponding sound has been detected, the sound is adopted as the chord component KEY3 according to the arrow of YES (step 3).
1). If not detected, the process proceeds to step X28, where K
The sound of EY (1) +5 (3rd higher than the reference sound) is stored in a register SK indicating a search key, and a key buffer search is performed in step X29 based on the SK. Subsequently, it is determined in a step X30 whether or not the flag "find" is "1".
If find is 1, the corresponding sound has been detected.
According to the arrow of ES, the value of the register SK indicating the sound is stored in the register KEY3. If not detected, the routine returns.

As described above, in the third key press search, the third higher key, the third higher key, the third key,
Search for higher sounds.

FIG. 8 shows a flowchart of the second key press search process. Steps X32 to X39 are steps equivalent to steps X21 to X27 and X31 in FIG. 7, but correspond to the second key press, and the pitch to be searched is KEY.
(1) +2 and KEY (1) +1. That is,
The sound of the reference sound of the second major and the minor second of the reference sound are searched.
When the second key press is detected, the sound is stored in the register KEY2.

FIG. 9 shows a flowchart of the seventh-degree key depression search process. First, in step X41, the flag find
Is set to 0, a search is made for a sound that is 7th above the reference sound in steps X42, X43, and X44, and if there is a corresponding key press, that sound is stored in KEY 7 (step X51). Subsequently, at steps X45, X46 and X47, KEY (1) +
11 (sound 7th above the reference sound) is searched, and if there is a corresponding sound, that sound is stored in the register KEY7 in step X51. In steps X48, X49 and X50, K
EY (1) +9 (seventh minor tone of the reference tone) is searched, and if there is a corresponding tone, the tone is stored in the register K in step X51.
Store it in EY7. Then, return.

FIG. 10 shows a flowchart of the fifth-degree key depression search process. First, in step X52, the flag fin
d is set to 0, and KEY is set in steps X53 and X54.
(1) +7 (sound 5th above the reference sound) is searched, and if there is a corresponding sound, the detection flag find is set to 1 and discriminated by this flag (step X55), and its pitch (SK value) ) Is stored in the register KEY5 (step X62). If not detected, KEY (1) +6
(Sound 5th above the reference sound) is stored in the search key register as a search target (step X56), a key buffer search is performed (step X57), and the detection result is determined based on whether or not flag find is 1 (step X57). Step X58). If detected, the value of SK indicating the sound is stored in the register KE.
It is stored in Y5 (step X62). If the result is NO in step X58 as well, the process proceeds to step X59, where KEY is used.
(1) Search key SK for +8 (sound 5 degrees higher than the reference sound)
And a key buffer search is performed in step X60. Subsequently, it is determined whether or not the flag find is 1 (step X61). If detected, the value of the register SK indicating the search key is registered in the register KEY5 according to the arrow of YES.
(Step X62). Then return.

As described above, when there is a key press satisfying the relationship of the third degree system, the second degree system, the seventh degree system, and the fifth degree system, the key depression is detected.

The key buffer search used in the flowcharts of FIGS. 7 to 10 will be described with reference to FIG.

When the key buffer search process is started, in step U1, the number i + 1 one higher than the counter i indicating the set root tone is set in the search counter k. In other words, since the root sound does not need to be searched, the search starts from the next key depression. Next, in step U2, k
It is determined whether or not the key buffer KB (k) indicating the key depression signal matches the search key SK.

When the key pressed and the search key match, YE
Move to step U5 according to the arrow of S, and register fin
Set 1 to d and return. When there is no corresponding key depression, the process proceeds to step U3 according to the arrow of NO, and the value of the counter k is incremented. When the incremented value of k exceeds the maximum value kM, it means that all key presses have been searched, and the routine returns. If the maximum value has not been exceeded, the process returns to step U2, and the same processing is repeated.

Next, a description will be given of a code establishment determination process performed in the code detection process of FIG.

FIG. 12 shows a flowchart of the code establishment determination processing. Prior to this processing, a sound that is a chord constituent sound candidate is stored in the register {KEY}.

