JP2013097302A - Automatic tone correction device, automatic tone correction method and program therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic tone correction device by which the results of tone determination that is performed for an input melody in real time can be determined in non-real time for a correction to more accurate tone, so that coding accuracy can be improved.SOLUTION: An automatic tone correction device 10 includes a key board 11 that performs a melody of a piece of music, and a CPU 21 performs, by the history of pitch of melody being performed, tone determination for the melody in real time. And after the melody performance finishes, the tone determination results performed in real time is corrected in non-real time by the CPU 21.

Description

  The present invention relates to an automatic tone correction device, an automatic tone correction method, and a program thereof.

  In an electronic musical instrument having a keyboard such as an electronic piano or an electronic organ, it is common to play a melody mainly with the right hand and an accompaniment with the left hand, or press a plurality of keys constituting a chord. Therefore, in such an electronic musical instrument, it is necessary to practice to move the right hand and the left hand independently according to a musical score or the like.

  As described above, in both the piano playing method and the organ playing method, it is necessary to move the right hand and the left hand simultaneously in different forms. For this purpose, appropriate practice is required. In particular, it is possible to move the right hand to play a melody, but there are many performers who find it difficult to perform different performances with the left hand at the same time, especially for beginners. Therefore, there is a need for an electronic musical instrument that can automatically create an accompaniment sound to be generated by playing a melody with the right hand by the performer.

  For example, in Patent Document 1, musical note data of music is stored for each of a plurality of sections, and when the chord name of the second section of the note data is given, the key data and the musical note data corresponding to the second section An apparatus for determining a new chord name by referring to the note data of the first section and the chord name previously given to the second section has been proposed.

  As described above, notes constituting the music are used as one of the elements for determining the chord, but each of these notes is not reflected in the chord determination evenly. For example, the weight varies depending on which beat the note is included in, and the weight varies depending on the temporal position within the beat. Therefore, it is desirable to determine the code name reflecting the weight. It is more desirable to determine the chord name according to the transition of a plurality of notes rather than referencing only a single note.

  Therefore, Patent Document 2 proposes an automatic accompaniment apparatus that can appropriately determine a chord name in real time based on the weights of notes constituting the music and the transitions thereof.

Japanese Patent No. 3099436 JP 2011-158855 A

However, if you code in real time, it will be decided to some extent while using a database created from rules of thumb, even if simple songs can be accurately coded in real time, but decorative sounds and borrowed chords are used In some cases, the accuracy of real-time coding is low.
One reason for the low accuracy of coding is that the accuracy of music key (key) determination, which is required for coding, is low.
The tone of a song is not always constant. Depending on the music, there are many cases in which a transition to another key occurs in the middle. When a modulation occurs, it is necessary to determine which key has been changed and to acquire a new key prior to the determination of the assigned code name.
Usually, the key of a music is often determined using a history of pitches of the melody constituting the music. However, when trying to determine the key from the melody of a song that is played in real time, there are few pitches included in the history at the start of the song / phrase, etc. It could be difficult.
Thus, in order to improve the accuracy of real-time coding, it is necessary to make the key determined in real time more accurate.

  An object of the present invention is to improve the accuracy of chords assigned to music in real time by further improving the accuracy of the key determination result of the music played in real time.

In order to achieve the above object, an automatic adjustment device according to one aspect of the present invention is provided.
A performance means for playing a song melody,
Real-time key determination means for determining the key of the song melody in real time based on the pitch history of the song melody being played,
A non-real-time tone correcting means for correcting the key determined by the real-time key determining means in non-real time after the melody performance by the playing means is completed;
Have
Moreover, the automatic tone correction method of one aspect of the present invention includes:
Determining in real time the song melody based on the pitch history of the song melody being played;
Correcting the determined key in non-real time after the melody performance is completed;
Have
Furthermore, a program according to one embodiment of the present invention includes:
Determining the key of the song melody in real time based on the pitch history of the song melody being played;
Correcting the determined key in a non-real time after the music melody performance is completed;
Causes the computer to execute.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the automatic tone correction apparatus which can raise the precision of coding more.

FIG. 1 is a diagram illustrating an appearance of an automatic tone correction device according to the present embodiment. FIG. 2 is a block diagram showing the configuration of the automatic adjustment device according to the embodiment of the present invention. FIG. 3 is a flowchart showing an example of a main flow executed in the automatic adjustment device according to the present embodiment. FIG. 4 is a flowchart showing in more detail an example of keyboard processing according to the present embodiment. FIG. 5 is a flowchart illustrating an example of the code name determination process according to the present embodiment. FIG. 6 is a flowchart illustrating an example of the first-beat / third-beat correspondence determination processing according to the present embodiment. FIG. 7 is a flowchart showing an example of the first beat / third beat corresponding note determination process according to the present embodiment. FIG. 8 is a flowchart showing an example of the first-beat / third-beat-corresponding note determination processing according to the present embodiment. FIG. 9 is a flowchart showing an example of the first-beat / third-beat correspondence note determination processing according to the present embodiment. FIG. 10 is a flowchart showing an example of the first dominant motion determination process according to the present embodiment. FIG. 11 is a flowchart showing an example of the second beat corresponding note determination process according to the present embodiment. FIG. 12 is a flowchart showing an example of the second beat-corresponding note determination process according to the present embodiment. FIG. 13 is a flowchart showing an example of the second beat corresponding note determination process according to the present embodiment. FIG. 14 is a flowchart showing an example of the second beat corresponding note determination process according to the present embodiment. FIG. 15 is a flowchart showing an example of the fourth beat corresponding note determination process according to the present embodiment. FIG. 16 is a flowchart showing an example of the fourth beat corresponding note determination process according to the present embodiment. FIG. 17 is a flowchart illustrating an example of the second dominant motion determination process according to the present embodiment. FIG. 18 is a flowchart illustrating an example of the code determination process according to the present embodiment. FIG. 19 is a flowchart illustrating an example of code determination processing according to the present embodiment. FIG. 20 is a flowchart illustrating an example of code determination processing according to the present embodiment. FIG. 21 is a flowchart illustrating an example of code determination processing according to the present embodiment. FIG. 22 is a diagram showing an example of a melody sequence table according to the present embodiment. FIG. 23 is a diagram illustrating an example of a first code table according to the present embodiment. FIG. 24 is a diagram illustrating an example of a second code table according to the present embodiment. FIG. 25 is a diagram showing an example of a melody function table according to the present embodiment. FIG. 26 is a diagram showing an example of a non-determination code table according to the present embodiment. FIG. 27 is a flowchart showing an example of non-real time code correction processing according to the present embodiment. FIG. 28 is a flowchart illustrating an example of the adjustment correction verification process according to the present embodiment. FIG. 29 is a flowchart illustrating an example of the adjustment correction verification process according to the present embodiment. FIG. 30 is a flowchart showing an example of the chord composition sound verification process according to the present embodiment. FIG. 31 is a flowchart showing an example of phrase break verification processing according to the present embodiment. FIG. 32 is a flowchart showing an example of phrase break verification processing according to the present embodiment. FIG. 33 is a flowchart illustrating an example of phrase break verification processing according to the present embodiment. FIG. 34 is a flowchart showing an example of phrase break verification processing according to the present embodiment. FIG. 35 is a flowchart illustrating an example of the coding correction process according to the present embodiment. FIG. 36 is a flowchart illustrating an example of the coding correction process according to the present embodiment. FIG. 37 is a flowchart showing an example of comparison processing between a melody and a chord scale according to the present embodiment. FIG. 38 is a flowchart showing an example of correction processing by the user according to the present embodiment. FIG. 39 is a diagram showing an example of a key confirmation table based on the final melody sound according to the present embodiment. FIG. 40 is a diagram illustrating an example of a cadence determination table according to the present embodiment. FIG. 41 is a diagram illustrating an example of a relational arbitration determination table according to the present embodiment. FIG. 42 is a diagram showing an example of a decorative sound database according to the present embodiment. FIG. 43 is a diagram illustrating an example of a code scale table according to the present embodiment. FIG. 44 is a diagram showing an example of a code priority database according to the present embodiment. FIG. 45 is a flowchart showing an example of automatic accompaniment processing according to the present embodiment. FIG. 46 shows an example of a score.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a diagram showing an external appearance of an electronic musical instrument to which the automatic tone correction device according to the present embodiment is applied. As shown in FIG. 1, the electronic musical instrument 10 according to the present embodiment has a keyboard 11. Also, on the upper part of the keyboard 11, there are switches (see reference numerals 12 and 13) for specifying a timbre, starting and ending automatic accompaniment, specifying a rhythm pattern, etc., and non-real time after the real time coding is finished. A non-real-time coding correction switch 14 for starting chord correction at, and a display unit 15 for displaying various information relating to the music to be played, for example, tone color, rhythm pattern, chord name, and the like. The electronic musical instrument 10 according to the present embodiment has, for example, 61 keys (C2 to C7). In addition, the electronic musical instrument 10 can perform under one of two performance modes: an automatic accompaniment mode for turning on automatic accompaniment and a normal mode for turning off automatic accompaniment.

  FIG. 2 is a block diagram showing the configuration of the automatic adjustment device according to the embodiment of the present invention. As shown in FIG. 2, the automatic adjustment device 10 according to the present embodiment includes a CPU 21, a ROM 22, a RAM 23, a sound system 24, a switch group 25, a keyboard 11, and a display unit 15.

  The CPU 21 controls the entire electronic musical instrument 10, detects keys pressed on the keyboard 11, detects operations of switches constituting the switch group 25 (for example, reference numerals 12, 13 and 14 in FIG. 1), and operates keys and switches. Accordingly, various processes such as control of the sound system 24, determination of the chord name according to the pitch of the depressed key, automatic accompaniment according to the automatic accompaniment pattern and chord name, and the like are executed.

  The ROM 22 performs various processes to be executed by the CPU 21, such as switch operation, key depression of any key on the keyboard, tone generation according to the key depression, and chord names according to the pitch of the depressed musical tone. Programs such as determination, automatic accompaniment pattern and automatic accompaniment performance according to chord names are stored. The ROM 22 stores a waveform data area for storing waveform data for generating musical sounds such as piano, guitar, bass drum, snare drum, and cymbal, and data (automatic accompaniment data) indicating various automatic accompaniment patterns. Automatic accompaniment pattern area. The RAM 23 stores a program read from the ROM 22 and data generated in the course of processing. In the present embodiment, the automatic accompaniment pattern has a chord pattern including constituent sounds for each chord name, a bass pattern including bass sounds, and a rhythm pattern including drum sounds. For example, the chord pattern and base pattern data records include the pitch of the musical tone, the sounding timing (sounding time), and the sound length. The rhythm pattern data includes the tone color of the musical tone and the sounding timing.

  The sound system 24 includes a sound source unit 26, an audio circuit 27, and a speaker 28. For example, when the sound source unit 26 receives information about the depressed key or information about the automatic accompaniment pattern from the CPU 21, the sound source unit 26 reads predetermined waveform data from the waveform data area of the ROM 22, and generates musical tone data of a predetermined pitch. Generate and output. The sound source unit 26 can also output waveform data, particularly waveform data of percussion instrument sounds such as a snare drum, bass drum, and cymbal, as musical sound data. The audio circuit 27 D / A converts and amplifies the musical sound data. Thereby, an acoustic signal is output from the speaker 28.

  The electronic musical instrument 10 according to the present embodiment generates a musical tone based on the key depression of the keyboard 11 under the normal mode. On the other hand, the electronic musical instrument 10 enters an automatic accompaniment mode when an automatic accompaniment switch (not shown) is operated. Under the automatic accompaniment mode, a musical tone having the pitch of the key is generated by pressing the key. Further, a chord name is determined based on the pitch of the depressed key, and a musical tone is generated according to an automatic accompaniment pattern including chord constituent sounds of the chord name. Hereinafter, the case where the electronic musical instrument 10 operates in the automatic accompaniment mode will be described.

  Hereinafter, the process executed in the electronic musical instrument 10 according to the present embodiment will be described in more detail. FIG. 3 is a flowchart showing an example of a main flow executed in the automatic adjustment device according to the present embodiment. Although not shown, timer increment processing for incrementing the counter value of the interrupt counter is also executed at predetermined time intervals during execution of the main flow.

  As shown in FIG. 3, when the power is turned on, the CPU 21 of the electronic musical instrument 10 executes initial processing (initialization processing) including clearing of the data in the RAM 23 and the image on the display unit 15 (step S1). . When the initial process (step S1) ends, the CPU 21 detects each operation of the switches constituting the switch group 25, and executes a switch process that executes a process according to the detected operation (step S2).

  For example, in the switch process (step S2), various switch operations such as a tone color designation switch, an automatic accompaniment pattern type designation switch, and an automatic accompaniment pattern on / off designation switch are detected. When the automatic accompaniment pattern is turned on, the CPU 21 switches the performance mode to the automatic accompaniment mode. Data indicating the performance mode is specified in a predetermined area of the RAM 23. Similarly, data indicating the type of timbre and automatic accompaniment pattern is also stored in a predetermined area of the RAM 23.

