JP4253997B2 - Musical sound information conversion apparatus, control method therefor, and storage medium storing program for realizing the control method - Google Patents

Description

[0001]
[Industrial application fields]
  The present invention is applied to an electronic musical instrument, an automatic accompaniment apparatus that performs automatic accompaniment based on performance information stored in advance or performance information supplied from outside, an automatic performance apparatus that performs automatic performance, and the like. Musical sound information conversion device that performs editing such as data conversion and correction,Its controlWay orStorage medium storing program for realizing control methodAbout.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, some sequencers that perform automatic performance or accompaniment based on performance data have a function of appropriately converting note data included in the performance data based on a chord designated by the user. Such a prior art is disclosed as a performance data editing device in, for example, Japanese Patent Laid-Open No. 8-16994. In this type of performance data editing apparatus, the root note and type of chords are detected in advance from the performance data, and all note data is once associated with the reference chord “Cmaj7” based on the detected chord information. Conversion to note data corresponding to a specified chord is performed by shifting to note data having a pitch, and then conversion to a desired chord is realized.
[0003]
[Problems to be solved by the invention]
By the way, chords are broadly divided into those having three notes (major chord, minor chord, etc.) and those having four notes or more (seventh chord, etc.). The performance data includes the fourth chord constituent sound (for example, the seventh chord is a seventh chord, and in this application, the ninth chord composing sound is also included including the ninth, eleventh, thirteenth, etc.). Yes, if the chord detected from the performance data is composed of four tones, the fourth chord component is converted to another one component such as once to convert it to a chord of any three-or four-component structure. It can be carried out.
[0004]
On the other hand, if the performance data does not include the fourth chord component and the chord detected from the performance data has a three tone configuration, the fourth chord component is converted to a four tone chord as in the above case. Must be generated. In such a case, in the prior art, a note code corresponding to “C”, which is a single component of the reference chord “Cmaj7”, is converted into a pitch “B” which is a fourth chord component. Was. However, even if the pitch of one component sound is simply converted into the fourth chord component sound in this way, there are still three types of chord component sounds that constitute the “Cmaj7” chord. In such a conventional technique, as a result of converting the pitch of one component sound into a fourth chord component sound, there is a problem that the combination of note codes after conversion does not become a chord to be converted.
[0005]
For example, "C₃"," E₃"," G₃"," C₄When trying to convert a chord consisting of “corresponding to“ Cmaj ”” to “Cmaj7”, in the prior art, “Cmaj7”₃"," C₄"Is the 7th component sound" B₂"," B₃"Pitch conversion into. As a result, the converted note code is" B₂"," E₃"," G₃"," B₃This note code combination is equivalent to the chord “Em”.₃"," E₃"," G₃If you try to convert a chord consisting of “corresponding to“ Cmaj ”” to “Cmaj7”,₃"Is the 7th component sound" B₂"Pitch conversion into. As a result, the converted note code is" B₃"," E₃"," G₃This note code combination corresponds to “Con B” which is a fractional chord.
[0006]
That is, in the prior art, when converting performance data into chords designated by the user, it is required that there are substantially four types of chord constituent sounds in the performance data, and there is a significant amount of performance data that can be converted. It was restricted.
[0007]
The present invention has been made to solve the above-mentioned problems in the prior art, and aims to correctly convert chords having four tones even if there are not four types of chords in the performance data. In addition, the processing to generate the 4th chord component is appropriately changed according to the characteristics of the performance part (bass, piano, guitar, etc.) to create natural performance data that is musically comfortable. The purpose is to do.