When the code satisfaction determination process starts, 0 is stored in a flag CMPLT in a step X65. In step X66, 0 is stored in the register PTN. The register PTN is a 12-bit register, and corresponds to a 1-octave range corresponding to each 12-bit bit. In step X67, 1 is stored in the counter l. A preparatory stage is performed by these steps.

At step X68, the register PTN in which the difference between the chord component candidate KEY (l) of interest and the reference tone KB (i) (root tone) is converted to an exponent with a base of 2 and reset to 0 To the number of new PTNs. That is, if this number is expressed in binary notation, a key depression pattern in 12-tone notation is obtained.

Next, in step X69, the counter 1 is incremented, and in step X70, it is determined whether or not the counter number 1 has become larger than the maximum value j of the array KEY adopted as the chord component. When l becomes larger than j, there is no longer any candidate sound, so step X7
Proceed to 1. If the maximum value is not exceeded, step X68
Return to In this manner, the processing of step X68 is performed on all the sounds in the array {KEY}, whereby the register P corresponding to the pattern of all chord constituent sound candidates
Get the value of TN. After fetching all the arrays {KEY}, in step X71, 0 is stored in the pointer k for referencing the chord table. A plurality of chord patterns {CPTN (k)} to be considered are stored in the chord table. The new chord candidate tone pattern PTN obtained by discriminating from the key depression input is the k-th chord pattern CPTN.
It is determined in order from k = 1 whether or not they are equal to (k). If not equal, the pointer k is incremented in step X73 according to the arrow of NO, and it is determined in step X74 whether k exceeds the maximum value 24 of the chord table. Note that the maximum value of the chord table need not be 24. If the maximum value has not been exceeded, the process returns to step X72. In this way, it is determined whether the detection pattern PTN is equal to any of the chord tables. If it is equal to any of the code patterns, the process proceeds from step X72 to step X75, and 1 is stored in the flag CMPLT. When all the tables have been referenced or when a chord is established, the routine returns.

The code detection process shown in FIG. 6 is performed by the various processes described above.

Next, the new key press addition process used in step T7 of FIG. 5 will be described. This process refers to the key code of the chord being sounded in addition to the key code that was pressed.

FIG. 13 shows a flowchart of the new key press addition process. When there is a new key press, a signal corresponding to the key press is latched in the key buffer KB. In step Y1, the counter m is set to 1. Then, in step Y2, the m-th sound KO of the chord currently being generated.
It is determined whether or not (m) (the lowest note is initially selected) is in the array KB by a new key press. In some cases, it is not necessary to employ the same sound repeatedly, so that m is immediately advanced (step Y6). If KO (m) is not in the array of new keypresses KB, the process proceeds to step Y3 according to the arrow of NO, and a sound related to KO (m) and the semitone is arranged in the array KB
Is determined. Since a semitone-related sound is not used as a chord component sound, when there is a semitone-related sound, the old tone KO (m) is usually ignored. That is, step Y4
In, it is determined whether or not KO (m) is the root sound of a chord, and if not, the KO (m) is ignored and the sound by a new key press is prioritized.

If KO (m) is the root sound, the process proceeds to step Y5 to adopt both the root sound and the sound of the new key depression, and the key number KN is increased by one, and
The sound of O (m) is converted to the KNth sound KB of the key buffer KB.
(KN). Also, when there is no semitone-related sound with KO (m), the process proceeds to step Y5 to adopt this sound and increase the key number by one.

Next, the routine proceeds to step Y6, where the counter m is incremented, and it is determined in step Y7 whether or not the maximum value j has been exceeded. If it does not exceed the maximum value, the process returns to step Y2. If it exceeds the maximum value, proceed to step Y8,
The arrangement KB is rearranged in ascending order of values (lower pitch).

That is, according to the new key press addition processing, both the currently sounding chord and the sound of the new key press are taken into account, the predetermined old key press is ignored, and the new chord is determined. Get the sound.