  Next, the CPU 21 executes keyboard processing (step S3). FIG. 4 is a flowchart showing in more detail an example of keyboard processing according to the present embodiment. In the keyboard process, the CPU 21 sequentially scans the keys on the keyboard 11 (step S11). The CPU 21 refers to the scanning result and determines whether an event has occurred for a certain key (step S12). When it is determined Yes in step S12, the CPU 11 determines whether or not the generated event is key-on (key depression) (step S13).

  When it is determined Yes in step S13, the CPU 21 executes a sound generation process corresponding to the key that has been turned on (step S14). In this sound generation process, the CPU 21 reads out the tone color data for the melody key stored in the RAM 23 and the data indicating the pitch corresponding to the key, and temporarily stores them in the RAM 23. In a sound source sound generation process (step S6 in FIG. 3), which will be described later, data indicating the tone color and pitch stored temporarily are given to the sound source unit 26. The sound source unit 26 generates musical tone data by reading waveform data from the ROM 22 in accordance with the data indicating the timbre and pitch. Thereby, a predetermined musical sound is generated from the speaker 28.

  Thereafter, the CPU 21 stores pitch information (for example, key number) and key pressing timing (for example, key pressing time) for the key that has been turned on in the RAM 23 (step S15). The key pressing timing can be calculated based on the counter value of the interrupt counter.

  If it is determined No in step S13, the generated event is key-off (key release). Therefore, the CPU 21 executes a mute process for the key that has been turned off (step S16). In the mute process, the CPU 21 extracts data indicating the pitch included in the generated event. Also in this case, in the sound source sound generation process (step S6) described later, data indicating the extracted pitch is given to the sound source unit 26. The tone generator 26 mutes the musical tone having the pitch based on the given data. Thereafter, the CPU 21 stores the key release time (key release time) in the RAM 23 for the key that has been key-off (step S17).

  The CPU 21 determines whether the processing has been completed for all key events (step S18). If it is determined No in step S18, the process returns to step S12.

When the keyboard process (step S3 in FIG. 3) is completed, the CPU 21 executes a code determination process (step S4). FIG. 5 is a flowchart illustrating an example of the code determination process according to the present embodiment. In the present embodiment, generally, a melody sound that is currently sounded is a current melody sound (Current Melody) CM, and a melody sound that is sounded immediately before that is a preceding melody sound (Previous Melody). ) PM, the chord name that was played immediately before the current chord name (Previous Chord name) PreCH is newly pronounced based on the current melody tone CM, preceding melody tone PM, and preceding chord name PreCH The current code name (Current Chord name) CurCH is determined. In the present embodiment, the tonality of the music is expressed in C major (CMaj) or A minor (Amin), the chord name is expressed in degrees with respect to the main sound, such as IMaj, IIm, and the data is stored in the RAM 23 or the like. doing. In the case of another tonality, it is possible to obtain a chord name accompanied by the root sound based on the pitch difference between the root sound of the tonality and the sound of “C” (or “A”). .
Note that the tonality of the music is not limited to the case where it is predetermined as C major (CMaj) or A minor (Amin), and may be determined in real time including other majors and minors. In this case, one or more key types are determined. Further, the determined key is stored in the RAM 23.

  However, the time position at which the key was pressed, such as at what beat the key was pressed, whether the key was the beginning of the beat, or the sound type by multiple key presses (continuous, progressive ), The current melody sound CM and the preceding melody sound PM may be determined. That is, when a key other than the key currently pressed is to be the current melody sound CM, or when a key other than the key pressed immediately before the key currently pressed is the preceding melody sound There are cases where it should be PM. Hereinafter, in the chord determination process, steps S24 to 30 mainly relate to determination of the current melody sound CM and the preceding melody sound PM. The subsequent step S31 is a process for specifically determining the current chord name CurCH based on the current melody sound CM, the preceding melody sound PM, and the preceding chord name PreCH.

  First, the CPU 21 specifies beat information and key press information (key-on time and time until key-off) to which the current time belongs, specifies the key pressed in the current beat, and sets the current time The information on the key that was pressed in the previous section (previous beat section) of the beat to which the key belongs is acquired (step S21). In step S21, the information on the key pressed at the current beat becomes the initial value of the current melody sound CM, and the information on the key pressed at the head of the previous beat section becomes the initial value of the preceding melody sound PM.

  Next, the CPU 21 determines whether there is a key being pressed at the head of the beat to which the current time belongs based on the beat information and the key press information (step S22). If it is determined No in step S23, the code name determination process ends. If it is determined Yes in step S22, the CPU 21 copies the current code name CurCH to the preceding code name PreCH (step S23).

  The CPU 21 sets table designation information for designating a code table, which will be described later, to information for designating the second code table (step S24). The code table mainly includes a first code table used when the key is pressed at the first beat and a second code table used at other times, and each is stored in the ROM 22. The table designation information is information indicating which of the first code table or the second code table is used, and is stored in the RAM 23.

  Next, the time position of the pressed key, that is, the beat at which the key is pressed is determined (steps S25, 26, and 28). When the key press is the first beat (Yes in step S25) or the key press is the third beat (Yes in step S26), the CPU 21 executes the first beat / third beat corresponding note determination process. (Step S27). If the pressed key is the second beat (Yes in step S28), the second beat corresponding note determination process is executed (step S29). If NO is determined in step S508, that is, if the depressed key is the fourth beat, the fourth beat corresponding note determination process is executed (step S30).

  In the present embodiment, the music has a four-quarter beat, and one measure is composed of four beats. The key depression being the nth beat means that the timing of the key depression is temporally earlier than the beginning of the (n + 1) th beat after the beginning of the nth beat.

  The components of music have the concept of time and beat. Further, in the time signature, there is a weight for each beat, and the melody progresses in consideration of the weight of the beat. In some cases, the weight of the beat may move, such as syncopation. In the present embodiment, taking into account the weight of the beat, constituent sounds constituting an optimal melody flow are extracted, and the current melody sound CM and preceding melody sound PM that are optimal for use in chord determination are specified.

  6 to 9 are flowcharts showing examples of the first-beat / third-beat-corresponding note determination processing according to the present embodiment. As shown in FIG. 6, it is determined whether the key depression is related to the first beat (step S41). When it is determined Yes in step S24, the CPU 21 changes the table designation information to information indicating the first table (step S42). Next, the CPU 12 executes a first dominant motion determination process (step S43).

  The dominant motion determination process is to extract a dominant motion (that is, progression from a genus chord (dominant) to a main chord (tonic)) from the melody flow. In the present embodiment, a first dominant motion determination process in which the code name is considered in the process and a second dominant motion determination process in which the code name is not considered are used. FIG. 10 is a flowchart showing an example of the first dominant motion determination process according to the present embodiment.

  As shown in FIG. 10, the CPU 21 determines whether the preceding code name PreCH stored in the RAM 23 corresponds to one of the major dominant codes (step S81). Here, in the first dominant motion determination process according to the present embodiment, for example, the major dominant codes are “VMaj”, “V7”, and “VIIm7 (−5)”. When it is determined Yes in step S81, the CPU 21 determines that (PM, CM), which is a set of values of the preceding melody sound PM and the current melody sound CM, is (F, E), (B, C), In step S82, the movement from the preceding melody sound PM to the current melody sound CM is resolved from dominant to tonic with major chord progression. It is judged whether or not it is a movement.

  If it is determined Yes in step S82, the CPU 21 determines the current code name CurCH to be “IMaj” and stores the information in the RAM 23 (step S83). Thereafter, the CPU 21 stores information indicating that it is the first determination result in the dominant motion processing in the RAM 23 (step S84). If it is determined No in step S81 or No in step S82, the CPU 21 determines whether the preceding code name PreCH corresponds to any of the minor dominant codes (step S85). In this embodiment, for example, the minor dominant codes are “IIIMaj” and “III7”.

  When it is determined Yes in step S85, the CPU 21 determines whether (PM, CM) is any one of (G #, A), (B, A), and (D, C) ( Step S86). In step S86, it is determined whether or not the movement from the preceding melody sound PM to the current melody sound CM is a movement when resolving from dominant to tonic by progression of minor chords. When it is determined Yes in step S86, the CPU 21 determines the current code name CurCH to “VImin” and stores the information in the RAM 23 (step S87). Thereafter, the process proceeds to step S84.

  If it is determined No in step S85, or if it is determined No in step S86, the CPU 21 stores information indicating that the result of the second determination in the dominant motion process is in the RAM 23 (step S85). S88).

  When the first dominant motion determination process in step S43 ends, the CPU 21 determines whether or not the result of the first dominant motion determination process is the second process result (step S44). If it is determined No in step S44, that is, if the result of the first dominant motion determination process is the first process result, the preceding melody sound PM and the current melody sound CM are changed from the initial values. Without being stored in the RAM 23, the process is terminated (step S45). When it is determined Yes in step S44, the CPU 21 refers to the key pressing information stored in the RAM 23 and determines whether there is a key pressing immediately before the beat corresponding to the current time (step S46).

  When it is determined Yes in step S44, the CPU 21 refers to the key pressing information in the RAM 23 and determines whether there is a key pressing after the head of the immediately preceding beat (step S47). If it is determined No in step S47, it means that a quarter note has been pressed on the previous beat and the current beat. Processing in this case will be described later. If it is determined Yes in step S47, the CPU 21 refers to the sound generation time of the key pressed after the head of the previous beat in the key press information stored in the RAM 23, and has it been sounded so far? Is determined (step S49). In step S49, even if the key is not depressed at the beginning of the beat, if it is depressed and maintained after the beginning of the previous beat, it corresponds to the syncopation, so it is determined whether there is a pressed key corresponding to the syncopation. doing.

  If it is determined Yes in step S49, the preceding melody sound PM remains the initial value, while the key pressed after the head of the immediately preceding beat is set as the current melody sound CM, and the syncopation flag SYN is set to “1”. And stored in the RAM 23 (step S50). The preceding melody sound PM and the current melody sound CM are stored in the RAM 23. Since the key depression corresponding to the syncopation has the same weight as the key depression at the beginning of the beat, it is handled in the same manner as the beginning of the beat.

  Next, with reference to FIG. 7, the case where it is No in step S46 will be described. When it is determined No in step S46, the CPU 21 determines whether or not the key depression at the beginning of the current beat corresponds to the music start sound (step S51). Step S51 can be realized by the CPU 21 referring to the key pressing information stored in the AM 23 to determine whether it is the first key pressing information. If it is determined Yes in step S51, the current melody sound CM is given as the preceding melody sound PM. Further, although the current melody sound CM is not changed from the initial value, the CPU 21 changes the table designation information to information for designating the second code table (step S52).

  If it is determined No in step S51, it is determined whether there is no key depression for a time of 8 beats or more (step S53). If it is determined Yes in step S53, the CPU 21 maintains the current melody sound CM as the initial value, while giving the current melody sound CM as the preceding melody sound PM and stores the information in the RAM 23 ( Step S54). When it is determined Yes in step S53, there is no new key pressing for two bars or more. In this case, since the meaning of the melody sequence becomes faint, the initial value of the preceding melody sound PM that is pressed two or more bars before is ignored. When it is determined No in step S53, the CPU 21 maintains the initial melody sound PM and the current melody sound CM as initial values (step S55).

  Next, the case where No is determined in step S47 will be described with reference to FIG. When it is determined No in step S47, the CPU 21 determines whether the current melody function (CurrentMelodyFunction: CMF) is a non-harmonic sound (OtherTone: OT) (step S61). Here, the current melody function CMF indicates the function of the current melody sound CM for the preceding chord name PreCH. In the present embodiment, the CMF indicates that the current melody tone CM is a tone of chord tone (ChordTone: CT) indicating that it is a chord tone of the preceding chord name PreCH, and a tone of the current scale (tonality). A scale note (ScaleNote: CN) to be shown, or an other tone (Other Tone) to show another sound (non-harmonic sound).

  More specifically, a melody function table that associates chord names with pitch names is stored in the ROM 22, and the CPU 21 refers to the values associated with the set of the current melody tone CM and the preceding chord name PreCH. Currently judging the melody function. FIG. 25 is a diagram showing an example of a melody function table according to the present embodiment. As shown in FIG. 25, in the melody function table 2500, a predetermined value can be acquired corresponding to a set of values including the current melody sound CM and the preceding chord name PreCH. 25, in the melody function table 2500, CT indicates a chord tone (for example, reference numerals 2501 to 2503), and SN indicates a scale note (for example, reference numerals 2511 to 2513). In FIG. 25, a column in which nothing is written in the melody function table 2500 (for example, see reference numerals 2521 and 2522) indicates an other tone.

