JP5712702B2 - Code detection apparatus and program for realizing code detection method - Google Patents

Description

  The present invention relates to a chord detection apparatus and chord detection method for detecting chords (chords) that are musically connected to inputted performance data.

  2. Description of the Related Art Chord detection devices that detect chords that are musically connected to input performance data are conventionally known.

  As such a chord detection device, after the performance data indicating the sequence of notes is inputted and stored, the performance data is divided into predetermined sections, and the above-mentioned division is performed from a chord progression database in which a large number of chord progressions are stored in advance. In addition, there is one that extracts chord progressions that match the performance data of a predetermined section (see, for example, Patent Document 1).

JP 2001-142462 A

  However, in the above-described conventional chord detection device, the performance data referred to extract the chord progression is input in advance, so that by looking at the flow of notes in the performance data before and after or in a wide range, Although it is possible to extract chord progressions that are musically connected to the performance data (notes that make up the performance data), the performance data being input by, for example, real-time performance is adopted as a reference object for extracting chord progressions. If an attempt is made to extract a chord progression that is musically connected to the performance data being input, it is unknown which sound will be input next, so the chord progression extracted from the chord progression database is being input. And may not fit.

  Also, when extracting any chord progression from the chord progression database with reference to the performance data being input, a chord detected at a certain timing, that is, a chord starting from a chord detected from the performance data input at a certain timing The progress is extracted from the chord progression database, but each time the chord output timing comes, the chord progression extracted in this way is sequentially output to detect the chord. In this case, the detected chord may be inconsistent with the performance data being input even though it is established as chord progression. Furthermore, in this case, since the chord progression starting from a certain chord stored in the chord progression database is used as it is, the types of chord progression patterns generated by the detected chords are reduced, and the same chord progression Is likely to appear, and as a result, the user may get bored.

  The present invention has been made paying attention to this point, and can detect a variety of chords that are musically connected to input performance data, particularly performance data input in real time. An object of the present invention is to provide a program for realizing the code detection device and the code detection method.

In order to achieve the above object, the chord detection apparatus according to claim 1 includes an input means for inputting performance data, an acquisition means for acquiring key information of the performance tune, a detection means for detecting a chord detection timing, When the detection timing of the code is detected by the detection means, a plurality of code candidates are extracted based on the key information acquired by the acquisition means, and input from the extracted plurality of code candidates by the input means a first selecting means for selecting one code based on the performance data, extraction means for extracting a solution destination code that may be chord progression solve destination including elected encoded by said first selecting means, each of the detection timing of one or a plurality of codes are detected by the detection means before reaching the resolution destination code extracted by said extraction means Nitsu Te, a generation unit based on the acquired key information, the elected out several extract the subsequent code candidates in the code, to produce a plurality of code candidate or al 1 or more code candidates that the extracted, when the detection timing of the code is detected by the pre-Symbol detection means, among the code candidates generated by said generating means, from the code candidates corresponding to the detected timing, the code based on the performance data that has been said input And a second selecting means for selecting one.

The code detection device according to claim 2 is the code detection device according to claim 1, wherein, when a plurality of code candidate groups are generated by the generation unit, the code selected by the second selection unit is next. proceeds reasonably code that is a candidate to lead the code, when in encoding candidate group corresponding to the detection timing after the detection timing of the selected code, the deletion means to delete the code candidate with the unreasonable Furthermore, it is characterized by having.

  The code detection device according to claim 3 is the code detection device according to claim 1 or 2, wherein when the generation unit generates the code candidate group, the code detection device is one before the plurality of code candidates constituting the code candidate group. Further, the present invention is characterized by further comprising a detour code candidate adding means for extracting one or a plurality of code candidates that can be detoured and adding to the code candidate group.

  The code detection device according to claim 4 is the code detection device according to any one of claims 1 to 3, wherein the selected code is further selected when the code selected by the second selection means is the solution destination code. Is determined to be the start point of the extended dominant, and if the selected code can be the start point of the extended dominant as a result of the determination, the detection means detects the detection timing of the next code. The first selecting means further includes an extended dominant candidate adding means for extracting one or a plurality of code candidates to be extended dominants and adding them to a plurality of code candidates extracted by the first selection means.

  The code detection device according to claim 5 is the code detection device according to claim 4, wherein the code selected by the second selection means is not the solution destination code but an extended dominant code. The generation means includes at least one extended dominant code candidate subsequent to the selected extended dominant code in the code candidate group corresponding to the detection timing next to the detection timing of the code, and the selection In this case, one or more code candidates to be solved by the generated code are extracted and included.

  In order to achieve the above object, the program according to claim 6 can be realized by the same technical idea as that of claim 1.

According to the first or sixth aspect of the invention, when the code detection timing is detected, a plurality of code candidates are extracted based on the acquired key information, and input from the extracted plurality of code candidates. One chord is selected on the basis of the selected performance data, a solution-destination chord that can be a solution to a chord progression including the selected chord is extracted, and the detection is performed until the extracted solution-destination chord is reached. About 1 to respective detection timings of the plurality of codes, based on the obtained key information, the selected code followed code candidate is issued multiple extraction, a plurality of code candidate or al 1 issued extract a plurality of code candidates is generated, when the detection timing of the code is detected, among the generated code candidates, code candidates corresponding to the detection timing Since one chord is selected based on the inputted performance data, that is, a progress code from the selected chord to the solution destination chord is also selected (detected), so a variety of chords Can be detected. Since the chord is detected based on the input performance data, the detected chord has a musical connection to the performance data.

According to the second aspect of the present invention, when a plurality of code candidate groups are generated by the generation unit, a code candidate that is impossible to connect as a code that advances next to the selected code is selected. code when in the corresponding code candidate group to the detection timing after the detection timing of, the code candidate with the irrational is deleted, as the code that follows to the selected code, reasonable in connection The code can be detected.

  According to the third aspect of the present invention, one or more code candidates that can be detoured are extracted and added to the code candidate group immediately before the plurality of code candidates constituting the code candidate group. It is possible to detect codes rich in.

  According to the fourth aspect of the present invention, when the selected code is the solution-destination code and can be the starting point of the extended dominant, one or more code candidates to be the extended dominant are also extracted. Therefore, a variety of codes can be detected.

  According to the fifth aspect of the present invention, when the selected code is not the resolution target code but an extended dominant code, a code candidate corresponding to a detection timing next to the detection timing of the code The group includes at least extracted candidate codes 1 to plural of the extended dominant following the selected extended dominant code and code candidates 1 to plural to be resolved by the selected code. A variety of codes can be detected.

It is a block diagram which shows schematic structure of the electronic musical instrument to which the chord detection apparatus which concerns on one embodiment of this invention is applied. It is a figure which shows the example of a setting of a code detection timing. It is a figure which shows an example of the correspondence table of the code | cord | chord function and degree name in a measure. It is a figure which shows an example of the correspondence table of the function of the code | cord | chord in a minor tone, and a degree name. It is a figure which shows an example of the data format of code data. It is a flowchart which shows the procedure of the performance process using an automatic accompaniment which the electronic musical instrument of FIG. 1, especially CPU performs. It is a flowchart which shows the procedure of the continuation of the performance process using the automatic accompaniment of FIG. 5A. It is a flowchart which shows the detailed procedure of the note event process in FIG. 5A. It is a flowchart which shows the detailed procedure of the code | cord | chord detection process in FIG. 5B. It is a flowchart which shows the detailed procedure of the basic code candidate extraction process in FIG. It is a flowchart which shows the procedure of the continuation of the basic code candidate extraction process of FIG. 8A. It is a flowchart which shows the detailed procedure of the extraction process of the degree name candidate with the function which can advance to the next in FIG. 8B. It is a flowchart which shows the procedure of the continuation of the extraction process of FIG. 9A. It is a flowchart which shows the detailed procedure of the additional code candidate extraction process in FIG. It is a flowchart which shows the detailed procedure of the code detection process for chord progression formation in FIG. It is a flowchart which shows the procedure of the continuation of the chord detection process code detection process of FIG. 11A. It is a flowchart which shows the procedure of the continuation of the chord detection process code detection process of FIG. 11A. It is a flowchart which shows the detailed procedure of the extraction process of the code candidate which advances to the interruption type in FIG. 11A. It is a flowchart which shows the procedure of the continuation of the extraction process of FIG. 12A.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a schematic configuration of an electronic musical instrument to which a chord detection device according to an embodiment of the present invention is applied.

  As shown in the figure, the electronic musical instrument of the present embodiment includes a performance operator 1 including a keyboard for inputting performance data including pitch information in accordance with a user's performance operation, and a plurality of information for inputting various information. A setting operation element 2 including a switch and a rotary encoder, a detection circuit 3 for detecting an operation state of the performance operation element 1, a detection circuit 4 for detecting an operation state of the setting operation element 2, and a CPU 5 for controlling the entire apparatus. A ROM 6 for storing a control program executed by the CPU 5 and various table data, a RAM 7 for temporarily storing performance data, various input information, calculation results, and the like, and clocking interrupt times and various times in timer interrupt processing. And an automatic accompaniment device 9 for generating performance data (for generating accompaniment sounds) based on chord information supplied by the CPU 5 as will be described later. Display device 10 for displaying various information, for example, LCD (liquid crystal display) and LED (light emitting diode), etc., and storage for storing various application programs including the control program, various music data, various data, etc. The device 11 is connected to an external device 100 such as an external MIDI (musical instrument digital interface) device, and a communication interface (I / F) 12 for transmitting / receiving data to / from the external device 100 and the performance operator 1 are input. A sound source circuit 13 for converting performance data, performance data generated by the automatic accompaniment device 9 and the like into a musical sound signal, an effect circuit 14 for applying various effects to the musical sound signal from the sound source circuit 13, and the effect Converts the musical sound signal from the circuit 14 into sound, for example, a DAC (digital-to-analog converter), an amplifier, a speaker It is constituted by a sound system 15 and the like.

  The above components 3 to 14 are connected to each other via a bus 16, the timer 5 is connected to the CPU 5 and the automatic accompaniment device 9, the external device 100 is connected to the communication I / F 12, and the sound source circuit 13 is connected to the sound source circuit 13. An effect circuit 14 is connected, and a sound system 15 is connected to the effect circuit 14.

  The automatic accompaniment device 9 is realized, for example, by causing the CPU 5 to execute sequencer software stored in advance in the ROM 6. As described above, the automatic accompaniment device 9 generates performance data based on the supplied chord information, and generates the generated performance data. By supplying the sound source circuit 13, an accompaniment sound is generated. The automatic accompaniment device 9 also has a function of generating performance data by reproducing any of the accompaniment style data selected by the user from various accompaniment style data stored in advance in the ROM 6, for example. When performing this function, the automatic accompaniment device 9 reproduces the accompaniment style data based on the time information supplied from the timer 8. Since the features of the present invention are not in the configuration and operation of the automatic accompaniment device 9, further description of the configuration and operation of the automatic accompaniment device 9 will not be given.