For example, if the key of Domiso is depressed and the chord of the C major is generated once, and then the sound of Fa is depressed, the old depressed key is ignored because Mi and Fa are in a semitone relationship, and Dofaso becomes a chord. As a configuration candidate tone, a chord of C sustain 4 is generated as can be seen from the chord table of FIG. When a key is pressed first to generate a C major chord and then a key is pressed, the key is depressed as a candidate sound. The de and shi have a semitone relationship. Is the root sound of the chord, so that it is left as it is, and the chord of the C major 7 is generated by the domiso.

By the above processing, ABC / KON of FIG.
The processing is executed. Next, KON processing and KOFF processing which are other processings included in FIG. 4 will be described.

FIG. 14 is a flowchart showing the KON process. This processing is performed when there is a key-on but the ABC mode is not selected. When the process starts, in step T21, the flag H
It is determined whether MNY is 1 or not.

When the flag HMNY is not 1, it is a normal key pressing process. In the next step T22, an empty channel is searched, and in a step T23, it is determined whether or not there is an empty channel.

If there is no empty channel, step T24
, A truncation process for terminating one of the currently sounding channels is performed, and in the next step T25, the key code KC and the timbre TC (0) are output to the channel.

If there is an empty channel in step T23, the process immediately proceeds to step T25,
The key code KC and the timbre TC (0) are output to the channel. In this way, a musical sound is generated.

When the flag HMNY is 1, the harmony mode is set, and a process of adding a sound having a predetermined relationship with the melody sound is performed. First, step T26
If there are multiple key presses, take out the highest key. That is, the melody sound is usually the highest key sound.

Subsequently, in step T27, the value of the register MD for storing the number of the harmonic mode is stored in the register n. Subsequently, in step T28, the KO for each mode is set.
N processing, that is, MDnKON processing is performed. An example of the processing for each mode will be described later.

Thus, the key-on process is executed.

FIG. 15 shows a flowchart of the MD1KON process as an example of the key sound process for each mode. It is assumed that the 0th to 4th sound channels are used for the melody sound and the additional sound. That is, it is assumed that up to four additional sounds can be generated.

When the process starts, in step Y11, sound generation stop information is output to the 0th to 4th sounding channels, and the sound currently sounding is first stopped.

Next, in step Y12, the offset information OFS (1) to OSS (4) and the threshold information T are obtained from the harmonize table according to the type of the code.
H (1) to TH (4), panning information PAN (1) to
PAN (4) and tone color information TC (1) to TC (4) are obtained. Subsequently, at step Y13, 1 is set to the counter p.

Next, in step Y14, the initial touch IT of the key depression is changed to the p-th threshold TH.
(P) It is determined whether or not it is larger. If the initial touch is larger, follow the arrow YES to step Y15
And the tone color TC (p), key code KC + OFS (p), and panning position PAN (p) are assigned to the p-th sounding channel.
Is output. In this manner, for example, a sound that is three degrees lower than the melody sound, a sound that is five degrees lower than the melody sound, and the like are generated with a predetermined tone color and localization.

Subsequently, the counter p is incremented (step Y).
16) In step Y16, it is determined whether the counter number p has reached 5 or not. If the number has not reached 5, step Y1
Returning to step 4, another additional sound generation process is performed.

If the initial touch IT is not larger than the threshold value TH, the process proceeds to step Y14.
According to the arrow O, the process bypasses step Y15 and proceeds to step Y16.

When the counter value p has reached 5, the process proceeds to step Y18, where the tone color TC (0), key code KC, and panning position 0 corresponding to the key depression of the melody tone are output to the 0th sounding channel. In other words, the melody sound due to key depression is pronounced from the center.

The case where the predetermined additional sound is generated when the initial touch IT is larger than the threshold value has been described above. However, information such as an after touch can be used instead of the initial touch. Further, the case where the accompanying sound is generated when the threshold value is larger than the threshold value has been described. However, it is also possible to generate the accompanying sound when entering a certain area. Also, Japanese Patent Laid-Open No. 2-17369
It is also possible to cause the sound to be generated in another harmony mode known in, for example, Japanese Patent Laid-Open No.