  When it is determined Yes in step S61, the CPU 21 maintains the preceding melody sound PM and the current melody sound CM as initial values (step S62). On the other hand, when it is determined No in step S61, the CPU 21 determines whether or not the current melody function CMF is the scale note SN (step S63). When it is determined Yes in step S63, the CPU 21 determines whether the difference between the preceding melody sound PM and the current melody sound CM is within two semitones (step S64). In step S64, it is determined whether the progress is so-called sequential. If it is determined Yes in step S64, or if it is determined No in step S63, the first dominant motion process is executed (step S65).

  Next, the CPU 21 determines whether or not it is the second determination result in the first dominant motion determination process (step S65) (step S66). When it is determined No in step S66, that is, when the result is the first determination result, the CPU 21 maintains the initial melody sound PM and the current melody sound CM as initial values (step S62). When it is determined Yes in step S66, the CPU 21 determines whether or not the current key press, that is, the key press to be processed is the first beat (step S63). When it is determined Yes in step S63, the CPU 21 maintains the preceding melody sound PM and the current melody sound CM as initial values (step S68).

  On the other hand, if it is determined No in step S67, that is, if the key to be processed is the third beat, the CPU 21 uses the initial preceding melody sound PM as the preceding melody sound PM. Further, the pitch PPM that was previously pressed is given (step S69). Note that the initial value of the current melody sound CM is maintained. This is because it is highly possible that the musical sound that sequentially progresses at the third beat is a decorative sound, and it is appropriate that the original musical sound that dominates the melody line is the musical sound one beat before that. Because it is considered.

  Next, the case where it is determined No in step S49 will be described. As shown in FIG. 9, the CPU 21 specifies the key pressing sound after the beginning of the previous beat (step S71), and determines whether the pitch of the specified key pressing sound is the same as the initial CM. (Step S72). If it is determined Yes in step S72, the CPU 21 maintains the current melody sound CM in the initial state, while giving the value of the current melody sound CM as the preceding melody sound PM (step S73). For example, consider a case where the melody progresses in the order of “D” and “C” in the eighth note in the immediately preceding beat, and the key depression at the current beat is “C”. In this case, the first “D” sound of the immediately preceding beat is considered to be a decoration sound, and the sequence “C” to “C” is appropriate instead of the sequence “D” to “C”. thinking. For this reason, the preceding melody sound is made the same as the current melody sound, and the same sound is continuous.

  When it is determined No in step S72, the CPU 21 determines whether or not all the key pressing sounds after the specified beat are equal to the preceding melody sound PM (step S74). If it is determined Yes in step S74, the process proceeds to step S63. For example, it is considered that four key presses of “D”, “C”, “C”, and “C” were made with a sixteenth note in the last beat. In this case, since “D” may be a decorative sound just because it is the beginning of a beat, “D”, which is the key to press the beginning of the beat, should not be the leading melody sound PM. There may be no. Therefore, the processing after step S63 is executed.

  On the other hand, when it is determined No in step S74, the CPU 21 maintains the initial melody sound PM and the current melody sound CM as initial values (step S75).

  Next, the second beat corresponding note determination process (step S29) will be described. FIGS. 11-14 is a flowchart which shows the example of the 2nd beat corresponding | compatible note determination process concerning this Embodiment. As shown in FIG. 11, the CPU 21 determines whether there is no key depression at the first beat (step S91). If it is determined Yes in step S91, that is, if there is no key depression, the CPU 21 changes the table designation information to information for designating the first code table (step S91). For example, when the sound of the previous measure is extended in the middle of a song, the first beat is rest and the next phrase starts from the second beat, in this embodiment, the key is depressed at the second beat. Considering that the sound has the same weight as the first beat, the first chord table that is the chord table for the first beat is used.

  In the second beat corresponding note processing, the first dominant motion determination processing (step S43) and the processing based on the determination result (steps S44 and 45) of the first beat / third beat corresponding note determination processing are omitted. Steps S93 to S97 are the same as steps S46 to S50 in FIG.

  FIG. 12 is a process executed when No is determined in step S93. Steps S101 and S103 to 105 are the same as steps S51 and S53 to 55 in FIG. 7, respectively. Step S102 is the same as step S52 in FIG. 7 except that the table designation information is not changed.

  Next, the case where it is determined No in step S94 will be described. As shown in FIG. 13, the CPU 21 determines whether or not the current melody function CMF is a non-harmonic sound OT (step S111). Step S112 that proceeds to the case of Yes in step S111 and step S111 is the same as steps S61 and 62 in FIG.

  When it is determined No in step S111, the CPU 21 determines whether the preceding code name PreCH is a code other than the non-determined code (step S113). As described in FIG. 21 (particularly step S205), for the non-determined code, the modulation flag has a value of “1” or more in the previous process. Therefore, in step S113, it may be determined whether the modulation flag stored in the RAM 23 is “1” or more.

  If No in step S113, that is, if the preceding code name PreCH is a non-determined code, the CPU 21 gives the current melody sound CM as the preceding melody tone PM (step S114). On the other hand, if it is determined Yes in step S113, that is, if the preceding chord name PreCH is other than a non-determined code, it is determined whether the current melody function CMF is the scale note SN (step S115). If it is determined Yes in step S63, the CPU 21 determines whether the difference between the preceding melody sound PM and the current melody sound CM is within two semitones (step S116). Steps S115 and S116 are the same as steps S63 and 64 in FIG. When it is determined No in step S115 or when it is determined Yes in step S116, the CPU 21 gives the preceding code name PreCH as the current code name CurCH (step S117). That is, the preceding code name PreCH is held.

  In the 2nd beat and the 4th beat, since the chord hold in which the preceding chord name is maintained may be performed, an appropriate chord hold is realized. In the present embodiment, when the current melody function CMF is a chord tone (CT), or when the current melody function CMF is a scale note (SN) and progresses sequentially, chord hold is performed. Yes. In a 4-beat music, the second and fourth beats are weak. Therefore, unless the melody emphasizes weak beats, the second and fourth chords basically maintain the first and third beat chords.

  For example, in the music shown in FIG. 46, there are sounds of “C”, “D”, “E”, “F”, “E”, “D”, “C” at the beginning, and the sequence of these sounds is in sequence. It is progressing. In practice, a suitable code name for this sequence is IMaj (CMaj). However, if the processing shown in steps S115 to S117 is not executed, the chord name will be other than IMaj (CMaj) at the second and fourth beats. Therefore, as shown in steps S115 to 116, code holding is performed under a certain condition so that an appropriate code name can be obtained.

  After step S117, the CPU 21 sets the code determination flag in the RAM 23 to “1” (step S118). In this case, since the current code name CurCH is already determined in step S117, the subsequent code determination process is not required.

  If it is determined No in step S116, the process proceeds to step S112, and the preceding melody sound PM and the current melody sound CM are maintained.

  The processing in the case where it is determined No in step S96 is shown in FIG. Steps S121 to S125 in FIG. 14 are the same as steps S71 to S75 in FIG. If it is determined Yes in step S124, the process proceeds to step S115 in FIG. 10 to determine code hold.

  Next, the fourth beat corresponding note determination process (step S30) will be described. 15 to 16 are flowcharts showing an example of the fourth beat corresponding note determination process according to the present embodiment. The fourth beat corresponding note determination process is similar to the second beat corresponding note determination process.

  As shown in FIG. 15, in the fourth beat corresponding note determination process, the presence / absence of the first beat key depression and the process (steps S91 to S92 in FIG. 11) are omitted. In FIG. 15, steps S131 to S135 are the same as steps S93 to 97 of FIG. If it is determined Yes in step S131, the process of FIG. 12 is executed.

  If it is determined No in step S132, the process shown in FIG. 16 is executed. In FIG. 16, steps S141 to 146 are the same as steps S111 to 116 in FIG. In the note processing corresponding to the fourth beat, when it is determined No in step S145 or Yes in step S146, the second dominant motion determination process (step S147) is further executed, and the code hold is performed based on the result. Judging what should be done.

  FIG. 17 is a flowchart illustrating an example of the second dominant motion determination process according to the present embodiment. In the second dominant motion determination process, only the transition of the melody sound is determined, and the chord type of the preceding chord name PreCH is not considered. As shown in FIG. 17, the CPU 21 determines whether (PM, CM) is any one of (F, E), (B, C), and (D, C) (step S151). This is the same as step S82 in FIG. When it is determined No in step S151, the CPU 21 determines whether (PM, CM) is any one of (G #, A), (B, A), and (D, C) ( Step S153). This is the same as step S86 in FIG.

  If it is determined Yes in step S151, or if it is determined Yes in step S153, the CPU 21 stores information indicating that it is the first determination result in the dominant motion processing in the RAM 23 (step S151). S152). On the other hand, when it is determined No in step S153, the CPU 21 stores information indicating that it is the first determination result in the dominant motion processing in the RAM 23 (step S154).

  When the second determination result is obtained by the second dominant motion determination process (Yes in step S148), the CPU 21 gives the preceding code name PreCH as the current code name CurCH (step S149). That is, the preceding code name PreCH is held. Next, the CPU 21 sets the code determination flag in the RAM 23 to “1”. On the other hand, if Yes is determined in step S148, the process proceeds to step S142, and the preceding melody sound PM and the current melody sound CM are maintained.

  If it is determined No in step S134, the process shown in FIG. 14 is executed.

  When the note determination process corresponding to each beat shown in steps S27, 29, and 30 is completed, the chord determination process is executed based on the preceding melody sound PM and the current melody sound CM corrected by the process (step S31). ). 18 to 21 are flowcharts illustrating an example of the code determination process according to the present embodiment.

  As shown in FIG. 18, the CPU 21 determines whether or not the current code name CurCH has already been determined by the code hold, and the code determination flag is “1” (step S161). If it is determined No in step S161, that is, if the chord determination flag is “1”, the CPU 21 stores the current chord name CurCH and its sounding time in a predetermined area of the RAM 23 (step S175).

  When it is determined Yes in step S161, that is, when the code determination flag is not “1”, the CPU 21 acquires the preceding melody sound PM and the current melody sound CM from the RAM 23 (step S162). The CPU 21 determines whether the preceding melody sound PM is a song start sound (step S163). In step S163, for example, it may be determined that there is no sound pressed at a time prior to the preceding melody sound PM. When it is determined No in step S163, the CPU 21 determines whether the preceding chord sound PreCH is a non-determined code (step S164).

  If it is determined Yes in step S163, or if it is determined Yes in step S164, the CPU 21 gives the current melody sound CM as the preceding melody sound PM (step S165). This is because if it is determined Yes in step S164, it is more appropriate to start a new melody sequence because the preceding chord name PreCH is an undetermined code.

  Next, the CPU 21 refers to the melody sequence table and acquires a set of values corresponding to (PM, CM) (step S166). FIG. 22 is a diagram showing an example of a melody sequence table according to the present embodiment. As shown in FIG. 22, the melody sequence table 2200 stores a set of values for a predetermined preceding melody sound PM and a current melody sound CM. If a set of values corresponding to (PM, CM) exists in the melody sequence table 2200, the set of values may be temporarily stored in the RAM 23. On the other hand, if the value set corresponding to (PM, CM) does not exist in the melody sequence table 2200, information indicating that the value set does not exist may be stored in the RAM 23.

  Next, the CPU 21 specifies the sound that was pressed immediately before the current melody sound CM (the sound immediately before the CM) (step S167). The sound pressed immediately before here is a sound actually pressed immediately before, and may not be the same as the preceding melody sound PM. The CPU 21 compares the sound immediately before the CM specified in step S167 with the current melody sound CM, and determines whether the pitch difference is 5 semitones or more (step S168). If it is determined Yes in step S168, the CPU 21 determines whether or not the current melody sound CM is related to the key depression of the first beat (step S169).

  The melody sequence includes a melody sound that should be a core, and there may be a melody sound that decorates the core melody sound before and after that. Normally, the melody sound to be decorated does not have much pitch difference from the core melody sound. On the other hand, when jumping so that the pitch difference is not less than a certain range (for example, about 4 degrees), the weight after the jump is often relatively large. Therefore, in step S168, the pitch difference is determined, and different processing is performed based on the difference.

  When it is determined Yes in step S169, the CPU 21 determines whether the preceding code name PreCH continues for two bars or more (step S171). When it is determined Yes in step S171, the CPU 21 determines to refer to the column “jump2” as the type in the first code table, and sets a predetermined code name in the first code table. Obtain (step S172). On the other hand, when it is determined No in step S172, the CPU 21 determines to refer to the column “jump1” as the type in the first code table, and the predetermined code in the first code table. A code name is acquired (step S173).

  FIG. 23 is a diagram illustrating an example of a first code table according to the present embodiment. FIG. 23 shows a part of the first code table. As shown in FIG. 23, in the code table 2300, at least a set of a preceding code name function (preceding chord function (PreviousChordFunction): see reference numeral 2310) and (preceding melody sound PM, current melody sound CM) (for example, reference numeral 2301) 2303), the code name is determined.