  The storage device 11 is, for example, a storage medium such as a flexible disk (FD), a hard disk (HD), a CD-ROM, a DVD (digital versatile disc), a magneto-optical disk (MO), and a semiconductor memory, and its driving device. The storage medium may be detachable from the drive device, or the storage device 11 itself may be detachable from the electronic musical instrument of the present embodiment. Alternatively, neither the storage medium nor the storage device 11 may be detachable. As described above, the storage device 11 (the storage medium) can also store the control program executed by the CPU 5. If the control program is not stored in the ROM 6, the control program is stored in the storage device 11. By reading it into the RAM 7, it is possible to cause the CPU 5 to perform the same operation as when the control program is stored in the ROM 6. In this way, control programs can be easily added and upgraded.

  Although the external device 100 is connected to the communication I / F 12 in the illustrated example, the present invention is not limited to this. For example, a server computer is connected via a communication network such as a LAN (local area network), the Internet, or a telephone line. You may make it do. In this case, if the above programs and various parameters are not stored in the storage device 11, the communication I / F 12 is used for downloading the programs and parameters from the server computer. The electronic musical instrument serving as a client transmits a command for requesting downloading of a program and parameters to the server computer via the communication I / F 12 and the communication network. Upon receiving this command, the server computer distributes the requested program or parameter to the electronic musical instrument via the communication network, and the electronic musical instrument receives the program or parameter via the communication I / F 12 and stores it. 11 completes the download.

  The electronic musical instrument of the present embodiment is constructed on an electronic keyboard instrument as can be seen from the above configuration, but is not limited to this, and is constructed on a general-purpose personal computer to which a keyboard is externally connected. May be.

  A control process executed by the electronic musical instrument configured as described above will be described first with reference to FIGS. 2 to 4 and then in detail with reference to FIGS.

  FIG. 2 is a diagram showing an example of setting chord detection timing. In the figure, 4/4 time tunes are selected as performance songs, the first and third beats are set as chord detection reference positions, An example is shown in which a period from 250 ms before to 50 ms after the code detection reference position is set as the code detection timing. In the illustrated example, the chord detection timing is not drawn for the first beat, but if the chord detection timing is drawn for one of the beat positions, the chord detection timing is similarly applied to the other beat position. This is because it is easy to see that is set. Although details will be described later, the user can freely select the code detection timing at which position, how much time width, and how many (or how often) the code detection timing is set.

In the electronic musical instrument of the present embodiment, when a user performs using the keyboard, the performance data input in real time is supplied to the tone generator circuit 13 as it is for sound generation, and the performance data input in real time is generated. By detecting a chord with musical connection and supplying it to the automatic accompaniment device 9, the accompaniment sound is also pronounced. The chord detection timing period (time width) is a period during which performance data to be referenced for detecting chords is input. That is, only performance data input within the chord detection timing period is referred to when detecting a chord for generating an accompaniment sound. The code detection starts with
(A) Basic code candidate extraction processing (B) Additional code candidate extraction processing is used to extract code candidates as detection candidates.

  In the extraction process of (A) above, the code function and degree name correspondence table in FIG. 3A (in the case of major tone) and FIG. 3B (in the case of minor tone) is referred to the one corresponding to the current tone, Extract all functions that can proceed to the next detected code function. Then, referring to the correspondence table, all the names having the extracted functions belonging to the diatonic scale code and those belonging to other than the diatonic scale code are extracted. Further, the code name corresponding to each extracted degree name is detected based on the current key, and code data including each detected code name and the degree name corresponding to each code name is generated, It is registered in a code list CList (not shown) secured in the RAM 7. However, when a predetermined condition is satisfied, a sample belonging to the diatonic scale code may not be extracted, but will not be further described here.

  FIG. 4 is a diagram illustrating an example of the data format of the code data. In the illustrated example, the code data includes a code name, a degree name, and information data, and the information data includes resolution code information, this code (solution Degree information (hereinafter abbreviated as “degree information”) and flag information up to the first code). However, it does not mean that the information data is always composed of the three types of information. That is, depending on the code data, there is resolution destination code information but no flag information, there is no resolution destination code information, but there is flag information, and there is neither resolution destination code information nor flag information. However, when there is solution destination code information, there is always degree information, and when there is no solution destination code information, there is always no degree information. None of the code data registered in the code list CList by the extraction process (A) includes information data. The meaning contents of the solution code information, the degree information, and the flag information will be described in the following description of the extraction process (B).

The extraction process (B) above is
(B1) Chord candidate extraction process for forming an interrupted chord progression (B2) Chord candidate extraction process for forming an extended chord progression. And in the present embodiment, the interrupt type chord progression is
(B1a) Major secondary dominant / two-five → Predetermined code of diatonic scale code (major system solution target code)
(B1b) Minor secondary dominant / two-five → Predetermined code of diatonic scale code (minor resolution target code)
(B1c) Double dominant → Diatonic scale code dominant 7th code (dominant 7th system solution code)
In the present embodiment, the extended chord progression is
(B2a) Extended dominant → diatonic scale code (B2b) Extended dominant → extended dominant → diatonic scale code.

  In the extraction process of (B1), first, the code data registered in the code list CList by the extraction process of (A) becomes the major system solution target code of (B1a), the minor of (B1b) The system solution destination code and the (B1c) dominant 7th system solution code are extracted. Here, the solution destination code is a stable chord progression, that is, a chord progression to be a skeleton formed by connecting chord data extracted by the extraction process (A) (hereinafter referred to as “skeleton chord”). When a different chord progression (that is, the above-mentioned interrupted chord progression) is interrupted, the chord is stabilized again in harmony with this extraordinary chord progression.

  Next, each code that is the starting point of the interrupt type code progression that proceeds to each extracted solution destination code is detected, and each of the code data includes a code name and a degree name corresponding to each detected code, The code data information data storing “resolution code information”, “degree information” and “flag information” is generated and added to the code list CList. Here, the “resolved code information” is information of the resolved code for the code data, and “resolved code information” stores “the frequency of the root of the resolved code” in the present embodiment. The The “degree information” is information indicating the transition of the degree name (code progression) up to the solution destination code. In the present embodiment, the “degree information” is “first degree name of the second degree”. "Degree name of the frequency of the route of the solution code" (however, there may be no "second degree name") is stored. “Flag information” indicates the type of the code data, and “Flag information” stores “interrupt” indicating a code for forming an interrupt type code progression.

  In the extraction process of (B2), each chord that is a starting point of the extended chord progression is detected, and each chord data includes a chord name and a degree name corresponding to each chord detected, and information on each chord data Data in which “flag information” is stored is generated and added to the code list CList. At this time, either “Extended V7 / V7sus4” or “Extended ♭ II7” is stored in the “flag information” according to the type of the detected code.

  When the extraction process of (B) is executed, the extraction process of (B1) is always executed, but the extraction process of (B2) is performed when a predetermined condition is satisfied (EP (extended plus described later) ) When the flag is set). In which case the EP flag is set / reset will be described later in detail of the control process.

  Thus, one chord data is selected from the chord data registered in the chord list CList based on the input performance data, and the chord data thus selected is detected code Chord (not shown) secured in the RAM 7. ).

If the detection code Chord is a basic code extracted by the extraction process of (A), the extraction process of (A) and (B) is executed at the next code detection timing, but the detection code Chord is If it is an additional code extracted by the extraction process of (B), at the next code detection timing,
(C) The chord progression forming chord detection process is executed.

  In this (C) code progression formation code detection process, the code data detected last time stored in the start code SChord (not shown) secured in the RAM 7 is used as the start point, and the type of the code data (the code data Depending on whether the flag information in the information data is “Interrupt”, “Extended V7 / V7sus4” or “Extended ♭ II7”), the code data to form either interrupt type code progression or extended type code progression One is selected, and the selected code data is stored in the detection code Chord.

  This (C) chord progression forming chord detection process is subsequently executed until the chord data serving as the end point of the interrupt chord progression or the extended chord progression is detected, and the chord data serving as the end point is detected. Then, the flag information in the information data of the detection code Chord is set to “End”. Thereby, at the next code detection timing, the extraction processes (A) and (B) are executed again.

  During the detection of the code data for forming the interrupt type chord progression, the code data that becomes the “detour” (the “departure” is stored in the flag information in the information data) can also be detected. Yes.

  In addition, the detection code Chord detected at each chord detection timing is output to the automatic accompaniment device 9 and pronounced.

  The chord detection process (from “control process” to be described later) from the start of the extraction process of (A) and (B) or the detection process of (C) until a single chord is selected and output to the automatic accompaniment device 9 In the “detail”, “corresponding to (5) chord detection timing processing” is executed on the first beat and the third beat in the example of FIG. 2, but more strictly, the chord of each beat The start of execution is instructed at the end of the detection timing period, that is, at time T3. This is because the performance data detection period referred to in the chord detection process is the entire period of the chord detection timing, that is, the periods T1 to T3.

  As described above, in the electronic musical instrument of the present embodiment, chord candidates that are musically connected are extracted as chords to proceed next, and chords that match the performance data input in real time are extracted from the extracted chord candidates. Since it has been detected, it is possible to detect chords that are musically connected to real-time performances. The detected chord is always in harmony with the performance data inputted in real time.

  All chord progressions used in tonal music include chord progressions that interrupt skeletal chord progressions (for example, secondary progressions), in addition to skeletal chord progressions that are classified as functional progressions such as tonic to dominant and non-functional progressions such as cliche. Dominant and its two-five, extended dominant). The chord progression that interrupts the skeleton cannot be a candidate from the viewpoint of a chord that can proceed to a chord at a certain point in time. The reason is that a code that passes until reaching a certain solution destination code is interrupted. In the present embodiment, when a plurality of chord candidates for forming a skeleton chord progression are extracted, a chord that can be a solution-destination code is extracted from the plurality of chord candidates and resolved to each extracted solution-destination code. Since chord candidates and chord candidates that have passed until they are resolved to the respective chords are detected and added to the chord candidates, out of all chord progressions used in tonal music, interrupt the skeletal chord progression It is possible to add chord candidates and detect chords that form chord progressions.

  When a code (candidate) that resolves to one of the code candidates extracted at a certain code detection timing is detected, at the next code detection timing, the code that was detected at the immediately preceding code detection timing is resolved or the code The possibility that a surrogate code is detected is extremely high. In the present embodiment, chord candidates that are very likely to be detected are all listed, so that chord progressions that are more flexible and variety than conventional methods using chord progression databases can be generated. Further, in the present embodiment, since there is a possibility that a plurality of chords will elapse until the solution chord is detected, it is possible to generate a more varied chord progression.

  Furthermore, in the present embodiment, it is possible to detect chord progressions that appear during the performance, such as solutions from secondary dominants, two-five progressions, and continuation of dominant 7th chords from there, and it is not determined how long they will continue.