FIG. 16 shows a flowchart of the key-off process. When the key-off process starts, a step T3
At 1, it is determined whether or not the harmony mode flag HMNY is 1. If the flag HMNY is not 1, the routine proceeds to step T32, where it is determined whether there is a key-on KON corresponding to the key-off. If there is a corresponding KON, the key release information is output to the channel, and the generation of the musical sound ends (step T33). If there is no corresponding KON, the process returns immediately because the mute processing has already been performed.

When the harmony mode is selected, the program proceeds to step T31 according to the YES arrow.
34, the value of the mode register MD of the harmonic mode is stored in the register n, and the KOFF processing for each mode (MDn
OFF) (step T35).

FIG. 17 shows MD1KOFF processing as an example of the key-off processing for each mode. When the process starts,
In step Y21, sound generation stop information is output to the 0th to 4th sounding channels. In this way, the 0th to the 4th
The pronunciation by the pronunciation channel stops. Then, return.

Although the key processing steps in the main processing routine have been described above, the main processing includes panel processing and operator processing.

FIG. 18 shows a flowchart of the panel processing. As shown in FIG. 2, an ABC key 33, a harmony key 34, and tempo setting keys 32a, 32
b, keys such as a tone color selection switch group 37, an ABC style selection switch group 38, and a harmony mode selection switch group 39 are provided. Panel processing is performed according to these key processing.

When the process starts, it is determined in step S31 whether or not the ABC key has been operated.
When the ABC key is operated, the process proceeds to step S32,
Perform ABC processing. If it is not an ABC key operation, the process proceeds to step S34, and it is determined whether a harmony key is operated. When the harmony key is operated, step S
Proceed to 35 to perform harmony processing.

If the operation is not the operation of the harmonic key, the flow advances to step S36 to determine whether or not the tempo setting key is operated. If the tempo setting key has been operated, the process proceeds to step S37, where an interrupt interval is set according to the key. That is, the tempo setting key is a key 32a indicating acceleration.
And 32b indicating deceleration, and the interrupt interval is changed according to the key.

If it is not the operation of the tempo setting key, it is determined in step S38 whether or not the tone color setting key has been operated. When the timbre setting key is operated, the process proceeds to step S39, where the timbre TC (0) corresponding to the key is set.

If the operation is not the operation of the tone color setting key, the ABC
It is determined whether or not the style setting key has been operated in step S40.
Is determined. When the ABC style setting key is operated, the process proceeds to step S41, and a style corresponding to the key is set.

If it is not the operation of the ABC style setting key, the flow advances to step S42 to determine whether or not the harmony style setting key is operated. When the harmony style setting key is operated, the process proceeds to step S43, and the mode MD corresponding to the key is set. If it is not the key operation described above, other key processing is performed in step S44. For example, it corresponds to a key operation for instructing vibrato or the like.

After these processes, the panel display is updated according to the settings (step S33). Thus, panel processing is performed.

FIG. 19 is a flowchart of the ABC processing shown in step S32 of FIG. When the process starts, in step T41, the ABC flag is inverted. That is, the ABC flag alternately takes a value of 0 (off) and 1 (on). When the ABC flag is 1, the accompaniment mode is selected.

Next, in step T42, sound generation stop information is output to all sound generation channels. Subsequently, in a step T43, it is determined whether or not the flag ABC is "1".
If the flag ABC is 1, the traveling flag RUN is set to -1, and the routine returns as a standby state.

If it is determined in step T43 that the flag ABC is not 1, the routine proceeds to step T45, where the travel flag RU is set.
After N is set to 0 and the processing is performed so as not to perform the ABC processing, the routine returns.

FIG. 20 is a flowchart of the harmony process shown in step S35 of FIG. When the process starts, in step T46, the flag HMNY is inverted. This flag HMNY also alternates between 0 (off) and 1 (on). Next, in step T47,
Outputs sound stop information to all sound channels. Then return.

Next, referring to FIG. 25, step S2 in FIG.
The operation process shown in FIG.

When the operation process starts, step S
At 51, it is determined whether there is a foot switch on event. If there is a foot switch on event, the process proceeds to step S52 according to the YES arrow, and 1 is stored in the flag FS to indicate that the foot switch is on. When there is no foot switch on event,
Proceeding to step S53, 0 is stored in the flag FS, indicating that the foot switch is off.