  Further, in the present embodiment, three types (see reference numeral 2311) of “no jump”, “jump1”, and “jump2” are provided, and a set of a preceding chord function and (preceding melody sound PM, current melody sound CM). The code name is determined based on the type.

  There are three preceding code functions: tonic (TO), subdominant (SU), and dominant (DO). Code names corresponding to tonics include “IMaj”, “IM7”, “IIImin”, “IIIm7”, “VImin”, and “VIm7”. Code names corresponding to the subdominant include “IImin”, “IIm7”, “IIm7 (−5)”, “IVMaj”, “IVM7”, “IVmin”, and [IVmM7]. As code names corresponding to dominant, there are “IIIMaj”, “III7”, “III7sus4”, “VMaj”, “V7”, “V7sus4”, and “VIIm7 (−5)”. The corresponding code name for each preceding code function is stored in advance in the RAM 23, for example.

  In addition, “jump2” in the type increases the degree of chord change in consideration of the jump in the melody sequence and the continuation of the preceding chord name. On the other hand, the degree of code change is smaller in “jump1” than in “jump2”. The type “no jump” is used when there is no “jump1” or “jump2” type even when the first code table is used, as will be described later.

  For example, if the preceding code function of the preceding code name PreCH is “Tonic”, (preceding melody sound PM, current melody sound CM) = (C, G), and the type is “jump2”, the first “VMaj” (see reference numeral 2321) is acquired from the code table of FIG. Also, if the preceding code function of the preceding code name PreCH is “Tonic”, (preceding melody sound PM, current melody sound CM) = (C, G), and the type is “jump1”, the first "IMaj" (see reference numeral 2322) is obtained from the code table.

  The CPU 21 stores the code name specified from the first code table in the predetermined area of the RAM 23 as the current code name CurCH, and stores the sound generation time in the predetermined area of the RAM 23 (steps S174 and 175).

  Next, the case where it is determined No in step S168 or No in step S169 will be described. The CPU 21 determines whether or not the current melody sound CM is a chord tone (CT) of the preceding chord name PreCH (step S181). When it is determined Yes in step S181, the CPU 21 determines whether the sounding time of the preceding chord name PreCH is within 2 beats based on the sounding time of the musical tone of the preceding chord name PreCH and the current time (step S182). ). When it is determined Yes in step S182, the CPU 21 further determines whether the syncopation flag is “1” (step S183).

  If it is determined No in step S183, the CPU 21 executes the second dominant motion process (step S185), and determines whether the transition from the preceding melody sound PM to the current melody sound CM is a dominant motion. To do. When the result of the second dominant motion process is the second determination result (Yes in step S185), the CPU 21 gives the preceding code name PreCH as the current code name CurCH (step S186). That is, the preceding code name PreCH is held.

  If NO in step S181, NO in step S182, or NO in step S185, CPU 21 determines that a set of values corresponding to (preceding melody sound PM, current melody sound CM) is in the melody sequence table. (Step S187). In step S166 in FIG. 18, information indicating that a value set or a value set does not exist is stored in the RAM 23, so that the determination in step S187 can be made by referring to the information.

  If it is determined Yes in step S187, it is determined whether the current melody sound CM relates to the first or second beat and the table designation information indicates the first table (step S188). ). If it is determined Yes in step S188, the CPU 21 determines to refer to the “no jump” field as the type in the first code table, and the predetermined code name in the first code table. Is acquired (step S189). On the other hand, when it is determined No in step S188, the CPU 21 determines to refer to the second code table, and acquires a predetermined code name in the second code table (step S190). ).

  FIG. 24 is a diagram illustrating an example of a second code table according to the present embodiment. FIG. 24 shows a part of the second code table. As shown in FIG. 24, in the code table 2300, a set of preceding code name functions (preceding chord function (Previous Chord Function): see reference numeral 2410) and (preceding melody sound PM, current melody sound CM) (for example, reference numerals 2401, 2402). The code name can be determined based on (see below). For example, if the preceding code function of the preceding code name PreCH is “subdominant” and (preceding melody sound PM, current melody sound CM) = (C, G), “VMaj” is obtained from the second code table. (See reference numeral 2421).

  Thereafter, the CPU 21 stores the code name specified from the first code table or the second code table in the predetermined area of the RAM 23 as the current code name CurCH, and stores the sound generation time in the predetermined area of the RAM 23. (Steps S191 and 175).

  If it is determined Yes in step S187, an appropriate chord name is acquired by referring to the chord table because there is a set of (preceding melody tone PM, current melody tone CM) in the melody sequence table. On the other hand, when it is determined No in step S187, modulation or temporary determination of a non-determined code is performed. If it is determined No in step S187, the CPU 21 determines whether the sounding time (key pressing time) of the current melody sound CM is longer than a quarter note, that is, whether it is sounding longer than one beat (step). S201).

  When it is determined No in step S201, the CPU 21 does not change the preceding code name PreCH, and uses the current code CurCH as it is (step S207). If No in step S201, the performer did not intentionally press the key, but there is a high possibility that the wrong key was pressed due to a mistouch. Therefore, in this case, the previous code name PreCH is directly used as the current code name CurCH without changing the code name.

  If it is determined Yes in step S201, the CPU 21 determines whether the sounding time (key pressing time) of the current melody sound is 3 beats or less (step S202). If it is determined Yes in step S202, it is determined whether the modulation flag is “2” or less. When it is determined Yes in step S203, the CPU 21 increments the value of the modulation flag stored in the RAM 23 (step S205). When it is determined No in step S202 or No in step S203, the CPU 21 executes a modulation process (step S204).

  In the present embodiment, basically, the melody sound including the current melody sound CM and the preceding melody sound PM is processed with a scale (tonality) of “C”. Therefore, in the modulation process, a pitch difference between the tone after the modulation and “C” is calculated, and the pitch difference may be stored in the RAM 23 as an offset. After the modulation process, it is possible to continue the process with the tone of “C” by reducing the pitch name specified by the key number actually pressed by the offset.

  After step S205, the CPU 21 determines whether the current melody sound CM is the chord tone (CT) or scale note (SN) of the preceding chord name PreCH (step S206). In step S206, as in step S61, the CPU 21 refers to the melody function table to determine whether the value associated with the set of the current melody sound CM and the preceding chord name PreCH is a chord tone or a scale note. Just judge. If it is determined Yes in step S206, the CPU 21 gives the preceding code name PreCH to the current code name CurCH to hold the code (step S207).

  If it is determined No in step S207, the CPU 21 refers to the non-determined code table and gives the code of the diminished (dim) or the augment (aug) to the current code name CurCH (step S208). FIG. 26 is a diagram showing an example of a non-determination code table according to the present embodiment. In the undetermined code table 2600 shown in FIG. 26, a predetermined value can be acquired corresponding to a set of values including the current melody sound CM and the preceding chord name PreCH.

  In 2600 of the undetermined code table, a blank (for example, reference numeral 2601) indicates that the current melody function (CMF) associated with the set of values including the current melody sound CM and the preceding chord name PreCH is the chord tone (CT). Or it means a scale note (SN) (see FIG. 25). Therefore, a value set of the current melody sound CM and the preceding chord name PreCH that is blank in the undetermined code table 2600 is not stored because the current chord name CurCH does not become an undetermined code. Therefore, the non-determined code table 2600 stores information specifying either diminished (dim) or augmented (aug) when the current melody function (CMF) is the other tone (OT).

  The CPU 21 acquires, from the undetermined code table 2600, information specifying either a diminished (dim) or an augment (aug) associated with a set of values including the current melody sound CM and the preceding chord name PreCH. Thus, the chord name having the current melody tone CM as the root tone is obtained. For example, if the current melody sound is “C #” and the preceding chord name is “IMaj”, the chord name is “I # dim” (see reference numeral 2611). If the current melody sound is “A ♭” and the preceding chord name is “IM7”, the chord name is “IV ♭ aug”. In this way, the CPU 21 determines the chord name of the diminished (dim) or augment (aug) having the current melody sound CM as the root tone as the current chord name CurCH and stores it in the RAM 23.

  As described above, in the present embodiment, first, how many beats the current melody sound CM related to the key pressing at the beginning of the current beat and the preceding melody sound PM related to the key pressing at the beginning of the previous beat. Is corrected in accordance with the information indicating the preceding code name, PreCH, key pressing timing, and the like (steps S21 to 30 in FIG. 5). Then, the current chord name CurCH is determined based on the current melody sound CM, the preceding melody sound PM, and the preceding chord name PreCH (step S31).

  When the chord name determination process (step S4 in FIG. 3) ends, the CPU 21 executes an automatic accompaniment process (step S5). FIG. 45 is a flowchart showing an example of automatic accompaniment processing according to the present embodiment. First, the CPU 21 determines whether or not the automatic tone correction device 10 is operating in the automatic accompaniment mode (step S411). If it is determined Yes in step S411, it is determined whether or not the current time has reached the event execution timing for the melody sound data in the automatic accompaniment data with reference to a timer (not shown) of the CPU 21. (Step S412).

  The automatic accompaniment data includes three types of musical sounds, that is, chord sound, bass sound, and rhythm sound data. The bass sound data and chord sound data include the pitch, sound generation timing, and sound generation time for each musical sound to be generated. The rhythm sound data includes the sound generation timing for each musical sound (rhythm sound) to be generated.

  When it is determined Yes in step S412, the CPU 21 executes bass sound generation / mute processing (step S413). In the melody pronunciation / mute process, it is determined whether the event related to the process is a note-on event. A note-on event can be determined by the fact that the current time substantially coincides with the tone generation timing of a predetermined musical sound in the bass sound data. On the other hand, a note-off event can be determined by the fact that the current time substantially coincides with the time obtained by adding the sound generation time to the sound generation timing.

  If the event related to the process is a note-off event, the CPU 21 executes a mute process. On the other hand, if the event related to the process is a note-on event, the sound generation process according to the bass sound data is executed.

  Next, the CPU 21 refers to a timer (not shown) of the CPU 21 to determine whether the current time has reached the event execution timing for the chord data in the automatic accompaniment data (step S414). When it is determined Yes in step S414, the CPU 21 executes chord sound generation / mute processing (step S415). In the chord sound generation / mute process, the sound generation process is executed for the chord sound that has reached the sound generation timing, while the mute process is executed for the chord sound that has reached the mute timing.

  Thereafter, the CPU 21 determines whether or not the current time has reached the event execution timing for the rhythm data in the automatic accompaniment data (step S416). When it is determined Yes in step S416, the CPU 21 executes a rhythm sound generation process (step S417). In the rhythm sound sound generation process, a note-on event is generated for the rhythm sound that has reached the sound generation timing.

  When the automatic accompaniment process (step S5 in FIG. 3) ends, the CPU 21 executes a sound source sound generation process (step S6). In the sound source sound generation process, the CPU 21 gives data indicating the tone color and pitch of the tone to be generated to the sound source unit 26 based on the generated note-on event, or indicates the tone color and pitch of the tone to be muted. Data is supplied to the sound source unit 26. The sound source unit 26 reads the waveform data in the ROM 22 in accordance with data indicating tone color, pitch, tone length, etc., and generates predetermined musical tone data. Thereby, a predetermined musical sound is generated from the speaker 28. Further, the CPU 21 instructs the sound source 26 to mute the pitch indicated by the note-off event based on the note-off event.

  When the sound source sound generation process (step S6) is completed, the CPU 21 stores other processes (for example, storing the current chord name CurCH and the current melody sound CM in the RAM 23, displaying an image on the display unit 15, and lighting an LED (not shown)). , Extinguishing, etc .: Step S7) is executed, and the process proceeds to Step S8.

  In step S8, the CPU 21 determines whether or not the performance is finished. Specifically, it is determined whether or not the non-real time coding correction switch 14 is pressed. When this determination is YES, the CPU 21 moves the process to step S9, and when NO, moves the process to step S2. If it is determined that the performance is completed, the RAM 23 stores the chord data and the melody data as a result of real-time chording by a series of performances by the process of step S7.

  In step S9, the CPU 21 performs non-real time code correction processing which will be described later with reference to FIG. When the non-real-time code correction process ends, the CPU 21 ends the main flow process.

  FIG. 27 is a flowchart showing an example of non-real time code correction processing according to the present embodiment.

In step S <b> 211, the CPU 21 performs adjustment correction verification processing described later with reference to FIGS. 28 and 29.
In step S212, the CPU 21 performs a chord composition sound verification process which will be described later with reference to FIG.
In step S <b> 213, the CPU 21 performs a phrase break verification process described later with reference to FIGS. 31 to 34.
In step S214, the CPU 21 performs a coding correction process to be described later with reference to FIGS.