  Further, in the present embodiment, the chord data detection process for forming the skeleton chord progression and the chord data detection process for forming the chord progression interrupting the skeleton chord progression are performed independently. Therefore, when a chord that begins the chord progression that interrupts the skeleton chord progression is detected, the chord data is detected only in the latter detection process, so that the chord progression of the interrupting chord continues thereafter. Data detection can be controlled. Thereby, even if the inputted performance data is the same, the detected chord data may change depending on which detection process is used to detect the chord data. As a result, it is possible to obtain a chord progression that is not monotonous while considering musical connections.

  Furthermore, since both the detection processes are automatically switched to one of the detection processes according to the performance flow, the user does not have to perform a conscious performance in order to detect chords, but concentrates on a free performance. be able to.

  Also, in this embodiment, the possibility of proceeding to an extended dominant (continuous V7 / V7sus4 / continuous II7) is also considered, and one of the chords that may advance to the next in the chord progression that interrupts the skeleton. Code candidates that may detour before are also extracted. As a result, when a chord progression that interrupts the skeletal chord progression occurs, it is possible to deal with more chord progressions that may occur thereafter.

  Next, this control process will be described in detail.

  5A and 5B are flowcharts showing a procedure of performance processing using automatic accompaniment executed by the electronic musical instrument of the present embodiment, particularly the CPU 5. In the electronic musical instrument according to the present embodiment, the performance data input using the performance operator 1 without operating the automatic accompaniment device 9 is used as a performance mode when the user performs a real-time performance using the performance operator 1. The performance data generated by the automatic accompaniment device 9 together with the musical performance corresponding to the performance data input using the performance operator 1 by operating the automatic accompaniment device 9 by generating the first performance mode for generating the corresponding musical sound. There are two types of performance modes: a second performance mode that also generates musical sounds (accompaniment sounds) according to the sound. The normal performance mode, that is, the performance mode that is selected first when the electronic musical instrument of the present embodiment is turned on is the first performance mode, so to shift to the second performance mode, A predetermined transition instruction from the user is required. The “performance process using automatic accompaniment” is a process that is started when this transition instruction is issued. In this embodiment, all the features of the present invention are included in the “performance processing using automatic accompaniment”, and the features of the present invention are explained using this “performance processing using automatic accompaniment”. , “Performance processing without automatic accompaniment” will not be described. Therefore, “performance processing using automatic accompaniment” is hereinafter abbreviated as “performance processing”. However, all the features of the present invention are included in the “performance processing that does not use automatic accompaniment”, and the features of the present invention can be described using this “performance processing that does not use automatic accompaniment” (in this case, Since no accompaniment sound is generated), the feature of the present invention has been described using “performance processing using automatic accompaniment” for convenience only.

This performance process is mainly
(1) Start-up processing (steps S1 to S4)
(2) Automatic accompaniment start process (step S7)
(3) Automatic accompaniment stop process (step S9)
(4) Note event processing (step S12)
(5) Processing at code detection timing (steps S14 to S19 in FIG. 5B)
It is constituted by.

  When the performance process is started, after the start-up process of (1) is executed once, it enters a standby state until an automatic accompaniment start instruction is issued (steps S6 → S5 → S6), and an automatic accompaniment start instruction is issued. If there is, the automatic accompaniment start process of (2) is executed, and subsequently, the processes of (4) and (5) are executed as appropriate. The processes (4) and (5) are repeatedly executed as appropriate until an instruction to stop automatic accompaniment (step S8) or an instruction to return to the first performance mode (step S5). If there is an automatic accompaniment stop instruction, it is determined whether the accompaniment style or the like has been changed after the automatic accompaniment stop process (3) has been executed (step S10). On the other hand, the process returns to the start-up process of (1) above, but returns to the standby state when there is no setting change. If there is an instruction to return to the first performance mode, the performance process ends (step S5 → end).

The startup process of (1) is as follows.
(11) Setting process before start of performance (step S1)
(12) Code detection timing and various rule setting processing (step S2)
(13) Initialization process (step S3)
(14) Calculation setting process of start point and end point of code detection timing (step S4)
It is constituted by.

  When the process proceeds to the (1) start-up process, the CPU 5 first executes the setting process before the start of the performance (11). In the setting process before the start of the performance (11), the CPU 5 sets items such as the performance tempo, time signature, accompaniment style, volume value, and performance tone color. Each item inquires about the value or type to be set by the user, and the value or type input or selected by the user may be set accordingly, or it may be set by default or previously set May be set as they are or with some changes.

  Next, the CPU 5 advances the processing to (12) code detection timing and various rule setting processing. In this (12) code detection timing and various rule setting processing, the CPU 5 sets a code detection timing setting rule and other various rules.

  Specifically, the code detection timing setting rule is whether a time width is provided as a timing for detecting a code, whether a time width is provided without a time width, or if a time width is provided. With the time point as a reference position, how much time width is provided before and after that, or if it is time point, what time point is set. As an example of providing the time width, as shown in FIG. 2, the beat detection position is the chord detection reference position every two beats (for example, the first and third beats when 4/4 time is selected). As a period from 250 ms before to 50 ms later. In the illustrated example, “ms” is adopted as a unit for setting the time width, but it is not limited to this, and may be a note length. In addition, the chord detection reference position is not limited to a beat position every two beats, and may be a beat position for every beat. When the tempo changes, for example, the chord detection reference position changes from every two beats to every beat. Alternatively, the position is not limited to a beat, and may be a predetermined position (for example, the start position) for each measure. Further, where to set the chord detection reference position may be determined by the tempo value and the accompaniment style. On the other hand, as an example of the time point, various code detection reference positions exemplified in the case of providing a time width, that is, a beat position for each predetermined beat, a predetermined position for each measure, or a user is included in the setting operator 2, for example. It is conceivable that the predetermined operating element is operated or the user operates the predetermined operating element within a predetermined beat. In this embodiment, it is assumed that the rule shown in the example of FIG. 2 is selected and set as the code detection timing setting rule.

  Examples of the various rules include (R1) tone information detection rules and (R2) chord selection rules.

  As the detection rule of the key information (R1), for example, a rule that the key information is inquired to the user before the performance and the key information input by the user is detected (acquired) in response thereto, or the user's performance operation is used. There may be a rule of analyzing performance data input in response to detecting key information at any time. In the present embodiment, it is assumed that the key information is input or detected as the main note name + measure / minor. As a method for analyzing performance data and detecting key information, a known method may be used. When a method for detecting key information at any time is adopted, every time a chord is selected (see step S47 in FIG. 7 to be described later), the key information detected at that time is used as the selected chord. It is good to store it in association.

As a rule for selecting the code of (R2), for example,
(R2a) Prioritize each code in advance (for example, the tonic code of the diatonic scale code is made higher), and the rule (R2b) of selecting the highest code among the candidate codes is referred (R2b) A rule that selects the chord with the highest degree of match (complete match, match 3 or more, 2 match, etc.) between all note names of the target performance data and the constituent sounds of the candidate chords. be able to. Further, an additional condition may be that the root of each chord of the candidate chord is included in the performance data to be referenced.

  Also, (12) In the chord detection timing and various rule setting processing, it is also set whether or not the dominant (D) → subdominant minor (SM) progression (interrupted cadence) is allowed in normal mode. It can be done. Here, “Allow at normal time” means extraction of a chord progression (a degree name having a chord function that forms an interrupted cadence in a major tone) even during the basic chord candidate extraction process of (A). Means to allow. If “Allow at normal time” is set, the function is “D: dominant” in the correspondence table of chord functions and degree names in the major tone of FIG. "SM" is added to "Function", and "Telefunction that can proceed to next" has three types of "T", "D", and "SM", while "Unauthorized during normal use" is set In this case, “SM” is not added to the “function that can proceed to the next”, and there are two types of “T” and “D”. In FIG. 3A, “SM” is shown in parentheses because “SM” can be selected as a function and not selected as a function. is there. In the present embodiment, the correspondence table in FIG. 3A is stored in advance in the ROM 6 and is read out and referred to when necessary, but the correspondence table depends on whether “Allow normal” is set. It is cumbersome to decide whether or not to use some data ("SM"), so the correspondence table when "Allow at normal time" is set and "Not allowed at normal time" are set A correspondence table for each case may be created and stored separately in advance, and the necessary one may be read and referenced as necessary.

  In the setting process of (12) chord detection timing and various rules, each rule inquires about the contents of the rule to be set by the user, and the user responds accordingly. The input or selected content may be set, or the default setting or the previous setting may be set as it is or after being partially changed.

Next, the CPU 5 advances the process to the initialization process (13). In this (13) initialization process, the CPU 5 initializes each next area and each flag secured on the RAM 7. That is,
Note event list NList: a list for registering note event information (pitch + input timing) corresponding to a note-on event input within the chord detection timing period;
Code detection timing start point sTime: an area for storing the code detection timing start point;
Code detection timing end point eTime: an area for storing the code detection timing end point;
Key information Key: an area for storing key information detected according to the set key detection rule;
Code list CList: a list for registering code data candidates to be detected;
Detection code Chord: an area for storing one code data selected from the code list CList based on the note event information registered in the note event list NList as a detection code;
Start code SChord: In some cases, during generation of the skeleton code progression, the interrupt type chord progression or extended type chord progression that interrupts the skeleton chord chord progression is generated. In this case, the interrupt type chord progression or extended chord progression is generated. An area for storing chord data that will be the start (start point) of progress;
Next code list NCList: This is a list for registering candidates of code data to proceed next from the current detection code Chord when generating an interrupt type code progression;
Output result list RList: a list for registering key information Key and detection code Chord;
Previous degree name LDN: an area for storing the degree name of the previous detection code Chord;
Previous function LF: an area for storing a function corresponding to the previous detection code Chord;
While clearing each area of
Normal code detection flag: set (“1”) when a normal code (in this embodiment, code data forming a skeleton code progression) is detected, and reset when a code other than a normal code is detected ( The flag to be “0”);
EP flag: This is set when the code data candidate to be an extended dominant is extracted and added to the code list CList in the additional code candidate extraction process of (B), and when the code data candidate is not extracted. The flag to be reset;
Set the former flag and reset the latter flag.

  Next, the CPU 5 advances the processing to the calculation setting processing of the start point and end point of the code detection timing (14). In this (14) code detection timing start point and end point calculation setting process, the CPU 5 calculates the start point and end point of the first code detection timing in accordance with the set code detection timing setting rule, and the calculated start point is The start point sTime is stored (set), and the calculated end point is stored (set) in the end point eTime. However, in this embodiment, as the chord detection timing, the beat position of the first beat of the song is taken as the chord detection reference position, and 250 ms before the beat position is taken as the start point of the chord detection timing. Since it starts from the first beat, it is meaningless to calculate the position before the start position and set it as the start point sTime when the music starts. Therefore, in this case, it is preferable to set the beat position of the first beat, that is, the chord detection reference position, as the start point sTime. However, in this case, as a rule, even if the position 250 ms before the beat position of the first beat is calculated and set as the start point sTime, the control process is only started from the time after the set start point. Then, no problem occurs in the subsequent control processing.