The main processing shown in FIG. 3 has been described above. Here, an interrupt processing routine performed separately from the main processing routine will be described.

In FIG. 21, when the interrupt process starts, in step U11, the interrupt is first prohibited. Subsequently, at step U12, the travel flag RU
It is determined whether N is -1 indicating that the accompaniment mode is waiting.

If the traveling flag RUN is -1, YES
The process proceeds to step U13 in accordance with the arrow of (3) to obtain the data of the position of the timer count TCNT from the basic accompaniment pattern corresponding to the type of chord and the style of the accompaniment (Latin, jazz, etc.). FIG. 24 shows an example of the data of the position of this TCNT.
Shown in

Next, in step U14, it is determined whether or not the timer count TCNT is 0. The timer count takes a value from 0 to 31, for example, and divides one bar into 32.

When TCNT is 0, the routine proceeds to step U15, where it is determined whether or not the data is the key-on of the currently sounding data. That is, it is a case where the same sound is continuously produced but a key code is given since the beginning of the measure. If yes, step U19 without sounding again
Proceed to.

When the harmony flag HMNY is not 1,
Steps U20 and U21 are bypassed. When the determination in step U14 and step U15 is NO, the process first proceeds to step U16 to make the normal sound generation, and it is determined whether or not the operator (foot switch) flag FS is 1 or not. If the control flag FS is not 1, the process proceeds to step U18, where the data is processed and sounded according to the type of chord, the root sound, and the part. When the flag FS is 1, the process proceeds to step U17, where the data is processed according to the type of chord, the root sound, and the part, and the specific sound is changed with reference to a table as shown in FIG. I do. Thereafter, the flow merges with Step U19.

At Step U19, the harmony flag HM
It is determined whether NY is 1 or not. If the harmony flag HMNY is 1, the process proceeds to step U20, and the mode register M
The value of D is stored in the register N, and a clock routine for each mode is performed in the next step U21.

The mode-specific clock routine is independently executed according to each mode as shown in FIG. 22, for example.

Thereafter, the process proceeds to step U22, in which the value of the timer count TCNT is incremented. In step U23, it is determined whether or not the timer count TCNT has exceeded 31. T
When CNT exceeds 31, TCNT is reset to 0 (step U24), the interrupt is re-enabled (step U25), and the interrupt returns.

If TCNT does not exceed 31, interrupts are immediately re-enabled (step U25).
Return from interrupt. If the running flag RUN is not -1 in step U12, the interrupt is immediately re-permitted and the process returns to the interrupt.

For data processing in accordance with the type of chord, root sound, and part, a known technique can be employed.

FIG. 23 shows an example of the code table. This table has a column indicating the number K and a code type CT.
(K) and a chord pattern CPT serving as a detection rule to be compared with a chord component sound candidate detected based on a key press or the like.
An N (k) column and a reference pattern column are included. For example,
When the detected chord component is 1-3-5 in the notation of the 12th scale, it corresponds to k = 0 of CPTN (k) of the detection rule, the chord type is M, and the reference pattern is major.
Here, according to this table, the reference patterns are divided into three: major, minor, and 7th (seventh).

FIG. 24 shows an accompaniment track of a basic pattern set according to these reference patterns. The timer count TCNT is changed according to each of the major M, the minor m, and the seventh 7th from the pattern 4 from t0 to t31.
2, 44 and 46 are set. Accompaniment is performed according to these.

Although the case where the reference pattern is divided into M, m, and 7th has been described, one accompaniment track may be employed without being divided. Further, the number of accompaniment tracks may be arbitrarily set. Furthermore, accompaniment tracks may be separately prepared according to the style of waltz, pops, jazz big band, jazz quartet, classical music, Japanese folk songs, and the like.

FIG. 26 shows a specific sound conversion table.