In step S215, the CPU 21 performs a melody and chord scale comparison process, which will be described later with reference to FIG.
In step S216, the CPU 21 performs correction processing by the user, which will be described later with reference to FIG.
Thereafter, the CPU 21 ends the non-real time code correction process.

  28 and 29 are flowcharts illustrating an example of the adjustment correction verification process according to the present embodiment.

In step S <b> 221, the CPU 21 determines whether there is one song key. Specifically, the CPU 21 determines whether or not the number of types of keys (in-music keys) stored in the RAM 23 is one. This in-music key is a key determined based on the history of the pitch of the melody input by the keyboard 11. This can be obtained, for example, by extracting a pitch name group corresponding to this pitch history and searching for keys (keys) in which this pitch name group is all included in the diatonic scale. For the search, it is conceivable to use a table storing diatonic scale notes for each key.
If the determination in step S221 is YES, the CPU 21 moves the process to step S222, and if NO, moves the process to step S234.

In step S222, the CPU 21 compares the key confirmation table based on the final melody sound with the mid-music key. Specifically, the CPU 21 refers to the key confirmation table (FIG. 39) based on the final melody sound, and compares the mid-music key with the final melody sound. For example, according to FIG. 39, if the song key is “C” and the last melody sound is any one of “C, E, G”, the portion corresponding to the song key and the last melody sound is “ Since “C (1)” is stored, it is matched in step S223 described later.
On the other hand, if the song key is “C” and the last melody sound is “A”, the data is not stored in the corresponding portion of the song key and the last melody sound, so a match is made in step S223 described later. Will not.

  In step S223, the CPU 21 determines whether or not the result in step S222 matches. If this determination is YES, there is no problem with the key determination, so the CPU 21 ends the key determination verification process. If NO, the process proceeds to step S224 to correct the key.

In step S224, the CPU 21 sets a key having the final sound listed in the key determination table of the final melody sound as a candidate key according to the priority order. Specifically, the CPU 21 refers to the key confirmation table (FIG. 39) based on the final melody sound and determines a candidate key. For example, according to the key confirmation table (FIG. 39), when the final melody sound is “A”, F (1), D (2), and A (3) are stored in the column “A”. . F, D, and A indicate keys that are most likely when the melody ends with A, and the numbers indicate the priority order. Therefore, in this case where the middle key is determined as “C” in the real-time key determination and the final melody sound is A, the F key having a higher priority is the candidate key for the final part of the music.
In FIG. 39, only the key determination table for the major key is disclosed, but a key determination table for the minor key (not shown) also exists.

In step S225, the CPU 21 codes the last four bars with the candidate key, and in step S226, the CPU 21 determines whether or not the last four bars match the cadence determination table. These processes are processes for verifying whether there is no problem with the candidate key for the final part of the song.
For example, if the candidate key is “F”, the last four measures are coded with the key “F”, and the chord function of the first four measures coded is the chord of the song end portion of the cadence determination table (FIG. 40). Determine whether it matches the function. If this determination is YES, the CPU 21 moves the process to step S227, and if NO, moves the process to step S231.

  The cadence determination table of FIG. 40 is a table showing the shape of a possible cadence (end shape) divided into a song end part and a part in the song. Note that an intro part, a rust part, or the like may be added to further subdivide. In FIG. 40, “T” represents a tonic, “D” represents a dominant, and “S” represents a subdominant.

  In step S227, the CPU 21 verifies whether there is no problem with the candidate key from the beginning of the song. This process is performed when the last four measures of the candidate key match the chord function of the song end portion of the cadence determination table. For example, if the candidate key is “F”, the last four measures of the song are confirmed with the key “F”, but the CPU 21 determines whether or not the key “F” is established from the beginning of the song. The melody sound data from the beginning of the song input in real time (keyboard processing in FIG. 4) is read from the RAM 23 and verified.

  In step S228, the CPU 21 determines whether there is no problem with the candidate key. Specifically, the CPU 21 determines whether or not a candidate key has been established from the beginning of the song in step S227. If this determination is YES, the CPU 21 moves the process to step S230, and if NO, moves the process to step S229.

  In step S229, the CPU 21 corrects the key to two at the contradiction. Specifically, the CPU 21 determines from the verification result in step S227 which part from the beginning of the song is not established as a candidate key, and the compensation key is established from the beginning of the song to a portion that is not established as a candidate key. Divide the key in two from the beginning to the end of the song. For example, if the in-music key is determined to be “C” in the real-time key determination and the candidate key is “F”, the key is divided into two keys “C” and “F”. Thereafter, the process proceeds to step S221.

  In step S230, since the candidate key is established from the beginning of the song, the CPU 21 determines the key and ends the key determination verification process.

  In step S231, the CPU 21 determines whether there is a next candidate in the key determination table. Specifically, it is determined by searching the key confirmation table (FIG. 39). For example, when the final melody sound is “A” and the candidate key is “F”, “D” that is the second priority key exists as the next candidate key. On the other hand, when the final melody sound is “A” and the candidate key is “A”, there is no next candidate key. If this determination is YES, the CPU 21 moves the process to step S232, and if NO, moves the process to step S233.

  In step S232, the CPU 21 replaces the candidate key with the next candidate key determined in step S231, and proceeds to step S225.

  In step S233, the CPU 21 does not correct the key and ends the key determination verification process.

  In step S234, the CPU 21 determines whether there are two keys in the song. If this determination is YES, the CPU 21 moves the process to step S235, and if NO, moves the process to step S237.

  In step S235, the CPU 21 determines whether or not the difference between the opening key and the song key is +1 semitone, +2 semitone, or +3 semitone. This process is a process for determining whether or not so-called upscaling modulation is performed. If this determination is YES, the CPU 21 moves the process to step S236, and if NO, moves the process to step S240.

  In step S236, the CPU 21 determines the key and ends the key determination verification process.

  In step S237, the CPU 21 determines whether or not the initial key and the final key are the same. If this determination is YES, the CPU 21 moves the process to step S238, and if NO, moves the process to step S240.

  In step S238, the CPU 21 determines whether or not the last four measures match the cadence determination table. Specifically, the CPU 21 refers to the beginning of the song in the cadence determination table (FIG. 40), and the chord function of the last four measures of the opening key is in the cadence determination table (FIG. 40). It is determined whether or not the code function matches. On the other hand, for the last key, the chord function of the last four bars of the last key is referred to as the chord function of the song end part of the cadence judgment table (FIG. 40) with reference to the song end part of the cadence judgment table (FIG. 40). Determine whether they match. If this determination is YES, the CPU 21 moves the process to step S239, and if NO, moves the process to step S240.

  In step S239, the CPU 21 determines the initial key and the final key. For example, if it is determined in step S238 that only the first key matches, only the first key is confirmed. If it is determined that only the last key matches, only the last key is confirmed. Furthermore, if it is determined that both the initial key and the final key match, the initial key and the final key are determined.

  In step S240, the CPU 21 determines whether there is an unverified key. If this determination is YES, the CPU 21 moves the process to step S241, and if NO, ends the key determination verification process.

In step S241, the CPU 21 refers to the relational arbitration determination table and determines whether there is a code in this table within four measures after the key is determined. Specifically, the CPU 21 refers to the relational adjustment determination table (FIG. 41) and determines whether the main tone unique code of the table exists within 4 bars after the key is determined.
For example, when there are three keys, the first key is “C”, and the key between the first key and the last key is in the genus (the genus of the main key C as the first key is G). It is determined whether V7 (G7) or IIm (Dm), which is the main tone unique code, exists within the first four measures of the genus.
Here, the main tone unique code will be described. For example, when the main tone is C, the genus tone G is a tone in which only the fourth tone (Fa) of the main tone is raised by a semitone, and the main tone C (comprised of de, les, mi, fa, so, la, shi) And the genus G (composed of Seo, La, Shi, Do, Les, Mi, and Fah #), the genus G does not include fa. Therefore, G7 or Dm, which is a code including a fa, is not a genus G code but a main tone C code. In this embodiment, the name of the main tone unique code is used in this sense.

  In step S242, the CPU 21 determines whether or not there is a match in step S241. If this determination is YES, the CPU 21 moves the process to step S243, and if NO, moves the process to step S248.

  In step S243, the CPU 21 corrects the original key from the match. For example, the main tone (first key) is corrected from the location where the main tone unique code exists.

  In step S244, the CPU 21 determines whether or not the last four measures match the cadence determination table. In this process, the chord function of the last four measures of the corrected key is referred to in the cadence determination table (FIG. 40) by referring to the tune in the cadence determination table (FIG. 40) with the corrected key in step S243. Determine if it matches the code function in the middle. If this determination is YES, the CPU 21 moves the process to step S245, and if NO, moves the process to step S246.

  In step S245, the CPU 21 determines the key of the part (key after correction), and moves the process to step S247.

  In step S246, the CPU 21 determines whether or not there is one remaining key. If this determination is YES, the CPU 21 moves the process to step S222, and if NO, moves the process to step S247.

  In step S247, the CPU 21 determines whether there is an unverified key. If this determination is YES, the CPU 21 moves the process to step S241, and if NO, ends the key determination verification process.

  In step S248, the CPU 21 determines whether or not the last four measures match the cadence determination table. In this process, the chord function of the last four measures of the unverified key is referred to as the chord function in the song of the cadence determination table (FIG. 40) by referring to the song in the cadence determination table (FIG. 40) with the unverified key. Determine whether they match. If this determination is YES, the CPU 21 moves the process to step S245, and if NO, moves the process to step S249.

  In step S249, the CPU 21 determines whether or not the same key is checked. That is, the CPU 21 determines whether the presence / absence of the main tone unique code is checked for the same main tone which is the lowermost row of the relational borrowing determination table (FIG. 41). If this determination is YES, the CPU 21 moves the process to step S246, and if NO, moves the process to step S250.

  In step S250, the CPU 21 moves the process to step S241 on the next relational tone as a confirmation target. For example, if the genus is checked, the lower genus that is the next (lower) relation is set as the next confirmation target.

  FIG. 30 is a flowchart showing an example of the chord composition sound verification process according to the present embodiment. By the way, when a decoration sound such as a stuttering sound or a progress sound is present in a strong beat in real-time automatic coding, the decoration may affect the coding so that the coding may be wrong. Therefore, in order to correct such chording errors, this dominant tone verification process extracts chord constituent sounds obtained by removing decoration sounds from melody sounds for each measure in this embodiment, and this chord constituent sound is extracted. It is verified whether or not the code formed by is the same as the code for real-time coding.

  First, in step S261, the CPU 21 extracts three sounds in descending order of the sound length of a sound having one beat or more in the measure. Specifically, the chord data and melody data as a result of real-time chording by a series of performances stored in the RAM 23 are read out, and three sounds are extracted in order from the longest of the sounds having one or more beats in that measure. Here, the measure is a measure to be processed by the CPU 21, and the first measure is initially set as the process target.

In step S262, the CPU 21 determines whether or not the extracted three sounds are included in a chord constituent sound with a real-time code. Specifically, the CPU 21 makes a determination with reference to the code scale table of FIG. For example, when the code for real-time coding is Imajor, it is determined whether or not the three sounds extracted in step S261 or S263 are “C, E, G” that are chord constituent sounds (CT). If this determination is YES, the CPU 21 moves the process to step S264, and if NO, moves the process to step S263.
Here, if the extracted three sounds are chord constituent sounds (CT), these three sounds become dominant sounds.

  In step S263, the CPU 21 extracts three sounds in descending order of the sound length by halving the bars. Thereafter, the process proceeds to step S262.

  In step S264, the CPU 21 determines whether or not there is a melody sound other than the chord in the measure. Specifically, a sound other than the three sounds extracted in step S261 or step S263 exists in the measure to be processed by the CPU 21, and the sound is a chord component sound (see FIG. 43). If not (CT), the sound is a melody sound other than chords. If this determination is YES, the CPU 21 moves the process to step S265, and if NO, moves the process to step S270.

  In step S265, the CPU 21 determines whether or not these sounds include decoration sounds. Specifically, the CPU 21 makes a determination with reference to the decorative sound database (FIG. 42). If this determination is YES, the CPU 21 moves the process to step S266, and if NO, moves the process to step S267.

Here, the decorative sound database will be described with reference to FIG. This decoration sound database is a database for determining whether or not there is a decoration sound (a sound other than a chord constituent sound) in a strong beat (1 beat and 3 beats in the case of 4 beats).
For example, according to the example of FIG. 42, in the case of four time signatures, there is a decoration sound (for example, stuttering) other than the chord component sound in the first beat (strong beat), and the chord structure in the second beat (weak beat). If there is a sound and the chord component sound of the second beat is higher or lower than the sound of the first beat, the sound of the first beat is determined as a decoration sound. . The same applies to the third and fourth beats.
Although not shown in FIG. 42, it is also possible to determine the presence / absence of a decoration sound other than four time signatures such as two time signatures and three time signatures.