  Thus, (1) when the start-up process is executed once, the CPU 5 waits for an automatic accompaniment start instruction from the user, and when an automatic accompaniment start instruction is given, the process proceeds to the automatic accompaniment of (2). The process proceeds to the start process (steps S6 → S7). In this (2) automatic accompaniment start process, the CPU 5 starts the timer 8 and starts the time measurement by the timer 8. The time measured by the timer 8 is also supplied to the automatic accompaniment device 9 as described above. Further, in this performance processing, the automatic accompaniment device 9 is in the operating state as described above, and therefore the accompaniment style data is selected and set. Then, the automatic accompaniment apparatus 9 reproduces the accompaniment style data based on the time measured (time information) supplied from the timer 8 independently of the performance process.

  When the note-on event is accepted until the user instructs to stop the automatic accompaniment, the CPU 5 executes the note event process (4) (steps S8 → S11 → S12), and the timer 8 When the detection timing (the end point eTime) is counted, the code detection timing processing (5) is executed (steps S13 to S14 in FIG. 5B).

  On the other hand, when an automatic accompaniment stop instruction is given by the user, the CPU 5 advances the process to the automatic accompaniment stop process of (3) (steps S8 → S9). In this (3) automatic accompaniment stop process, the CPU 5 stops the timer 8. Accordingly, naturally, the reproduction of the accompaniment style data by the automatic accompaniment device 9 is stopped.

  FIG. 6 is a flowchart showing a detailed procedure of the note event processing (4).

As shown in the figure, this note event processing
(41) When the timer 8 is measuring the time within the code detection timing period, that is, the time from the start point sTime to the end point eTime (41a) Processing when a note-on event is received (step S33) , S34)
(41b) Processing when a note-off event is received (steps S35 and S36)
(42) When the timer 8 is measuring a time outside the code detection timing period (42a) Processing when a note-on event is received (step S38)
(42b) Processing when a note-off event is received (step S39)
It is constituted by.

  The processing of (41a) is the same as the processing of (42a), that is, the note event corresponding to the note-on event in the sound generation processing performed by outputting the received note-on event to the tone generator circuit 13. The processing of adding information (pitch + input timing) to the note event list NList is added, and the processing of (41b) is the processing of (42b), that is, the received note-off event is a tone generator circuit. 13 is added with a process of deleting the note event information corresponding to the note-off event from the note event list NList in addition to the mute process performed by outputting to the note 13.

  In the code detection timing processing (5), first, the CPU 5 executes code detection processing (step S14). Details of the code detection process will be described later based on the flowchart of FIG. 7. When the code detection process is executed, one code data is detected (selected) and stored in the detection code Chord. Next, the CPU 5 reads out the code name from the detection code Chord and outputs it to the automatic accompaniment device 9 (step S15).

  Further, if the normal code detection flag is set, the CPU 5 adds the key information and the detection code Chord to the end of the output result list RList (step S16 → S17), and sets the code detection timing setting contents (setting rule). Based on this, the start point and end point of the next chord detection period are calculated, the start point sTime and the end point eTime are updated (step S18), and the note event list NList is initialized (cleared) (step S19). If the normal code detection flag is in the reset state, the CPU 5 skips step S17 and proceeds to the next step S18. When the normal code detection flag is in the reset state, as described above, it means that the interrupt type code progress or the extended type code progress is being generated. The chord data (and key information) for forming the interrupt type chord progression or the extended chord progression is not additionally registered in the output result list RList. In this case, the CPU 5 performs step S17. The additional registration process is not executed.

  FIG. 7 is a flowchart showing a detailed procedure of the code detection process in step S14.

This code detection process mainly consists of
(51) Normal code detection processing (steps S42 to S48)
(52) Transition process to non-ordinary code detection process (steps S50 to S52)
(53) Unusual code detection processing (steps S53, S56 to S58)
(54) Transition process to normal code detection process (step S55)
The (51) normal code detection process includes the (A) basic code candidate extraction process and the (B) additional code candidate extraction process, and the (53) non-normal code detection process includes The (C) chord progression forming chord detection process is included.

  The processing sequence related to normal code detection, that is, the processing sequence of (51) and (52) above is executed when the normal code detection flag is set (step S41 → S42), That is, the processing sequence of (53) and (54) is executed when the normal code detection flag is in the reset state (steps S41 → S53).

  In the (51) normal code detection process, first, the CPU 5 acquires key information according to the key detection rule set by the code detection timing and the various rule setting process of (12) (step S42), and acquires the acquired key code. Information is stored in the key information Key (step S43), and the code list CList is initialized (cleared) (step S44).

  Next, the CPU 5 executes the (A) basic code candidate extraction process (details will be described later with reference to FIGS. 8A and 8B) (step S45) and the (B) additional code candidate extraction process (details thereof). Executes (described later using FIG. 10) (step S46). When both the extraction processes are executed, the chord list CList includes a plurality of chord candidates for forming a skeleton chord progression and a plurality of start points of non-skeletal chord progressions (interrupt type chord progression and extended chord progression). Code candidates are registered.

  Then, based on the note event list NList and the chord list CList, the CPU 5 refers to the note event information (pitch) registered in the note event list NList, and is registered in the chord list CList. One code data is selected from a plurality of code candidates (code data) (step S47), and the selected code data is stored in the detection code Chord (step S48). Note that “code selection” in step S47 is the (R2) code selection rule set in the code detection timing and various rule setting processing in (12), that is, in the present embodiment, (R2a ) Or (R2b). This situation is not limited to the above step S47, but also applies to other steps that perform the same processing as step S47, for example, steps S126, S145, and S150 in FIG.

  Next, the CPU 5 determines whether or not there is flag information in the information data of the detection code Chord (step S49). If there is flag information, one of “interrupt”, “extended V7 / V7sus4”, “extended ♭ II7”, “detour”, and “End” is stored in the flag information, but the process is step S49. When there is flag information in the information data of the detection code Chord when proceeding to, the flag information is one of “interrupt”, “extended V7 / V7sus4” and “extended ♭ II7” There is no “detour” or “End”. This is because the setting of “detour” and “End” is set only in the processing sequence related to the abnormal code detection, more specifically, (C) chord progression formation included in the (53) abnormal code detection processing. This is because the “detour” and “End” are not set after the transition to the processing sequence related to the normal code detection from this processing sequence.

  As a result of the determination in the step S49, when there is flag information in the information data of the detection code Chord, the CPU 5 executes the process of shifting to the non-normal code detection process of (52) (steps S49 → S50). On the other hand, when there is no flag information in the information data of the detection code Chord, the CPU 5 ends this code detection process without executing the transition process (step S49 → return).

  (52) In the transition process to the non-ordinary code detection process, the CPU 5 resets the normal code detection flag (step S50), initializes (clears) the start code SChord, and starts each information of the detection code Chord. The code is stored in the code SWord (step S51), and the code list CList and the next code list NCList are both initialized (cleared) (step S52).

  In the non-ordinary chord detection process (53), first, the CPU 5 executes a chord progression forming chord detection process (step S53). When the chord progression forming chord detection process is executed, one chord data is selected and stored in the chord code Chord. Next, the CPU 5 determines whether or not the flag information in the information data of the detection code Chord is “End” (step S54). In the processing sequence related to non-ordinary chord detection, the chord data for forming the non-skeletal chord progression is basically detected. However, it is necessary to return to the chord data detection for forming the skeletal chord progression. is there. The chord data whose flag information is “End” is chord data that is the end point of the non-skeletal chord progression. That is, the determination in step S54 determines whether or not the detection of the chord data for forming the non-skeletal chord progression is to be terminated. As a result of this determination, when the detection of the chord data for forming the non-skeletal chord progression is continued (flag information ≠ “End”), the CPU 5 determines that the flag information is “Extended V7 / V7sus4” or “Extended It is determined whether it falls under any of “II7” (step S56). In the determination in step S56, it is determined whether to generate the extended chord progression or the interrupt chord progression as the non-skeleton chord progression. As a result of this determination, when generating an extended chord progression, the CPU 5 stores each piece of information of the detection code Chord in the start code SChord (step S57) and initializes the chord list CList (step S58). When generating the type code progression, the CPU 5 immediately ends the code detection process (step S56 → return).

  As a result of the determination in step S54, when the detection of the chord data for forming the non-skeletal chord progression is to be ended (flag information = “End”), the CPU 5 performs the normal chord detection process of (54). The normal code detection flag is set (step S55).

  8A and 8B are flowcharts showing the detailed procedure of the basic code candidate extraction process in step S45.

This basic code candidate extraction process mainly includes
(61) Start-up processing (steps S61 to S63)
(62) Preprocessing of next extraction processing (63) (steps S64 to S72)
(63) Degree name candidate extraction process having a function capable of proceeding to the next step (step S75 in FIG. 8B)
It is constituted by.

When this basic code candidate extraction process is activated, the CPU 5 first proceeds to the activation process (61). In this (61) startup process, the CPU 5
(61a) When the note event list NList is empty, that is, when no note event information is recorded in the note event list NList: the basic code candidate extraction process is immediately terminated (step S61 → return);
(61b) When the note event list NList is not empty: and (61b1) When the output result list RList is empty: The degree name of the initial code based on the note event information and the key information Key recorded in the note event list NList The code name is detected, code data including the detected degree name and code name is generated and registered in the code list CList, and then the basic code candidate extraction process is terminated (steps S61 → S62 → S63 → return);
(61b2) When the output result list RList is not empty: The process proceeds to the preprocessing of (62) (steps S61 → S62 → S64).

  In the process (61b1), the degree name and code name of the initial code may be detected by a conventional method. In this case, since the previous chord to be referred to is not yet stored, for example, a major chord or minor tone tonic chord having a key main tone as a root may be used as a reference code.

In the preprocessing (62), the CPU 5 sets / resets the tonic prohibition flag, the subdominant prohibition flag, and the dominant prohibition flag. In particular,
(62a) Four or more pieces of code data are registered in the output result list RList, and all four pieces from the end, that is, four pieces retroactively from the most recently registered one, are tonics of diatonic scale codes (T ) (Step S64),
When all of the four code data correspond to diatonic scale code tonics: a tonic prohibition flag is set (“1”) (step S65);
If at least one of the four pieces of code data does not correspond to the tonic of the diatonic scale code: reset the tonic prohibition flag (“0”) (step S66);
(62b) It is determined whether or not two or more code data are registered in the output result list RList, and all the two from the end correspond to the subdominant (S) of the diatonic scale code (step S67). ,
When all the two code data correspond to the subdominant of the diatonic scale code: set the subdominant prohibition flag (step S68);
If at least one of the two pieces of code data does not correspond to the subdominant of the diatonic scale code: reset the subdominant prohibition flag (step S69);
(62c) It is determined whether two or more pieces of code data are registered in the output result list RList, and all the two pieces from the end correspond to the dominant (D) of the diatonic scale code (step S70).
When all of the two code data correspond to the dominant of the diatonic scale code: set the dominant prohibition flag (step S71);
If at least one of the two pieces of code data does not correspond to the dominant of the diatonic scale code: reset the dominant prohibition flag (step S72).