For example, when C7th is selected as a code, ABC is turned on when the foot switch FS is turned on.
All 9th (sound D in C tone) in the pattern is converted to Db, and all 13th (sound A in C tone) is converted to Ab. This is one of the automatic accompaniment for high level users.
It is a form. When the pedal is off, 9th, 13th and the like in the ABC pattern are output as they are.

The present invention has been described in connection with the preferred embodiments.
The present invention is not limited to these. For example,
As the automatic accompaniment mode, any mode other than the above-mentioned mode can be adopted. In addition, it is possible to set a desired accompaniment form for a user having advanced musical knowledge.

[0179]

As described above, according to the present invention,
The required pitch information input operation can be simplified and a desired chord can be generated.

Further, it is possible to provide an electronic musical instrument in which when unnecessary pitch information is input by mistake, the input is ignored.

[Brief description of the drawings]

FIG. 1 is a block diagram of the overall configuration.

FIG. 2 is a schematic plan view of an operation panel.

FIG. 3 is a flowchart of a main processing routine.

FIG. 4 is a flowchart of a key processing routine.

FIG. 5 is a flowchart of ABC / KON processing.

FIG. 6 is a flowchart of a code detection process.

FIG. 7 is a flowchart of a third-degree key press search process.

FIG. 8 is a flowchart of a second key press search process.

FIG. 9 is a flowchart of a seventh-degree key press search process.

FIG. 10 is a flowchart of a fifth-degree key press search process.

FIG. 11 is a flowchart of a key buffer search process.

FIG. 12 is a flowchart of a code establishment determination process.

FIG. 13 is a flowchart of a new key press addition process.

FIG. 14 is a flowchart of a KON process.

FIG. 15 is a flowchart of MD1KON processing.

FIG. 16 is a flowchart of a KOFF process.

FIG. 17 is a flowchart of MD1KOFF processing.

FIG. 18 is a flowchart of panel processing.

FIG. 19 is a flowchart of an ABC process.

FIG. 20 is a flowchart of a harmony process.

FIG. 21 is a flowchart of an interrupt processing routine.

FIG. 22 is a flowchart of MD1CLK processing.

FIG. 23 is an example of a code table.

FIG. 24 is a conceptual diagram of an accompaniment track of a basic pattern.

FIG. 25 is a flowchart of an operator processing routine.

FIG. 26 is an example of a specific sound conversion table.

[Explanation of symbols]

11 keyboard, 12 keyboard interface, 13 bus, 14 operator, 15 operator interface, 1
6 CPU, 17 ROM, 18 RAM, 21 operation panel, 22 panel interface, 23 sound source,
24 ... Sound system, 25, 26 ... Speaker, 31
... Power key, 32 ... Tempo key, 33 ... ABC key, 3
4: Harmonic key, 35: Display, 37: Tone selection switch group, 38: ABC style selection switch group, 39:
Harmony mode selection switches

Claims (3)

(57) [Claims] 1. Pitch information input means for inputting pitch information; first chord determining means for determining a root and a type of a chord based on a combination of pitch information input by the pitch information input means; Pitch input determining means for determining whether or not new pitch information has been input when the first chord determining means has failed to determine the root and type of a chord; When it is determined by the means that new pitch information is input, a second chord is determined based on a combination of the previously input pitch information and the newly input pitch information. Chord determination means; pattern storage means for storing time information and an accompaniment pattern composed of accompaniment sounds associated with the time information; root and type of chord determined by the first or second chord determination means Based on the The accompaniment sound Kanade pattern is converted into an accompaniment tone information, the electronic musical instrument, characterized in that an output means for outputting the converted accompaniment tone information based on the time information of the accompaniment pattern. 2. The electronic musical instrument according to claim 1, wherein said second chord discriminating means determines whether the previously inputted pitch information and the newly inputted pitch information have a semitone interval. An electronic musical instrument that gives priority to newly input pitch information unless the previously input pitch information is a root sound. 3. The electronic musical instrument according to claim 1, wherein a root sound is used as a base sound when a lowest sound of the inputted pitch information is not a constituent sound of a chord. An electronic musical instrument having means for using the lowest note as a bass sound when the lowest note is a constituent sound of a chord.