  In step S266, the CPU 21 omits the decoration sound and compares it with the code of the real-time coding result. Specifically, the CPU 21 reads the code with the real-time code stored in the RAM 23, and the sound of the measure to be processed without the decoration sound corresponds to each code in the code scale table of FIG. It is determined whether or not the sound (CT).

  In step S267, the CPU 21 compares the code with the result of real-time coding. Specifically, the CPU 21 reads the real-time coded code stored in the RAM 23, and the sound of the bar to be processed is a chord constituent sound (CT) corresponding to each chord in the chord scale table of FIG. Judge whether there is.

  In step S268, the CPU 21 determines whether or not the comparison results in steps S266 and S267 match (whether the determination in step S266 or S267 is YES (match) or NO (unmatch)). If this determination is YES, the CPU 21 moves the process to step S270, and if NO, moves the process to step S269.

  In step S269, the CPU 21 determines whether or not this dominant sound is established as a chord. Specifically, the CPU 21 refers to the chord scale table in FIG. 43 and determines whether or not the sound compared in step S266 or S267 is a chord constituent sound (CT). If this determination is YES, the CPU 21 moves the process to step S272, and if NO, moves the process to step S270.

  In step S272, the CPU 21 turns on the code correction flag. Specifically, the CPU 21 turns on the code correction flag in association with the measure to be processed. The code correction flag exists in a predetermined area of the RAM 23. Thereby, the code of the target measure can be corrected in the code correction process (FIG. 35) described later.

  In step S270, the CPU 21 determines whether it is the last measure. If this determination is YES, the CPU 21 ends the chord composition sound verification process, and if NO, the process proceeds to step S271.

  In step S271, the CPU 21 sets the next measure as the target of the chord composition sound verification, and moves the process to step S261.

  FIG. 31 to FIG. 34 are flowcharts showing examples of phrase break verification processing according to the present embodiment.

  In step S281, the CPU 21 compares the chord progressions of the first bar and the (8N + 1) bar. Specifically, the chord data and the melody data as a result of real-time chording by a series of performances stored in the RAM 23 are read and compared. The initial value of N is 1.

  In step S282, the CPU 21 determines whether the chord progressions of the first bar and the (8N + 1) bar are the same. If this determination is YES, the CPU 21 moves the process to step S296, and if NO, moves the process to step S283.

  In step S283, the CPU 21 compares melody rhythms in units of eight measures (1st to 8th measures and (8N + 1) to (8N + 8) measures). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S284, the CPU 21 determines whether or not the melody rhythms in units of eight measures (1st to 8th measures and (8N + 1) to (8N + 8) measures) match. If this determination is YES, the CPU 21 moves the process to step S285, and if NO, moves the process to step S290.

  In step S285, the CPU 21 compares melody pitches in units of eight measures (measures 1 to 8 and measures (8N + 1) to (8N + 1) to (8N + 8)). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S286, the CPU 21 determines whether or not the melody pitches in units of eight measures (1st to 8th measures and (8N + 1) to (8N + 8) measures) match. If this determination is YES, the CPU 21 moves the process to step S287, and if NO, moves the process to step S288.

  In step S287, the CPU 21 sets the corresponding measure to the phrase 8 type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S288, the CPU 21 determines whether or not the melody pitches in units of eight measures (measures 1 to 8 and measures (8N + 1) to (8N + 1) to (8N + 8)) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S289, and if NO, moves the process to step S296.

  In step S289, the CPU 21 sets the corresponding measure to the phrase 8β type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S290, the CPU 21 determines whether or not the melody rhythm in units of 8 bars (1st to 8th bars and (8N + 1) to (8N + 8) bars) matches 70% or more. If this determination is YES, the CPU 21 moves the process to step S291, and if NO, moves the process to step S296.

  In step S291, the CPU 21 compares the melody pitches in units of eight measures (measures 1 to 8 and measures (8N + 1) to (8N + 1) to (8N + 8)). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S292, the CPU 21 determines whether or not the melody pitches in the 8 bar units (1st to 8th bar and (8N + 1) to (8N + 8) bar) match. If this determination is YES, the CPU 21 moves the process to step S293, and if NO, moves the process to step S294.

  In step S293, the CPU 21 sets the corresponding measure to the phrase 8Rα type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S294, the CPU 21 determines whether or not the melody pitches in units of 8 measures (1st to 8th measures and (8N + 1) to (8N + 8) measures) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S295, and if NO, moves the process to step S296.

  In step S295, the CPU 21 sets the corresponding measure to the phrase 8Rβ type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S296, the CPU 21 determines whether or not the song has ended. If this determination is YES, the CPU 21 moves the process to step S298, and if NO, moves the process to step S297.

  In step S297, the CPU 21 increments N by 1, and moves the process to step S281.

  In step S298, the CPU 21 compares the chord progressions of the first bar and the (4N + 1) bar. Specifically, the code data by real-time coding stored in the RAM 23 is read and compared. Here, the initial value of N is 1.

  In step S299, the CPU 21 determines whether or not the chord progressions of the first bar and the (4N + 1) bar are the same. If this determination is YES, the CPU 21 moves the process to step S313, and if NO, moves the process to step S300.

  In step S300, the CPU 21 compares melody rhythms in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S301, the CPU 21 determines whether or not the melody rhythms in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures) match. If this determination is YES, the CPU 21 moves the process to step S302, and if NO, moves the process to step S307.

  In step S302, the CPU 21 compares the melody pitches in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S303, the CPU 21 determines whether or not the melody pitches in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures) match. If this determination is YES, the CPU 21 moves the process to step S304, and if NO, moves the process to step S305.

  In step S304, the CPU 21 sets the corresponding measure to the phrase 4 type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S <b> 305, the CPU 21 determines whether or not the melody pitches in units of four measures (1 to 4 measures and (4N + 1) to (4N + 4) measures) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S306, and if NO, moves the process to step S313.

  In step S306, the CPU 21 sets the corresponding measure to the phrase 4β type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S307, the CPU 21 determines whether or not the melody rhythm of the four measures (1 to 4 measures and (4N + 1) to (4N + 4) measures) matches 70% or more. If this determination is YES, the CPU 21 moves the process to step S308, and if NO, moves the process to step S313.

  In step S308, the CPU 21 compares melody pitches in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S309, the CPU 21 determines whether or not the melody pitches in units of four measures (1st to 4th measures and (4N + 1) to (4N + 4) measures) match. If this determination is YES, the CPU 21 moves the process to step S310, and if NO, moves the process to step S311.

  In step S310, the CPU 21 sets the corresponding measure to the phrase 4Rα type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S <b> 311, the CPU 21 determines whether or not the melody pitches in units of four measures (1 to 4 measures and (4N + 1) to (4N + 4) measures) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S312, and if NO, moves the process to step S313.

  In step S312, the CPU 21 sets the corresponding measure to the phrase 4Rβ type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S313, the CPU 21 determines whether or not the song has ended. If this determination is YES, the CPU 21 moves the process to step S315, and if NO, moves the process to step S314.

  In step S314, the CPU 21 increases N by 1 and moves the process to step S281.

  In step S315, the CPU 21 compares the chord progressions of the first bar and the (2N + 1) bar. Specifically, the code data by real-time coding stored in the RAM 23 is read and compared. Here, the initial value of N is 1.

  In step S316, the CPU 21 determines whether or not the chord progressions of the first bar and the (2N + 1) bar are the same. If this determination is YES, the CPU 21 moves the process to step S330, and if NO, moves the process to step S317.

  In step S317, the CPU 21 compares the melody rhythms of the two measures (the measures 1 and 2 and the measures (2N + 1) to (2N + 2)). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S318, the CPU 21 determines whether or not the melody rhythms in units of two measures (1st and 2nd measures and (2N + 1) to (2N + 2) measures) match. If this determination is YES, the CPU 21 moves the process to step S319, and if NO, moves the process to step S324.

  In step S319, the CPU 21 compares the melody pitches in units of two measures (the measures 1 and 2 and the measures (2N + 1) to (2N + 2)). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S320, the CPU 21 determines whether or not the melody pitches in the two measures (the first and second measures and the (2N + 1) to (2N + 2) measures) match. If this determination is YES, the CPU 21 moves the process to step S321, and if NO, moves the process to step S322.

  In step S321, the CPU 21 sets the corresponding measure to phrase 2 type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S322, the CPU 21 determines whether or not the melody pitches in the units of two measures (the measures 1 and 2 and the measures (2N + 1) to (2N + 2)) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S323, and if NO, moves the process to step S330.

  In step S323, the CPU 21 sets the corresponding measure to the phrase 2β type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S324, the CPU 21 determines whether or not the melody rhythm in two measures (the first and second measures and the (2N + 1) to (2N + 2) measures) matches 70% or more. If this determination is YES, the CPU 21 moves the process to step S325, and if NO, moves the process to step S330.

  In step S325, the CPU 21 compares the melody pitches in units of two measures (the measures 1 and 2 and the measures (2N + 1) to (2N + 2)). More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison.

  In step S326, the CPU 21 determines whether or not the melody pitches in units of two measures (1st and 2nd measures and (2N + 1) to (2N + 2) measures) match. If this determination is YES, the CPU 21 moves the process to step S327, and if NO, moves the process to step S328.

  In step S327, the CPU 21 sets the corresponding measure to the phrase 2Rα type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S328, the CPU 21 determines whether or not the melody pitches in the units of two measures (the measures 1 and 2 and the measures (2N + 1) to (2N + 2)) match 70% or more. If this determination is YES, the CPU 21 moves the process to step S329, and if NO, moves the process to step S330.

  In step S329, the CPU 21 sets the corresponding measure to the phrase 2Rβ type. Specifically, the CPU 21 stores bar data and type data in the phrase type area of the RAM 23.

  In step S330, the CPU 21 determines whether or not the song has ended. If this determination is YES, the CPU 21 moves the process to step S332, and if NO, moves the process to step S331.

  In step S331, the CPU 21 increases N by 1 and moves the process to step S315.

  In step S332, the CPU 21 determines whether there is no phrase type. Specifically, the CPU 21 determines whether or not there is any type data recorded in the phrase type area of the RAM 23. If this determination is YES, the CPU 21 moves the process to step S333, and if NO, ends the phrase break verification process.

  In step S333, the CPU 21 compares the melody pitches of the first bar and the (N + 1) th bar. More specifically, the melody data used for real-time coding stored in the RAM 23 is used for comparison. Here, the initial value of N is 1.

  In step S334, the CPU 21 determines whether or not the melody pitches of the first bar and the (N + 1) th bar match. If this determination is YES, the CPU 21 moves the process to step S335, and if NO, moves the process to step S336.

  In step S335, the CPU 21 sets the corresponding measure to the phrase X type. Also, the number of bars up to X is stored. Specifically, the CPU 21 stores measure data and type data in the phrase type area of the RAM 23.

  In step S336, the CPU 21 determines whether it is the last measure. If this determination is YES, the CPU 21 ends the phrase delimitation verification process, and if NO, the process proceeds to step S337.

  In step S337, the CPU 21 increases N by 1 and moves the process to step S333.

  35 and 36 are flowcharts showing an example of the coding correction process according to the present embodiment.

  In step S341, the CPU 21 compares the key with the real-time code with the key corrected in the key determination verification process (FIGS. 28 and 29).

  In step S342, the CPU 21 determines whether there is a key change portion. If this determination is YES, the CPU 21 moves the process to step S343, and if NO, moves the process to step S346.

  In step S343, the CPU 21 codes the changed part of the key with a new key (key corrected in the key determination verification process (FIGS. 28 and 29)).

  In step S344, the CPU 21 determines whether or not there is a key confirmation suspension place. If this determination is YES, the CPU 21 moves the process to step S345, and if NO, moves the process to step S346.

  In step S346, the CPU 21 determines whether or not there is a chord correction designation portion with dominant sound. Specifically, the CPU 21 determines in step S272 (FIG. 30) whether or not there is a measure for which the code correction flag is turned on. If this determination is YES, the CPU 21 moves the process to step S347, and if NO, moves the process to step S350.

  In step S347, the CPU 21 determines whether or not the last four bars of the designated portion are in agreement with the cadence determination table. Specifically, for the four bars including the bar for which the chord correction flag is turned on, the CPU 21 determines whether the four bars are in the song or the song end in the cadence determination table (FIG. 40) depending on whether the four measures are in the song or the song end part. Referring to the section, it is determined whether or not the four bar code function matches the code function of the cadence determination table (FIG. 40). If this determination is YES, the CPU 21 moves the process to step S377, and if NO, moves the process to step S348.

  In step S348, the CPU 21 specifies the key confirmation hold range, performs key determination on the melody of that part, and confirms the key.

  In step S349, the CPU 21 codes a key change portion with a new key. Thereafter, the process proceeds to step S347.

  In step S377, the CPU 21 corrects the code. Specifically, the CPU 21 corrects the chord of the bar for which the chord correction flag is turned on, using the chord corresponding to the chord constituting sound determined in step S269 and corresponding to the dominant sound.