  When the above three types of prohibition flags are respectively set, it corresponds to a case where a predetermined number of diatonic scale codes having corresponding functions are selected and sounded (see steps S15 and S17 in FIG. 5B). To do. That is, each prohibition flag has a function of restricting a predetermined number of diatonic scale codes having a corresponding function from being successively selected (steps S83 and S87 in FIG. 9A described later, and steps in FIG. 9B). (See S91). By providing such a prohibition flag, monotonous chord selection and its pronunciation can be suppressed. In addition, the above-mentioned number of examples of prohibiting selection of a new diatonic scale code having the same function when the number of diatonic scale codes having the same function continues is merely an example. If it is plural, it is not limited to two and four.

  Further, the CPU 5 refers to the code data at the end of the output result list RList, acquires the degree name and function of the previous code (step S73 in FIG. 8B), and stores the acquired degree name in the previous degree name LDN. The acquired function is stored in the previous function LF (step S74).

  Next, the CPU 5 proceeds to a degree name candidate extraction process having a function capable of proceeding after the process (63) (step S75).

  FIG. 9A and FIG. 9B are flowcharts showing the detailed procedure of extraction processing of degree name candidates having the function that can proceed to (63) next.

  As shown in FIGS. A and B, in this extraction process, the CPU 5 first extracts all the functions that can proceed to the next based on the previous function LF and key information Key (step S81). At this time, referring to the correspondence table between the code functions and degree names shown in FIG. 3A (when the key information key is major) or FIG. 3B (when the key information key is minor), the process proceeds to the next. Extract all possible functions. For example, if the previous function LF is a tonic (T) and the key information Key is a major key, the CPU 5 can select the tonic (T), subdominant (S), dominant (D) and subdominant minor (SM) are extracted. Further, when the previous function LF is dominant (D) and the key information Key is major, the major dominant (D) → subdominant minor (SM) progression (interrupted Depending on whether or not to allow cadence at normal times, the CPU 5 can select tonic (T), dominant (D), and subdominant minor (SM) ("Allow at normal time") as functions that can proceed to the next. ) And tonic (T) and dominant (D) (when “normally disallowed” is set) are extracted.

Next, the CPU 5 divides the case as follows according to the extracted function and the set state of the prohibition flag corresponding to the function, and executes processing corresponding to each case. That is,
(63a) If the extracted function has a tonic and the tonic prohibition flag value is 0 (reset state): The degree name corresponding to the key information is selected from the degree names corresponding to the tonic of the diatonic scale code. All code names are extracted for each of the degree names, code data including the extracted degree names and the detected code names are generated and registered in the code list CList ( Along with step S84), all of the degree names corresponding to the key information Key among the degree names corresponding to tonics other than the diatonic scale code are extracted, and each chord name is detected for each of the degree names. Each extracted degree name and the detected Each code data including each code name is generated and registered in the code list CList (step S85);
When the value of the tonic prohibition flag = 1 (set state): the process of step S84 is skipped and only the process of step S85 is performed;
(63b) When the extracted function has a subdominant and the subdominant prohibition flag value is 0: all of the degree names corresponding to the subdominant of the diatonic scale code that match the key information key Extract, detect each code name for each of the degree names, generate code data including the extracted degree names and the detected code names, respectively, and register them in the code list CList (step) In S88), all the degree names corresponding to the key information key among the degree names corresponding to the subdominants other than the diatonic scale code are extracted, and each code name is detected for each of the degree names. Each degree name extracted and previous Each code data including each detected code name is generated and registered in the code list CList (step S89);
When the value of the subdominant prohibition flag = 1: The process of step S88 is skipped and only the process of step S89 is performed;
(63c) If the extracted function has a dominant and the dominant prohibition flag value = 0, all of the degree names corresponding to the dominant of the diatonic scale code that match the key information key are extracted, Each code name is detected for each of all the degree names, and each piece of code data including each extracted degree name and each detected code name is generated and registered in the code list CList (step of FIG. 9B). In step S92), all the degree names corresponding to the key information key among the degree names corresponding to the dominants other than the diatonic scale code are extracted, and the respective code names are detected for each of the degree names. Each detected degree name and the detected Each code data including each code name is generated and registered in the code list CList (step S93);
When the value of the dominant prohibition flag = 1: The process of step S92 is skipped and only the process of step S93 is performed;
(63d) When there is a sub-dominant minor in the extracted function: From among the degree names corresponding to the sub-dominant minor, all those that match the key information Key are extracted, and each code name for each of all the degree names And each piece of code data including the extracted degree name and the detected code name is generated and registered in the code list CList (step S95).

  If the extracted function does not include tonic, subdominant, dominant, or subdominant minor, the following processes (63a) to (63d) are not performed.

  The detection of each code name in steps S84, S85, S88, S89, S92, S93 and S95 is performed by expanding the corresponding degree name into a code name based on the key information Key. Specifically, this development is performed by calculating a route from the degree name degree with the main tone of the key information as I degree. Thereby, for example, when the degree name “IIm7” is expanded by the key information “F measure”, the code name “Gm7” is obtained.

  In the correspondence table of FIG. 3A, “-S” is added to “# IVm7 (♭ 5)” belonging to “other than diatonic scale code” with “S: subdominant” in “function”. This “-S” means that “# IVm7 (♭ 5)” is treated as “subdominant (S)”. “# IVm7 (♭ 5)” is normally handled as “Tonic (T)”, but is also handled as “Subdominant (S)” in the present embodiment. Therefore, when “subdominant (S)” is extracted by the process of step S81, if the key information is “major key”, “# IVm7 (♭ 5)” is also extracted. However, in this case, if “Tonic (T)” is extracted together with “Subdominant (S)”, “# IVm7 (♭ 5)” is considered to belong to “Subdominant (S)” and “Tonic (T)”. ) ”Will be extracted. At this time, in the present embodiment, those belonging to “Tonic (T)” are preferentially extracted, but of course this is not limiting, and those belonging to “Subdominant (S)” are preferentially extracted. You may make it do. In this case, it is preferable to add information specifying whether the extracted “# IVm7 (♭ 5)” belongs to “subdominant (S)” or “tonic (T)”.

  FIG. 10 is a flowchart showing a detailed procedure of the additional code candidate extraction process in step S46.

  When the additional code candidate extraction process is started, the CPU 5 first refers to the key information Key, and among the code data registered in the code list CList, the Maj7, 6, add9, Maj of the diatonic scale code , Sus4, excluding the major I degree and the minor マ イ III degree, is extracted as a major solution target code (step S101). As a result, if the key information Key is a key, IVMaj7, IV6, IVadd9 and IV codes are extracted. If the key information Key is a minor key, ♭ VIMaj7, ♭ VI6, ♭ VIadd9 and ♭ VI The code is extracted.

Next, the CPU 5 executes processing for adding a major secondary dominant and its two-five candidates (step S102). In this additional processing (the details are not shown), the CPU 5
(71) Consider each route of the extracted all major system solution destination codes as I degree, detect a code that becomes V7 against them, and code data including a code name and a degree name corresponding to each detected code In the information data of each code data, “Solution code root” is set as “Solution code information”, “V7 of the frequency of the root of solution code” is set as “Flag information” as “Degree information” Generate what stores “interrupt” and add it to the code list CList (add a dominant candidate that resolves to a major system);
(72) Detect each code that becomes IIm7 and IIm7 (♭ 5) with respect to the root of the solution code stored as the solution code information in the information data of each code data added in (71) above, Code data that includes a code name and a degree name corresponding to each detected code. In the information data of each code data, “Solution code root” as “Solution code” and “Degree information” “IIm7 / IIm7 (♭ 5) of V7 / V7sus4 of resolution code root frequency” (“IIm7 / IIm7 (♭ 5)” is a proxy code of “IV / IVm”) and “flag information” Is generated and added to the code list CList (addition of two-five candidates to be resolved to the major system);
(73) Each route of the extracted all major solution destination codes is regarded as an I degree, code corresponding to V7sus4 is detected for them, and code data each including a code name and a degree name corresponding to each detected code Then, in the information data of each code data, “Solution code root” as “Solution code information”, “V7sus4 of the frequency of root of solution code” as “Degree information” and “Flag information” Generate what stores “interrupt” and add it to the code list CList (add a dominant candidate that resolves to a major system);
(74) Consider each route of the extracted all major system solution destination codes as I degree, detect the code that becomes ♭ II7 against them, and code data each including a code name and a degree name corresponding to each detected code In the information data of each code data, “Resolution Destination Code Information” is “Resolution Destination Code Route” and “Degree Information” is “♭ II7 of Frequency of Destination Code Route” (“♭ II7” Is a proxy code of “V7”) and generates “flag information” storing “interrupt”, and adds it to the code list CList (addition of secondary dominant candidates to be resolved to the major system);
(75) The route of each code data added in the above (74) is regarded as V degrees, code corresponding to IIm7 is detected for them, and code data including a code name and a degree name corresponding to each detected code In the information data of each code data, “Resolution Destination Code Information” as “Resolution Destination Code Route” and “Degree Information” as “IIm7 of ♭ II7 of Resolution Destination Code Route Frequency” (= Resolution Destination) Generate ♭ VIm7 when the frequency of the root of the code is regarded as I degree and store “interrupt” as “flag information” and add it to the code list CList (addition of two-five candidates to be resolved to the major system) );
Each process is executed.

  For example, when the solution destination code is “FMaj7”, “C7” is obtained as V7 for “FMaj7” by the process of (71), and IIm7 and IIm7 (♭ 5) for “FMaj7” by the process of (72). ) As “Gm7” and “Gm7 (♭ 5)”, V7sus4 as “V7sus4” for “FMaj7” by the processing of (73), and “G ♭ 7” as ♭ II7 for “FMaj7” by the processing of (74). 7 ”is detected as“ Dm7 ”as IIm7 when“ G ♭ ”of“ G ♭ 7 ”which becomes ♭ II7 with respect to“ FMaj7 ”by the processing of (75) above is regarded as V degree. Is done.

  Next, the CPU 5 refers to the key information Key, and among the code data registered in the code list CList, the flag information is “None”, and the diatonic scale code m7, m6, mMaj, madd9 Corresponding code data excluding minor I degree and V degree is extracted as a minor system solution destination code (step S103). As a result, if the key information Key is a major key, codes IIm7, IIIm7, and VIm7 are extracted, and if the key information Key is a minor key, codes IIm7, IVm7, and IVm6 are extracted.