  In step S269, the CPU 21 determines whether the dominant sound is established as a chord. Specifically, the CPU 21 refers to the chord scale table in FIG. 43 and determines whether or not the sound compared in step S266 or S267 is a chord constituent sound (CT). If this determination is YES, the CPU 21 moves the process to step S272, and if NO, moves the process to step S270.

  In step S350, the CPU 21 determines whether or not the phrase break at the corresponding location is 4 (*) type (4 type, 4β type, 4Rα type, or 4Rβ type). If this determination is YES, the CPU 21 moves the process to step S351, and if NO, moves the process to step S357.

  In step S351, the CPU 21 determines whether or not the phrase delimiter is type 4. If this determination is YES, the CPU 21 moves the process to step S352, and if NO, moves the process to step S353.

  In step S352, the CPU 21 sets the code of that portion to be the same as that of the comparison source. In the case of Type 4, since the rhythm and pitch of the melody are the same as the comparison source, the chord is also the same. For example, when the 5th to 8th bars are of the 4th type, the CPU 21 makes the codes of the 5th to 8th bars the same as the codes of the 1st to 4th bars. In addition, as will be described later, if the 9th to 12th bars are any of 4β type, 4Rα type, and 4Rβ type, the code functions of the portions that are different between the 9th to 12th bar codes and the 1st to 4th bar codes Are the same, the chords of the different parts are the same as the chords of the first to fourth measures.

  In step S353, the CPU 21 compares different portions of code. In this case, it is either 4β type, 4Rα type, or 4Rβ type, which means that either one or both of the melody pitch and the melody rhythm differ from the comparison source by less than 30%. Compare the codes. For example, if the 9th to 12th measures are any of 4β type, 4Rα type, and 4Rβ type, the codes of the 9th to 12th measures are compared with the codes of the 1st to 4th measures.

  In step S354, the CPU 21 determines whether or not the functions are the same. Specifically, the CPU 21 determines whether or not the functions of the codes in the different parts are the same. If this determination is YES, the CPU 21 moves the process to step S352, and if NO, moves the process to step S355.

  In step S355, the CPU 21 determines whether or not it is the last measure. If this determination is YES, the CPU 21 moves the process to step S357, and if NO, moves the process to step S356.

  In step S356, the CPU 21 proceeds to the next four bars, and proceeds to step S351.

  In step S357, the CPU 21 determines whether or not the phrase break at the corresponding location is the 8 (*) type (8 type, 8β type, 8Rα type, or 8Rβ type). If this determination is YES, the CPU 21 moves the process to step S358, and if NO, moves the process to step S364.

  In step S358, the CPU 21 determines whether or not the phrase delimiter is 8-type. If this determination is YES, the CPU 21 moves the process to step S359, and if NO, moves the process to step S360.

  In step S359, the CPU 21 sets the code of that portion to be the same as that of the comparison source. In the case of type 8, since the rhythm and pitch of the melody are the same as the comparison source, the chord is also the same. For example, when the 9th to 16th measures are 8 type, the CPU 21 makes the codes of the 9th to 16th measures the same as the codes of the 1st to 8th measures. As will be described later, when the 17th to 24th bars are any of 8β type, 8Rα type, and 8Rβ type, the code functions of the portions that are different between the codes of the 17th to 24th bars and the codes of the 1st to 8th bars If they are the same, the code of the different part is made the same as the code of the 1st to 8th measures.

  In step S360, the CPU 21 compares different portions of code. In this case, it is any of 8β type, 8Rα type, and 8Rβ type, which means that either one of both the melody pitch and the melody rhythm differs from the comparison source by less than 30%. Compare the codes. For example, if the 17-24th bar is any of the 8β type, the 8Rα type, and the 8Rβ type, the codes of the 17th to 24th bars are compared with the codes of the 1st to 8th bars.

  In step S361, the CPU 21 determines whether or not the functions are the same. Specifically, the CPU 21 determines whether or not the functions of the codes in the different parts are the same. If this determination is YES, the CPU 21 moves the process to step S359, and if NO, moves the process to step S362.

  In step S362, the CPU 21 determines whether it is the last measure. If this determination is YES, the CPU 21 moves the process to step S364, and if NO, moves the process to step S363.

  In step S363, the CPU 21 proceeds to the next eight bars, and proceeds to step S358.

  In step S364, the CPU 21 determines whether or not the phrase break at the corresponding location is 2 (*) type (type 2, 2β type, 2Rα type, or 2Rβ type). If this determination is YES, the CPU 21 moves the process to step S365, and if NO, moves the process to step S371.

  In step S365, the CPU 21 determines whether or not the phrase delimiter is type 2. If this determination is YES, the CPU 21 moves the process to step S366, and if NO, moves the process to step S367.

  In step S366, the CPU 21 sets the code of the portion to be the same as that of the comparison source. In the case of type 2, the melody rhythm and pitch are the same as the comparison source, so the chords are also the same. For example, when the 3rd to 4th measures are of type 2, the CPU 21 makes the codes of the 3rd and 4th measures the same as the codes of the 1st and 2nd measures. As will be described later, if the 5th to 6th bars are either 2β type, 2Rα type, or 2Rβ type, the function of the code of the portion that is different between the 5th and 6th bar codes and the 1st and 2nd bar codes Are the same, the chords of the different parts are the same as the chords of the first and second measures.

  In step S367, the CPU 21 compares different portions of the code. In this case, it is either 2β type, 2Rα type, or 2Rβ type, which means that either one or both of the melody pitch and the melody rhythm differ from the comparison source by less than 30%. Compare the codes. For example, if the 3rd to 4th measures are 2β type, 2Rα type, or 2Rβ type, the codes of the 3rd and 4th measures are compared with the codes of the 1st and 2nd measures.

  In step S368, the CPU 21 determines whether or not the functions are the same. Specifically, the CPU 21 determines whether or not the functions of the codes in the different parts are the same. If this determination is YES, the CPU 21 moves the process to step S366, and if NO, moves the process to step S369.

  In step S369, the CPU 21 determines whether it is the last measure. If this determination is YES, the CPU 21 moves the process to step S371, and if NO, moves the process to step S370.

  In step S370, the CPU 21 proceeds to the next two measures, and proceeds to step S365.

  In step S <b> 371, the CPU 21 determines whether or not the phrase break at the corresponding portion is X-type. If this determination is YES, the CPU 21 moves the process to step S372, and if NO, moves the process to step S376.

  In step S372, the CPU 21 confirms the start position.

  In step S373, the CPU 21 sets the code of that portion to be the same as that of the comparison source.

  In step S374, the CPU 21 determines whether it is the last measure. If this determination is YES, the CPU 21 moves the process to step S376, and if NO, moves the process to step S375.

  In step S375, the CPU 21 proceeds to the next X type, and proceeds to step S372.

  In step S376, the CPU 21 determines whether there are other phrase-type storage locations. If this determination is YES, the CPU 21 moves the process to step S350, and if NO, ends the coding correction process.

  FIG. 37 is a flowchart showing an example of comparison processing between a melody and a chord scale according to the present embodiment.

In step S381, the CPU 21 confirms the scale of the corrected chord in the chord scale table (FIG. 43), and the melody corresponding to the corrected chord section (the melody input at the time of real-time chording) is corrected. Check whether the code scale of the selected code matches. Here, the chord scale is a group of sounds combining the scale note (SN) and the chord constituent sound (CT) as referred to in the chord scale table of FIG.
For example, if the modified code is Imajor (CMajor) with the key C, the code scale of the modified code is “C, D, E, F, G, A, B” according to FIG. It is.
Therefore, when the modified chord is IMajor (CMajor) with the key C, the melody corresponding to the modified chord section is “C, D, E, F, G, A, B”. If there is, the melody matches the chord scale.

  In step S382, the CPU 21 determines whether or not they match in step S381. If this determination is YES, the CPU 21 moves the process to step S388, and if NO, moves the process to step S383.

  In step S383, the CPU 21 determines whether the off-scale sound is a decoration sound. Specifically, the CPU 21 makes a determination with reference to the decorative sound database (FIG. 42) as in step S265 (FIG. 30). If this determination is YES, the CPU 21 moves the process to step S388, and if NO, moves the process to step S384.

In step S384, the CPU 21 refers to a rehabilitation database (not shown). For example, the key is Am (the constituent sounds of Am are “A, B, C, D, E, F, G”), and the melody is an upward type of melodic minor (A, B, C, D, E, In the case of F #, G #), since “F #, G #” is not a constituent sound of the key Am, the chord can be changed to Bm7 with respect to the sound of F #.
Thus, in the rehabilitation database (not shown), patterns that can change codes are registered as a database.

  In step S385, the CPU 21 determines whether or not there is a match in the rehabilitation database. If this determination is YES, the CPU 21 moves the process to step S386, and if NO, moves the process to step S387.

  In step S386, the CPU 21 corrects the code. Specifically, the CPU 21 corrects the code to be registered in the rehabilitation database.

  In step S387, the CPU 21 records the nonconforming portion. That is, the part which did not match is recorded on RAM23.

  In step S388, the CPU 21 determines whether it is the last code. If this determination is YES, the CPU 21 ends the melody and chord scale comparison process, and if the determination is NO, the process proceeds to step S389.

  In step S389, the CPU 21 proceeds to the next code, and proceeds to step S381.

FIG. 38 is a flowchart showing an example of correction processing by the user according to the present embodiment.
Even non-real-time corrections may be frustrating for users with advanced ears. At that time, it is convenient if there is a correction function by the user. The correction process by the user is a process for that purpose.

  In step S401, the CPU 21 determines whether data reproduction by the Any key is being performed. If this determination is YES, the CPU 21 moves the process to step S402, and if NO, corrects the correction process by the user.

  In step S402, the CPU 21 determines whether an Edit switch (not shown) is on. If this determination is YES, the CPU 21 moves the process to step S404, and if NO, moves the process to step S403.

  In step S403, the CPU 21 outputs code candidates in the order of code priority database. Specifically, the CPU 21 outputs code candidates in the order stored in the code priority database (FIG. 44). For example, if the code before correction is IIm, “IIm7” is output as the first code candidate.

  In step S404, the CPU 21 determines whether or not an OK switch (not shown) has been pressed. If this determination is YES, the CPU 21 moves the process to step S405, and if NO, moves the process to step S406.

  In step S405, the CPU 21 corrects the code and moves the process to step S407. Specifically, the CPU 21 corrects the code with the code output in step S403.

  In step S406, the CPU 21 outputs the next candidate, and the process proceeds to step S404. For example, if the currently output code is “IIm7”, the CPU 21 outputs “IV” which is the second code candidate.

  In step S407, the CPU 21 determines whether or not the song has ended. If this determination is YES, the CPU 21 ends the correction process by the user, and if NO, the process proceeds to step S401.

The electronic musical instrument 10 of the present embodiment has a keyboard 11 for playing a melody of a song, and the CPU 21 assigns a chord to the melody in real time based on the pitch history of the melody being played. After the melody performance is finished, the chord assigned in real time is corrected in non-real time.
Therefore, after performing chording on the input melody in real time, the chording accuracy can be further improved by correcting the chord in non-real time.

Further, in the present embodiment, the CPU 21 verifies whether or not there is a sound in which a chord other than the chord given in the melody sound included therein exists for each predetermined section of the song. The specified section is specified as a correction location, the melody of the song is divided into a plurality of phrases of a predetermined length, the degree of musical coincidence between the phrases is verified, and the grant is based on the verification result Fix the code that was written.
Therefore, for example, when the phrase is divided into phrases of four bars as a predetermined length, code correction is performed based on the musical score matching degree for every four bars. It is possible to increase the matching degree of coding between each other.

Further, in the present embodiment, the CPU 21 verifies whether or not the key assigned to the predetermined section of the song matches the key determined based on the final sound of the melody in the predetermined section.
Therefore, it can be verified based on the final sound of the melody whether or not the key assigned to the predetermined section of the song is accurate.

Further, in the present embodiment, the CPU 21 determines whether or not the scale of each corrected chord and the scale of the melody in the corresponding section match. correct.
Therefore, the accuracy of the modified code is verified, and if it is not accurate, the code can be further corrected, so that the coding accuracy can be further improved.

In the present embodiment, the CPU 21 assigns the pitch belonging to a predetermined beat in the song data, the pitch belonging to the beat immediately before the beat, and the chord assigned to the beat immediately before in the real-time coding. The chord to be given to a predetermined beat is determined based on the above.
Therefore, real-time coding can be performed quickly and accurately.

In the present embodiment, the CPU 21 determines, for each measure of the song, whether or not there is a sound other than the constituent sounds of the given chord among the melody sounds included in the measure. If it is determined that the other sound is a decoration sound, it is determined whether or not the other sound is a decoration sound. If it is determined to be a decoration sound, the chord based on the sound excluding the decoration sound and the bar are included. Compares the assigned chords, and if they are not identified as decorative sounds, compares the chords based on the other sounds with the chords given to the bars, and if both chords are judged not to match , The section of the bar is designated as the correction part.
Therefore, the chord can be corrected with certainty using the dominant sound, excluding the decoration sound that has caused the coding error in real time.

In the present embodiment, the CPU 21 determines, for each measure of the song, whether or not there is a sound other than the dominant sound among the melody sounds included in the measure. For each measure, three long sounds are extracted from the melody sounds included in the measure, and whether these three sounds are constituent sounds of the chord assigned to the measure, are determined as chord constituent sounds. In such a case, these three sounds are set to be dominant sounds.
Therefore, the dominant sound can be a chord constituent sound.

In the present embodiment, the CPU 21 divides the tune into every four bars, every eight bars, and every two bars when dividing the melody of the music into a plurality of phrases having a predetermined length.
Accordingly, it is possible to verify the degree of coincidence of musical expressions using the number of measures that are musically repeated, such as 4 bar units, 8 bar units, and 2 bar units.

In the present embodiment, the song data further includes a rhythm sound, and the CPU 21 compares the pitches of melody sounds and rhythm sounds between phrases as a musical expression, and each phrase is both a pitch and a rhythm sound. Are matched, one of the pitch and the rhythm sound is matched and the other is matched by a predetermined value or more, and the pitch and the rhythm sound are both sorted by a predetermined value or more.
Accordingly, it is possible to increase variations of the verification method for the coincidence of the musical style.

Further, in the present embodiment, when there is a song section that is determined to be a corrected portion by the chord composition sound verification when the chord correction is performed, the CPU 21 assigns another chord in the section, thereby It is determined whether or not the last part is terminated in a form that conforms to cadence. If it is determined that it is terminated, the assigned code is corrected to the other code.
Therefore, more appropriate coding can be performed.

Further, in the present embodiment, when there is a song section that is determined to be a corrected portion by the chord composition sound verification when the chord correction is performed, the CPU 21 assigns another chord in the section, thereby It is determined whether or not the last part is terminated in a form that conforms to cadence, and if it is determined that it has not been terminated, the key is determined by the melody sound included in the section, and based on the determined key It is determined whether or not the newly assigned code is terminated again along the cadence.
Therefore, more appropriate coding can be performed.

In the present embodiment, the CPU 21 divides the melody of the song into a plurality of phrases having a predetermined length, and then the chords assigned to the phrases classified when both the pitch and the rhythm sound match. The phrase is classified as either the case where one of the pitch and rhythm sound matches and the other matches the predetermined value or more, and both the pitch and rhythm sound match the predetermined value or more. In the case where the functions of different portions of the code assigned to each are the same, the different portions of the code are made the same.
Therefore, the coding accuracy can be further increased.

  In the present embodiment, in the chord name determination process (step S4 in FIG. 3), the CPU 21 sets time information that defines the progress of the automatic accompaniment data being executed in the melody sequence that proceeds in a series of operations of the keyboard 11. In particular, based on beat information, the current melody sound CM related to the key pressed at the beginning of the current beat and the preceding melody related to the key pressed at the beginning of the previous beat, which is the previous beat Sound PM is determined. Further, the CPU 21 determines the current chord name CurCH based on the determined current melody sound information, the preceding melody sound information, and the preceding chord name PreCH which is the chord name in the previous beat (FIG. 5). Step S31) is executed. Further, when determining the melody sound, the CPU 21 determines the current melody sound CM and the preceding melody sound PM based on the number of beats in the measure.

  That is, according to the present embodiment, the current melody sound CM and the preceding melody sound PM are determined in consideration of the position of the pressed key (temporal position), and the determined melody sound sequence and the preceding The current code name CurCH is determined based on the code name PreCH.

  In the present embodiment, when the time signature is 4 beats, the CPU 21 determines whether the current beat is the first beat, the third beat, or another beat. The melody sound information and the preceding melody sound information are determined. That is, the weight of the beat is taken into account by determining the current melody sound CM and the preceding melody sound PM based on the strong beat (first beat, third beat) and the weak beat (second beat, fourth beat). It becomes possible to do.

  In the present embodiment, the CPU 21 determines that the key depression after the head of the previous beat extends to the current beat and determines that it is a syncopation, and extends the current melody sound CM to the current beat. Decide on the keystrokes you have. That is, even if it is not a key depression, it can be handled in the same manner as a key depression when a syncopation is formed.

  In addition, in the first dominant motion determination process (FIG. 10), the CPU 21 indicates a dominant code configuration in which the preceding chord name PreCH indicates a dominant chord and the current melody tone CM is predetermined from the preceding melody tone PM. When a transition from a sound to a constituent sound of a tonic code is indicated, a chord name corresponding to a tonic is given as the current chord name CurCH. In this way, when a dominant motion is clearly indicated from the melody sequence, the current chord name CurCH is used as a tonic to form an end in chord progression.

  Further, in the second dominant motion determination process (FIG. 17), the CPU 21 indicates a transition from the preceding melody sound PM to the predetermined melody sound to the tonic code sound from the preceding melody sound PM. In this case, a code name corresponding to a tonic is given as the current code name CurCH. Here, even if the preceding chord name PreCH is not a dominant chord, if a dominant motion is clearly indicated from the melody sequence, the current chord name CurCH is used as a tonic to form an end in chord progression.

  Further, when the code name corresponding to the tonic is not assigned as the current code name CurCH by the first dominant motion determination process or the second dominant motion determination process, the CPU 21 sets the preceding code name as the current code name CurCH. Give PreCH. Thereby, code retention can be realized.

  Further, in the present embodiment, when the current melody sound CM relates to the key depression of the first beat, the code names associated with the preceding melody sound PM, the current melody sound CM, and the preceding code name information PreCH are stored. When the chord table of 1 and the current melody sound CM relate to key depression other than the first beat, the second melody that stores the chord name associated with the preceding melody sound PM, the current melody sound CM, and the preceding chord name PrCH is stored. The CPU 21 refers to the first code table or the second code table according to the key pressing timing. This makes it possible to acquire different chord names according to beats. Also, by referring to the table, it is possible to determine the code name in real time.

  Furthermore, in the present embodiment, the current code name when there is no corresponding code name in the first code table or the second code table based on the determined preceding melody sound PM and current melody sound CM. A non-determined code such as augment or dim is assigned as CurCH. As a result, even if the preceding melody sound PM or the current melody sound is not a chord constituent sound or a scale note, it is possible to give some chord name that is not so uncomfortable in the music.

  Further, in the present embodiment, the CPU 21 refers to the non-determined code table, and from the current melody sound CM, the preceding melody sound PM, and the preceding chord name PreCH, either an augment (aug) or a diminished (dim) It becomes possible to judge whether it should be code.

  The present invention is not limited to the above embodiments, and various modifications can be made within the scope of the invention described in the claims, and these are also included in the scope of the present invention. Needless to say.

  For example, in the above-described embodiment, the time signature is set to 4 time signatures, but the present invention can be applied to 3 time signatures or 6 time signatures. For example, in the case of three time signatures, the processing for the first to third beats among the above processing may be used. In addition, six beats are considered to have two three beats, and the above first to third beat processing is used. Further, the fourth to sixth beats may be handled in the same manner as the first to third beats, and the same processing may be executed.

  In the above embodiment, the chord name using the frequency with respect to the main tone (root tone) is obtained in the tonality of C major (CMaj) or A minor (Amin), but the present invention is not limited to this. The present invention can also be applied to the tonality. In this case, for example, if the music is in major key, the pitch difference between “C” and the root tone of the tonality of the music is calculated, and the offset value is stored in the RAM 23 using the pitch difference as an offset. Just do it. In the process, it is possible to execute the process on a scale of “C” by subtracting the pitch name specified by the actually pressed key number by the offset.

The invention described in the scope of claims at the beginning of the filing of the present application will be appended.
[Appendix 1]
A performance means for playing a song melody,
Real-time key determination means for determining the key of the song melody in real time based on the pitch history of the song melody being played,
A non-real-time tone correcting means for correcting the key determined by the real-time key determining means in non-real time after the melody performance by the playing means is completed;
An automatic tone correction device.
[Appendix 2]
The non-real time tone correction means includes
A confirmation key DB representing the final pitch of the song melody and the relationship between the final pitch and the key,
It is determined whether the key read by the final pitch from the determination key DB matches the key determined by the real-time key determination means. Real-time tone judgment means;
The automatic tone correction device according to claim 1, comprising:
[Appendix 3]
The automatic tone correction device further includes:
When it is determined that the non-real-time key determination means does not match, the key candidate having a close relationship with the final pitch of the melody of the song is read from the confirmation key DB, and the last key of the song is determined based on the key candidate. When a code is provisionally assigned to a section, a cadence determination means for determining whether or not the provisionally provided code is terminated in a form along cadence,
When it is determined by the cadence determination means that the key candidate has no contradiction from the beginning of the song, and when it is verified that there is no contradiction, the key candidate is determined as a definite key. And an automatic tone correction device according to claim 2.
[Appendix 4]
The automatic tone correction device further includes:
An assigning means for assigning the key candidate to a section having no contradiction of music and assigning the determined key to other sections when it is determined by the verification means that there is a contradiction;
It is determined whether the relationship between the key assigned to the beginning of the song and the key assigned to the other sections is a climax modulation, and if it is determined to be a climax modulation, both of the assigned tones are determined. Swell-up modulation judging means with a definite tone,
The automatic tone correction device according to claim 3.
[Appendix 5]
The automatic tone correction device further includes:
It has a relational tone and a borrowed chord DB indicating the relation between the relational tone and the borrowed chord,
If there is a section in the song to which a plurality of keys are assigned and an unverified key is assigned, and if there is a chord that is not in the relation key of the unverified key in that section, The automatic tone correction device according to claim 1, wherein the verification tone is a definite tone.
[Appendix 6]
Determining in real time the song melody based on the pitch history of the song melody being played;
Correcting the determined key in non-real time after the melody performance is completed;
An automatic tone correction method.
[Appendix 7]
Determining the key of the song melody in real time based on the pitch history of the song melody being played;
Correcting the determined key in a non-real time after the music melody performance is completed;
That causes a computer to execute.

DESCRIPTION OF SYMBOLS 10 ... Electronic musical instrument 11 ... Keyboard 12, 13 ... Switch 14 ... Non-real-time code correction switch 15 ... Display part 21 ... CPU 22 ... ROM 23 ... RAM 24 ... Sound system 25 ... Switch group 26 ... Sound source part 27 ... Audio circuit 28 ... speakers

Claims (7)

A performance means for playing a song melody,
Real-time key determination means for determining the key of the song melody in real time based on the pitch history of the song melody being played,
A non-real-time tone correcting means for correcting the key determined by the real-time key determining means in non-real time after the melody performance by the playing means is completed;
An automatic tone correction device. The non-real time tone correction means includes
A confirmation key DB representing the final pitch of the song melody and the relationship between the final pitch and the key,
It is determined whether the key read by the final pitch from the determination key DB matches the key determined by the real-time key determination means. Real-time tone judgment means;
The automatic tone correction device according to claim 1, comprising: The automatic tone correction device further includes:
When it is determined that the non-real-time key determination means does not match, the key candidate having a close relationship with the final pitch of the melody of the song is read from the confirmation key DB, and the last key of the song is determined based on the key candidate. When a code is provisionally assigned to a section, a cadence determination means for determining whether or not the provisionally provided code is terminated in a form along cadence,
When it is determined by the cadence determination means that the key candidate has no contradiction from the beginning of the song, and when it is verified that there is no contradiction, the key candidate is determined as a definite key. And an automatic tone correction device according to claim 2. The automatic tone correction device further includes:
An assigning means for assigning the key candidate to a section having no contradiction of music and assigning the determined key to other sections when it is determined by the verification means that there is a contradiction;
It is determined whether the relationship between the key assigned to the beginning of the song and the key assigned to the other sections is a climax modulation, and if it is determined to be a climax modulation, both of the assigned tones are determined. Swell-up modulation judging means with a definite tone,
The automatic tone correction device according to claim 3. The automatic tone correction device further includes:
It has a relational tone and a borrowed chord DB indicating the relation between the relational tone and the borrowed chord,
If there is a section in the song to which a plurality of keys are assigned and an unverified key is assigned, and if there is a chord that is not in the relation key of the unverified key in that section, The automatic tone correction device according to claim 1, wherein the verification tone is a definite tone. Determining in real time the song melody based on the pitch history of the song melody being played;
Correcting the determined key in non-real time after the melody performance is completed;
An automatic tone correction method. Determining the key of the song melody in real time based on the pitch history of the song melody being played;
Correcting the determined key in a non-real time after the music melody performance is completed;
That causes a computer to execute.

Images