Next, the CPU 5 executes a process of adding a minor secondary dominant and its two-five candidate (step S104). In this additional processing (the details are not shown), the CPU 5
(81) Code routes including code names and degree names corresponding to the detected codes, respectively, in which each route of the extracted all minor solution destination codes is regarded as an I degree, and a code that becomes V7 is detected. In the information data of each code data, “Solution code root” is set as “Solution code information”, “V7 of the frequency of the root of solution code” is set as “Flag information” as “Degree information” Generate what stores “interrupt” and add it to the code list CList (add a dominant candidate that resolves to a minor system);
(82) Detect each code that becomes IIm7 and IIm7 (♭ 5) with respect to the root of the solution code stored as the solution code information in the information data of each code data added in (81) above, Code data that includes a code name and a degree name corresponding to each detected code. In the information data of each code data, “Solution code root” as “Solution code” and “Degree information” Generate "IIm7 / IIm7 (♭ 5) of V7 / V7sus4 of frequency of solution code root" with "interrupt" stored as "flag information" and add to code list CList (resolve to minor system) Add two-five candidates);
(83) Consider each route of the extracted all minor system-solved codes as I degree, detect codes that become V7sus4 against them, and code data that includes code names and degree names corresponding to the detected codes, respectively Then, in the information data of each code data, “Solution code root” as “Solution code information”, “V7sus4 of the frequency of root of solution code” as “Degree information” and “Flag information” Generate what stores “interrupt” and add it to the code list CList (add a dominant candidate that resolves to a minor system);
(84) Code data including the code name and the degree name corresponding to each detected code, regarding each route of the extracted all minor system solution destination codes as I degree, detecting the code that becomes ♭ II7 against them In the information data of each code data, “Resolution Destination Code Information” is “Resolution Destination Code Route”, “Degree Information” is “♭ II7 of Frequency of Destination Code Route”, “Flag Information” ”And“ Interrupt ”stored therein, and added to the code list CList (addition of secondary dominant candidates to be resolved to the minor system);
(85) The route of each code data added in the above (84) is regarded as V degrees, code corresponding to IIm7 is detected with respect to them, and code data each including a code name and a degree name corresponding to each detected code Then, in the information data of each code data, “Resolution Destination Code Information” as “Resolution Destination Code Route”, “Degree Information” as “IIm7 of ♭ II7 of Resolution Destination Code Route Frequency”, “Flag Generate information that stores “interrupt” as “information”, and add it to the code list CList (add two-five candidates to be resolved to the minor system);
Each process is executed.

  For example, when the solution destination code is “Dm7”, “A7” is obtained as V7 for “Dm7” by the process (81), and IIm7 and IIm7 (♭ 5) for “Dm7” by the process (82). ) As “Em7” and “Em7 (♭ 5)”, “A7sus4” as V7sus4 for “Dm7” by the processing of (83), and “E ♭ as ♭ II7 for“ Dm7 ”by the processing of (84). “B ♭ m7” is detected as IIm7 when “E ♭” of “E ♭ 7”, which becomes ♭ II7 with respect to “Dm7” by the process of (85) above, is regarded as V degrees. Is done.

  Next, the CPU 5 refers to the key information Key, and among the code data registered in the code list CList, the flag information is “none” and the dominant 7th code data of the diatonic scale code is dominant. It is extracted as a 7th system solution destination code (step S105). As a result, if the key information Key is a major key, the V7 and V7sus4 chords are extracted. If the key information Key is a minor key, the V7, V7sus4, ♭ VII7, and ♭ VII7sus4 (IV7) chords are extracted. Is done.

Next, the CPU 5 executes a double dominant candidate addition process (step S106). In this additional processing (the details are not shown), the CPU 5
(91) Considering each route of the extracted all dominant 7th system solution-destination codes as I degrees, detecting a code that becomes V7 against them, code data including a code name and a degree name corresponding to each detected code In the information data of each code data, “Resolution Destination Code Route” as “Resolution Destination Code Information”, “V7 of Resolution Destination Code Route Frequency” as “Degree Information”, “Flag Information” As the "interrupt" stored and added to the code list CList;
(92) Consider each route of the extracted all dominant 7th system solution destination codes as I degree, detect code that becomes V7sus4 against them, and code data including code name and degree name corresponding to each detected code In the information data of each code data, “Resolution Destination Code Information” is “Resolution Destination Code Route”, “Degree Information” is “V7sus4 of Frequency of Resolution Destination Code Route”, “Flag Information” As the "interrupt" stored and added to the code list CList;
(93) Considering each route of the extracted all dominant 7th system solution-destination codes as I degrees, detects codes that are ♭ II7 against them, and each includes a code name and a degree name corresponding to each detected code In the information data of each code data, “Resolution Destination Code Information” is “Resolution Destination Code Route”, “Degree Information” is “♭ II7 of Frequency of Destination Code Route”, “Flag Generate information that stores “interrupt” as “information” and add it to the code list CList;
Each process is executed.

  For example, when the solution destination code is “G7”, “D7” is obtained as V7 for “G7” by the process (91), and “D7sus4” is obtained as V7sus4 for “G7” by the process (92). In the process of (93), “A“ 7 ”is detected as“ ♭ 7 ”for“ G7 ”.

  Next, the CPU 5 determines the set / reset state of the EP flag (step S107). If the EP flag is in the set state, the CPU 5 executes an extended dominant candidate addition process (step S108) and resets the EP flag (step S107). After S109), the process proceeds to step S110. On the other hand, if it is in the reset state, the process skips steps S108 and S109 and proceeds to step S110.

In the extended dominant candidate addition process (the details are not shown), the CPU 5
(101) Detect 7 chords and 7sus4 chords (chord type) that have a pitch that is four times higher than the root pitch name of the previous degree name LDN in the root, and the chord name and degree name corresponding to each chord detected Each of the code data including “extended V7 / V7sus4” as “flag information” is added to the code list CList;
(102) Code data including a chord name and a degree name corresponding to the detected chord detected by detecting 7 chords having a pitch that is one semitone below the root pitch name of the previous degree name LDN. Generate data that stores “Extended ♭ II7” as “flag information” and add it to the code list CList;
Each process is executed.

  In step S110, if there is a duplicate code name (degree name) in the code data registered in the code list CList, the CPU 5 sets the flag information in the information data to “interrupt”, “extended V7”. ”Or“ Extended ♭ II7 ”is deleted from the code list CList.

  11A to 11C are flowcharts showing the detailed procedure of the chord progression forming chord detection process in step S53 of FIG.

  In this code progression forming code detection process, the CPU 5 detects code data for forming the interrupt type chord progression when the flag information in the information data of the start code SChord is “interrupt” (step S121 → S122) When the flag information is “Extended V7 / V7sus4” or “Extended ♭ II7”, the chord data for forming the extended chord progression is detected (Step S121 → S153 in FIG. 11B).

  In the code data detection process for forming an interrupt type code progression (hereinafter referred to as “interrupt type code progress formation code detection process”), first, the CPU 5 is a code candidate that proceeds next to the start code SChord. The type code progression (hereinafter referred to as “next interrupt type code candidate”) is extracted, and the code data corresponding to the next interrupt type code candidate is registered in the code list CList (steps S122 → S123).

  FIG. 12A and FIG. 12B are flowcharts showing the detailed procedure of the next interrupt code candidate extraction process.

In the next interruption type code extraction process, the CPU 5 executes the following process. That is,
(111) When IIm7 or IIm7 (♭ 5) scheduled to be resolved to a major system solution destination code or a minor system solution destination code is detected as the start code SChord:
(111a) Generate chord data including a chord name and a degree name of 7 chords whose root is a pitch name four times higher than the root pitch name of the start code SChord, and register it in the chord list CList (step S171 → S172);
(111b) Generate chord data including a chord name and a degree name of a 7sus4 chord whose root is a pitch four times higher than the root chord name of the start code SChord, and register it in the chord list CList (step S173) ;
(111c) The chord data including the chord name and degree name of the pitch name with the ♭ added to the root pitch name of the start code SChord (lowering the semitone without changing the root pitch name) is generated and registered in the chord list CList ( After step S174), the extraction process is terminated (return);
(112) When V7 or V7sus4 scheduled to be resolved to the major system solution destination code or the minor system solution destination code is detected as the start code SChord:
(112a) Detects all diatonic scale chords whose root is a pitch name four times higher than the root pitch name of the start code SChord according to the key information Key, and the chord name corresponding to each detected chord And code data including the degree names are registered and registered in the code list CList (steps S175 → S176);
(112b) A code having the same function as that of the code data registered by the registration process of (112a) is detected from the diatonic scale code and a code other than the diatonic scale code according to the key information Key. Then, code data each including a code name and a degree name corresponding to each detected code is generated and registered in the code list CList (step S177);
(112c) The code name corresponding to VIMaj7, IV6, IVadd9, IV when the solution code information (the frequency of the route thereof) in the information data of the start code SChord is VI degrees and the key information Key is major. And code data including each degree name are generated and registered in the code list CList (step S178);
(112d) Code data including code names and degree names of 7 chords and 7sus4 chords whose roots are completely four times higher than the root pitch name of the start code SChord, and each chord data information A data in which “extended V7 / V7sus4” is stored as “flag information” is generated and registered in the code list CList (step S179);
(112e) Code data including a chord name and a degree name of 7 chords whose root is a pitch name that is a semitone below the root pitch name of the start code SChord, and the information data includes “extended ♭ II7” as “flag information” Is stored in the code list CList (step S180), and then the extraction process is terminated (return);
(113) When ♭ II7 scheduled to be resolved to the major system solution destination code or the minor system solution destination code is detected as the start code SChord:
(113a) All of the diatonic scale chords whose root is the pitch name that is a second lower than the root pitch name of the start code SChord is detected according to the key information Key, and the chord name corresponding to each detected chord Code data including the degree name and the degree name are registered in the code list CList (steps S181 → S182);
(113b) A code having the same function as the function of the code data registered by the registration process of (113a) is detected from the diatonic scale code and a code other than the diatonic scale code according to the key information Key. Code data including a code name and a degree name corresponding to each detected code are generated and registered in the code list CList (step S183);
(113c) A process similar to the extraction process of (112d) is executed (step S179);
(113d) After performing the same processing as the extraction processing of (112e) (step S180), the extraction processing is terminated (return);
(114) When V7 / V7sus4 scheduled to be resolved to V7 / V7sus4 code is detected as start code SChord:
(114a) Generate chord data including each chord name and each degree name of 7 chords and V7sus4 (V degree only) chords whose roots are pitches four times higher than the root pitch name of the start code SChord Registered in the code list CList (steps S184 to S185 in FIG. 12B);
(114b) A code having the same function as that of the code data registered by the registration process of (114a) is detected from the diatonic scale code and a code other than the diatonic scale code according to the key information Key. Then, code data each including a code name and a degree name corresponding to each detected code is generated and registered in the code list CList (step S186);
(114c) The m7 (= IIm7 of V7) chord and IIm7 (♭ 5) (= IIm7 (♭ 5) of V7) chord names and degree names that have the same root note name as the start chord SChord Code data including “departure” as “flag information” is generated in the information data and registered in the code list CList (step S187), and then the extraction process is terminated (return);
(115) When ♭ II7 scheduled to be resolved to V7 / V7sus4 code is detected as start code SChord:
(115a) Generate chord data including each chord name and each degree name of 7 chords and V7sus4 (V degree only) chords rooted at a pitch that is a second lower than the root chord name of the start code SChord, Registered in the code list CList (steps S188 → S189);
(115b) A code having the same function as that of the code data registered by the registration process of (115a) is detected from the diatonic scale code and a code other than the diatonic scale code according to the key information Key. Code data including a code name and a degree name corresponding to each detected code are generated and registered in the code list CList (step S190);
(115c) The m7 (= IIm7 of V7) chord and IIm7 (♭ 5) (= IIm7 (= IIm7 ( ♭ 5) of V7) Code data including each code name and each degree name of the code, in which information data is stored as “flag information” and “detour” is generated and registered in the code list CList After (step S191), this extraction process is terminated (return);
Each process is executed. Note that when generating code data for registration in the code list CList, steps that do not mention flag information, specifically, steps S172 to S174, S176 to S178, S182, and S183, It is assumed that flag information is not set.

  Returning to FIG. 11A, the CPU 5 next detects the start code SChord as IIm7 or IIm7 (♭ 5) to be resolved to the major system solution destination code or the minor system solution destination code as the start code SChord. Extract the next next code candidate that will be an interrupt type code (hereinafter referred to as “next next interrupt type code candidate”) and correspond to the next next interrupt type code candidate Code data to be registered is registered in the next code list NCList (steps S124 → S125).

In the next process for extracting the next interrupt type code candidate (details thereof are not shown), the CPU 5 executes the following process. That is,
(121) The resolution code is detected from the resolution code information in the information data of the start code SChord (there are a plurality of different types depending on the frequency of the detected code route. If the frequency is IV degrees, IVMaj7, IV, IV6, and IVadd9 are detected as the solution destination codes), and the code data including the code name and the degree name corresponding to each detected solution destination code is generated, and the next Register in the code list NCList;
(122) A code having the same function as the function of the code data registered by the registration process of (121) is detected from the diatonic scale code and codes other than the diatonic scale code according to the key information Key. Code data including a code name and a degree name corresponding to each detected code are generated and registered in the next code list NCList;
(123) Chord data including 7 chord names and 7sus4 chord chord names and degree names, respectively, rooted at a pitch name that is four degrees higher than the fourth complete pitch name with respect to the root pitch name of the start code SChord In the information data of each code data, “flag information” storing “extended V7 / V7sus4” is generated and registered in the next code list NCList;
(124) Code data including a chord name and a degree name of 7 chords whose root is the pitch name one semitone below the pitch name with a コ ー ド added to the root pitch name of the start code SChord, “Extension ♭ II7” stored as “flag information” is generated and registered in the next code list NCList;
Each process is executed.

  Next, the CPU 5 selects one piece of code data based on the note event list NList and the code list CList (step S126), and stores the selected code data in the detection code Chord in the same manner as in the steps S47 and S48. (Step S127).

Then, the CPU 5 executes the following process according to the detection code Chord. That is,
(131) When the detection code Chord is registered in the code list CList by the registration processing of (111a) or (111b): From the next code list NCList, the flag information in the information data is “Extended ♭ II7”. After deleting certain code data (step S128 → S132), the process proceeds to step S131;
(132) When the detection code Chord is registered in the code list CList by the registration process of (111c): From the next code list NCList, code data whose flag information in the information data is “Extended V7 / V7sus4” Is deleted (step S128 → S129 → S130), and then the process proceeds to step S131;
(133) If none of the above cases (131) and (132) applies: Move to step S131 without doing anything;
Execute one of the processes.

  In step S131, the CPU 5 initializes (clears) the code list CList, and then stores all the code data in the next code list NCList in the code list CList. Then, the CPU 5 ends the chord progression forming chord detection process (return).

  In the deletion processes (131) and (132) described above, code candidates that are impossible to connect are deleted as code data that follows the detection code Chord.

  Next, when this code progression forming code detection process is activated, the flag information in the information data of the start code SChord is “interrupt”, and one or more code data is registered in the code list CList. Therefore, the process proceeds to step S144.

  In step S144, it is determined whether or not the flag information in the information data of the detection code Chord is “detour”. If it is not “departure”, the CPU 5 performs the note event list NList in the same manner as in steps S126 and S127. Based on the code list CList, one code data is selected (step S150), and the selected code data is stored in the detection code Chord (step S151). Then, if the flag information in the information data of the detection code Chord is “none”, the CPU 5 sets the flag information to “End” in order to end the interrupt type chord progression formation code detection process (step After S152 → S149), the chord progression forming code detection process is terminated (return). On the other hand, if the flag information is not “None”, nothing is performed to continue the interrupt chord progression forming code detection process. The chord progression forming chord detection process is terminated (step S152 → return).

  On the other hand, as a result of the determination in step S144, if the flag information in the information data of the detection code Chord is “detour”, the CPU 5 is based on the note event list NList and the code list CList similarly to steps S126 and S127. Then, one piece of code data is selected (step S145), and the selected code data is stored in the detection code Chord (step S146). If the detection code Chord (code type) corresponds to 7 code or 7sus4 code, the CPU 5 sets the EP flag (step S147 → S148), and sets the flag information in the information data of the detection code Chord to “End”. (Step S149), the chord progression forming code detection process is terminated (Return). On the other hand, if the detection code Chord does not correspond to the 7 code or 7sus4 code, the process skips Step S148 and performs the process. Advances to step S149.

In step S124, if IIm7 or IIm7 (♭ 5) scheduled to be resolved to the major system solution destination code or the minor system solution destination code is not detected as the start code SChord, the CPU 5 proceeds to steps S126 and S127. Similarly, one chord data is selected based on the note event list NList and the chord list CList (step S133 in FIG. 11C), and the selected chord data is stored in the detection code Chord (step S134). Then, the CPU 5 executes the following process. That is,
(141) When V7, V7sus4, or ♭ II7, which is scheduled to be resolved to a major system solution destination code or a minor system solution destination code, is detected as the start code SChord:
(141a) When the flag information in the information data of the detection code Chord is “None”: The flag information in the information data of the detection code Chord is set to “End” (Steps S136 → S137 → S138);
(141b) When the flag information in the information data of the detection code Chord is not “None”: The chord progression formation code detection process is terminated (Steps S136 → S137 → Return);
(142) When V7 or V7sus4 or ♭ II7 scheduled to be resolved to a major system solution destination code or a minor system solution destination code is not detected as the start code SChord:
(142a) When the flag information in the information data of the detection code Chord is “None”:
(142a1) When the detection code Chord corresponds to 7 code or 7sus4 code: After setting the EP flag (step S136 → S139 → S140 → S141), the flag information in the information data of the detection code Chord is set to “End” (Step S138);
(142a2) When the detection code Chord does not correspond to the 7 code or the 7sus4 code: The flag information in the information data of the detection code Chord is set to “End” (steps S136 → S139 → S140 → S138);
(142b) When the flag information in the information data of the detection code Chord is not “None”: Delete the code data in which the flag information in the information data is “detour” from the code list CList (steps S136 → S139 → S142);
(142b1) When the code list CList is empty: flag information in the information data of the detection code Chord is set to “End” (steps S143 → S138);
(142b2) If the chord list CList is not empty: the chord progression forming chord detection process is terminated (step S143 → return);
Each process is executed.

  When the process proceeds from the step S124 to the step S133, the code that proceeds next to the start code SChord, that is, the code that is detected as the detection code Chord is the code that becomes the end point of the interruption type chord progression or the code that becomes the “detour” Or any of the chords for forming an extended chord progression. The cases where the detection code Chord is the end point of the interruption type chord progression are the cases (141a) and (142a), and the case where the detection code Chord is the “detour” is (142b). This is the case (141b) where the detection code Chord is a code for forming an extended chord progression.

  In step S121 of FIG. 11A, when the flag information in the information data of the start code SChord is “Extended V7 / V7sus4” or “Extended ♭ II7”, the CPU 5 forms the extended chord progression as described above. Detect code data.

  In the chord data detection process for forming the extended chord progression (hereinafter referred to as “extended chord progression forming chord detection process”), first, the CPU 5 is a chord candidate that proceeds next to the start code SChord. The next extended type code candidate that extracts the type code progression (hereinafter referred to as “next extended type code candidate”) and registers the code data corresponding to the next extended type code candidate in the code list CList Is extracted (step S153 in FIG. 11B).

In the next extended code candidate extraction process (details not shown), the CPU 5
(151) The chord data including the chord name and the degree name corresponding to each detected chord are detected as 7 chords and 7sus 4 chords having roots that are four times higher than the root pitch name of the start code SChord. Then, the information data of each code data is generated by storing “extended V7 / V7sus4” as “flag information” and added to the code list CList (however, the same code name (degree name) in the code list CList) If the code data of is already registered, it will not be added));
(152) Code data including a chord name and a degree name corresponding to the detected chord are detected, and the chord data includes a chord name that is one semitone below the root pitch name of the start code SChord. , Generating “extension ♭ II7” as “flag information” and adding it to the code list CList (provided that code data of the same code name (degree name) is already registered in the code list CList) , No additional registration);
Each process is executed.

Next, in step S154, the CPU 5
(161) When the flag information in the information data of the start code SChord is “Extended V7 / V7sus4”: Maj7, 6, add9, m7 having the pitch name four times higher than the root pitch name of the start code SChord as the root , M6, mMaj7, madd9 code data including the degree name and the degree name are generated and registered in the code list CList;
(162) When the flag information in the information data of the start code SChord is “Extended ♭ II7”: Maj7, 6, add9, m7, m6 having a pitch one semitone below the root pitch name of the start code SChord as a root , Generate code data including code names and degree names of mMaj7 and madd9 and register them in the code list CList;
Execute one of the processes.

  Next, the CPU 5 selects one piece of code data based on the note event list NList, the code list CList, and all available codes (step S155), and stores the selected code data in the detection code Chord (step S155). S156). Here, all the chords that can be used mean all chords obtained by substituting the component names of all key diatonic scales into the degrees of the (usable) degree names.

Then, the CPU 5 determines whether or not the flag information in the information data of the detection code Chord is “none” (step S157). In this determination, it is determined whether or not the detection code Chord is an end code of the extended chord progression. Since the chord detection process for forming the type chord progression continues, the extended dominant is subsequently detected. If the flag information is not “none” as a result of the determination in step S157, the CPU 5 immediately ends the chord progression forming code detection process (step S157 → return), whereas if the flag information is “none”. The CPU 5 executes the following processing. That is,
(171) When the flag information in the information data of the start code SChord is “Extended V7 / V7sus4”:
(171a) When the detection code Chord corresponds to one of the codes of Maj7 / 6 / add9 / m7 / m6 / mMaj7 / madd9 whose pitch is completely four times higher than the start code SChord:
(171a1) When there is a code name that can be handled by the key information Key among the code names including the detected code Chord and its synonym code: Code name and degree name determined to be handled by the key information Key Is generated and stored in the detection code Chord (step S161);
(171a2) If there is no code name that can be handled by the key information Key among the code names including the detection code Chord and its synonym / sound code: It is determined as a key modulation and the same as step S42 in FIG. New key information is acquired (step S165), and the acquired key information is stored in key information Key as in step S43 (step S166), and the code name and degree name included in the detection code Chord are set to key information Key. And stored in the detection code Chord (step S167);
(171b) When the detection code Chord does not correspond to any of the codes of Maj7 / 6 / add9 / m7 / m6 / mMaj7 / madd9 whose pitch is completely four times higher than the start code SChord:
(171b1) When the detection code Chord is a code that can be handled by the key information Key: The process proceeds to the above (171a1);
(171b2) When the detection code Chord is not a code that can be handled by the key information Key: the process proceeds to the above (171a2);
(172) When the flag information in the information data of the start code SChord is not “Extended V7 / V7sus4”:
(172a) When the detection code Chord corresponds to one of the codes of Maj7 / 6 / add9 / m7 / m6 / mMaj7 / madd9 whose pitch is one semitone below the start code SChord: (171a1) or ( 171a2)
(172b) When the detection code Chord does not correspond to any of the codes of Maj7 / 6 / add9 / m7 / m6 / mMaj7 / madd9 whose pitch is one semitone below the start code SChord: (171b1) or ( 171b2)
Execute one of the processes.

  The key information Key used in the determination in the processes (171a1), (171a2), (171b1), and (171b2) is acquired and stored in steps S42 and S43 of FIG. That is, the key information Key is basically acquired in the (51) normal code detection process, and is not acquired in the (53) non-normal code detection process. The reason why the key is acquired in the process (171a2) is that the detection code Chord cannot be handled by the key information Key, and is determined to be a key modulation.

  Next, after setting the flag information in the information data of the detection code Chord to “End” (step S162), the CPU 5 ends the chord progression forming code detection process.

  In the present embodiment, the detection code Chord is additionally registered in the output result list RList together with the key information if the normal code detection flag is set after being used for sound generation (see step S15 in FIG. 5B). (See step S17), but may be additionally registered in the output result list RList together with the key information Key regardless of the state of the normal code detection flag. Although the number of detection codes Chord (and key information Key) additionally registered in the output result list RList is not mentioned, in the present embodiment, initialization of the output result list RList is performed in step S3 of FIG. 5A. Therefore, after the automatic accompaniment is started, it is performed only when the setting is changed when the automatic accompaniment is stopped. In other words, the initialization of the output result list RList is not performed except for a special case while the performance processing using the automatic accompaniment is activated, so most of the chord data detected after the automatic accompaniment is started. Registered. This is only for the sake of simplification of description, and an arbitrary predetermined number of code data detected at the latest may be stored, and updated whenever new code data is detected. Good. Further, in the case of registering detected code data, in this embodiment, the detection code Chord is registered as it is together with the key information Key. However, the present invention is not limited to this, and only the code name is registered. Alternatively, a degree name associated with the key information at that time may be registered. Further, both the code name and the degree name, that is, the code data of FIG. 4 excluding the information data may be stored. If only the code name is registered as the detected code, it may be expanded to the degree name using the key information each time it is referred to when detecting the next code.

  In this embodiment, the condition to be extracted as the solution destination code is the diatonic scale code, but the code other than the diatonic scale code is also extracted as the solution destination code. Also good. In this case, the solution destination codes are further diversified, and further various chord progressions are generated.

  Further, in the present embodiment, the solution destination code information is the frequency of the root of the solution destination code with respect to the key of the key, and all the solution destination codes are calculated from the key as much as possible. All the degree names of codes may be stored in advance. Further, “degree information” in the information data of the code data may be used for determining the solution destination code. In the present embodiment, “degree information” is stored only when the flag information is “interrupt”, but appropriate information may be stored every time code data is newly generated. In the present embodiment, “solution destination code information” and “degree information” are provided for each code data, but may be managed in a list and referred to when necessary.

  In the present embodiment, the input performance data is assumed to be performance data generated by a user performing a real-time performance. However, the present invention is not limited to this. For example, recorded performance data is sequentially reproduced without prefetching. You may enter what you did. In this embodiment, a keyboard is used as a performance operator used when a user performs a real-time performance. However, the keyboard is not limited to this, and may be a stringed instrument type such as a guitar. If the performance data to be input is an audio signal, it is converted into note information having at least sounding timing information and pitch information before reference to the performance data in order to select one chord from the candidate chords at each chord detection timing at the latest. It only has to be done.

  The method of holding the information of the code candidate and the next code candidate is not limited to the format adopted in the present embodiment. Further, the order of processing and the way of summarizing the code candidate extraction method and the code candidate selection method are not limited to those employed in the present embodiment.

  Among code candidate groups extracted at the time of normal code detection, code candidates that resolve to code candidates that can be a solution destination, and code candidates that correspond to codes that elapse until reaching a code candidate that can be a solution destination, In this embodiment, a flag is set so that it can be identified from other code candidates. However, the present invention is not limited to this. For example, two types of lists may be prepared, and identification may be performed by another method. In the present embodiment, when a chord candidate with a flag is detected in normal chord detection, it is considered that a unique chord progression starts, and chord detection is performed by a chord detection method that generates a unique chord progression from the next chord detection timing. Detection is performed. Unique chord progressions include resolving from secondary dominant, secondary dominant two-five progression, and a series of dominant 7th chords. The chord progression length is not determined in this embodiment, but may be limited.

  In this embodiment, chords that are musically connected to input performance data are detected and automatic accompaniment is performed based on the detected chords. However, the present invention is not limited to this. Based on the input performance data, a harmony sound may be generated.

  Further, in the present embodiment, no processing is performed on the performance data to be referred to, and the input data is used as it is. However, the present invention is not limited to this. If the chord detection is performed after removing the note data, the chord detection accuracy is further improved.

  Further, in this embodiment, the types of codes described in the correspondence table between the function of the code and the degree name in each key of FIG. 3 are stored in advance in the ROM 6, for example. I didn't mention how. Although it is preferable that additional registration can be arbitrarily performed, additional registration may not be performed.

Further, in the present embodiment, as a case of extracting the next next code candidate group, the secondary dominant to the solution target code and the progress of the two-five are described as an example, but the next code candidate group is further described. The case of extracting For example, it is conceivable to extract the above-mentioned progression using the extended dominant together, specifically, E7 (Extended Dominant) → A7 (V7 of II) → Dm7 (Solution IIm7). For this chord progression, the Extended Dominant (E7) and Secondary Dominant (A7) can each be two-five,
Bm7 (♭ 5) (= IIm7 (♭ 5) of Extended) → E7 (= Extended) → Em7 (♭ 5) (= IIm7 (♭ 5) of Secondary) → A7 (= V7 of II) → Dm7 (= IIm7 )
It becomes. Considering the extended dominant and its two-five, it is possible to extract further candidate groups in advance.

  A program in which a storage medium storing software program codes for realizing the functions of the above-described embodiments is supplied to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus is stored in the storage medium. It goes without saying that the object of the present invention can also be achieved by reading and executing the code.

  In this case, the program code itself read from the storage medium realizes the novel function of the present invention, and the program code and the storage medium storing the program code constitute the present invention.

  As a storage medium for supplying the program code, for example, a flexible disk, hard disk, magneto-optical disk, CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD + RW, magnetic A tape, a non-volatile memory card, a ROM, or the like can be used. Further, the program code may be supplied from a server computer via a communication network.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also the OS running on the computer based on the instruction of the program code performs the actual processing. It goes without saying that a case where the functions of the above-described embodiment are realized by performing part or all of the above and the processing thereof is included.

  Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

DESCRIPTION OF SYMBOLS 1 ... Performance operator (input means), 5 ... CPU (acquisition means, detection means, 1st selection means, 2nd selection means, extraction means, generation means, deletion means, detour code candidate addition means, extended dominant Candidate addition means)

Claims (6)

Input means for inputting performance data;
Acquisition means for acquiring key information of a performance song;
Detection means for detecting the detection timing of the code;
When the detection timing of the code is detected by the detection means, a plurality of code candidates are extracted based on the key information acquired by the acquisition means, and input from the extracted plurality of code candidates by the input means First selecting means for selecting one chord based on the performance data,
Extracting means for extracting a solution destination code that may be chord progression solve destination including code selected by said first selecting means,
Following the selected code based on the acquired key information for each of the detection timings of one or more codes detected by the detection means until reaching the solution destination code extracted by the extraction means code candidate out multiple extraction, a generating means for generating a plurality of code candidate or al 1 or more code candidates that the extracted,
When the detection timing of the code is detected by the pre-Symbol detection means, among the code candidates generated by said generating means, from the code candidates corresponding to the detected timing, the code based on the performance data that has been said input A code detection apparatus comprising: second selection means for selecting one. When a plurality of code candidate groups are generated by the generation unit, a code candidate that is impossible to connect as a code that proceeds next to the code selected by the second selection unit is detected timing of the selected code when in the code candidate group corresponding to the detection timing after more code detection apparatus according to claim 1, further comprising a deleting means to delete the code candidate with the unreasonable.   When the generating means generates a code candidate group, one or more code candidates that can be detoured before one of the plurality of code candidates constituting the code candidate group are extracted and added to the code candidate group The code detection device according to claim 1, further comprising candidate addition means.   When the code selected by the second selection means is the solution-destination code, it is further determined whether or not the selected code can be the starting point of the extended dominant, and as a result of the determination, the selected code is selected. When the detected code can become the starting point of the extended dominant, the plurality of code candidates extracted by the first selection means become the extended dominant according to the detection means detecting the detection timing of the next code. The code detection apparatus according to claim 1, further comprising an extended dominant candidate addition unit that extracts one or more code candidates and adds them.   When the code selected by the second selection means is not the resolution target code but an extended dominant code, the generation means generates a code candidate corresponding to a detection timing next to the detection timing of the code The group includes at least extracted code candidates 1 to plural of the extended dominant following the selected extended dominant code and code candidates 1 to plural to be solved by the selected code. The code detection device according to claim 4. Input procedure for inputting performance data,
An acquisition procedure for acquiring key information of a performance song;
A detection procedure for detecting the detection timing of the code;
When the detection timing of the code is detected by the detection procedure, a plurality of code candidates are extracted based on the key information acquired by the acquisition procedure, and input from the extracted plurality of code candidates by the input procedure A first selection procedure for selecting one chord based on the recorded performance data;
An extraction procedure for extracting the resolved destination code that may be resolved destination chord progression including code selected by said first selection procedure,
Following the selected code based on the acquired key information for each of the detection timings of one or more codes detected by the detection procedure up to the solution destination code extracted by the extraction procedure code candidate out multiple extraction, a generating step of generating a plurality of code candidate or al 1 or more code candidates that the extracted,
When the detection timing of the code is detected by the pre-Symbol detection procedure is out of code candidates generated by the generating procedure, the code candidates corresponding to the detected timing, the code based on the performance data that has been said input A program for causing a computer to execute a second selection procedure for selecting one.

Images