[0008]
[Means for Solving the Problems]
  In order to solve the above problems, a musical tone information conversion apparatus according to claim 1 of the present invention is provided.Supply means for supplying performance data including a plurality of pitch information arranged in pronunciation order, division means for dividing the entire section of the performance data supplied by the supply means into partial sections of a predetermined length, and the division means For each divided partial section, detection means for detecting chord information of a chord composed of a plurality of pitch information included in the partial section, input means for inputting chord information corresponding to the partial section, and the partial section Each time, a plurality of pitch information included in the partial section is converted into a chord corresponding to the input chord information based on the chord information detected by the detecting means and the chord information input by the input means. A conversion means for converting into pitch information as configured, and a chord composed of a plurality of pitch information included in the partial section to be converted by the conversion means has a three-tone configuration and is input When the chord information has a four-tone structure, pitch information corresponding to a predetermined part of the three chord constituent sounds among the plurality of pitch information included in the partial section to be converted is Extraction means for extracting as a pitch candidate for generating the fourth chord constituent sound, and generation means for generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction means. The fourth chord constituent sound that is preferentially generated in the overlapped portion of the key-on area of the partial section immediately before the partial section to be converted and the key-on area of the partial section immediately after, and the generation means Replacing the pitch information that is the basis for generating the fourth chord constituent sound with the pitch information of the generated fourth chord constituent sound, and the converting means includes the partial section in the partial section. Multiple pitch information included When converting, the pitch information is replaced by the replacement means converts the pitch information after the replacement thereofIt is characterized by that.
[0009]
  A music information conversion device according to claim 2 of the present invention isSupply means for supplying performance data including a plurality of pitch information arranged in pronunciation order, division means for dividing the entire section of the performance data supplied by the supply means into partial sections of a predetermined length, and the division means For each divided partial section, detection means for detecting chord information of a chord composed of a plurality of pitch information included in the partial section, input means for inputting chord information corresponding to the partial section, and the partial section Each time, a plurality of pitch information included in the partial section is converted into a chord corresponding to the input chord information based on the chord information detected by the detecting means and the chord information input by the input means. A conversion means for converting into pitch information as configured, and a chord composed of a plurality of pitch information included in the partial section to be converted by the conversion means has a three-tone configuration and is input When the chord information has a four-tone structure, pitch information corresponding to a predetermined part of the three chord constituent sounds among the plurality of pitch information included in the partial section to be converted is Extraction means for extracting as a pitch candidate for generating the fourth chord constituent sound, and generation means for generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction means. The fourth chord constituent sound that is preferentially generated in the overlapped portion of the key-on area of the partial section immediately before the partial section to be converted and the key-on area of the partial section immediately after, and the generation means Adding means for adding the pitch information of the fourth chord constituent sound generated by the step to the sounding position of the chord composed of a plurality of pitch information included in the partial section, and the converting means includes the partial section in the partial section. Change multiple pitch information included When converts the pitch information after the pitch information is added by said adding meansIt is characterized by that.
[0010]
  Other features of the invention are:A predetermined part of the three chord constituent sounds has a pitch of 5 degrees.
[0014]
  Further, according to claim 4 of the present inventionStorage mediaIsCPU,Supplying performance data containing multiple pitch information arranged in pronunciation ordermeansWhenA computer-readable storage medium storing a program for causing a computer to execute a control method for controlling a musical sound information conversion device comprising:,In the control method, the CPUSupplymeansDividing the whole section of the performance data supplied byStepWhen,The CPUSplitStepDetection for detecting chord information of a chord composed of a plurality of pitch information included in the partial section for each partial section divided byStepWhen,The CPU isFor each partial section, a plurality of pitch information included in the partial section is detected by the detection.StepChord information detected byEnteringPowermeansConversion based on the chord information input by, which converts into pitch information that forms a chord corresponding to the input chord informationStepAnd the transformationStepWhen the chord consisting of a plurality of pitch information included in the partial section to be converted by is a three-tone configuration, and the input chord information is a four-tone configuration,The CPU isAmong the plurality of pitch information included in the partial section to be converted, pitch information corresponding to a predetermined part of the three chord constituent sounds is generated to generate the fourth chord constituent sound. Extraction extracted as a pitch candidateStepWhen,The CPUExtractionStepGenerating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidates extracted byStepA preferentially generating fourth chord constituent sound that is included in the overlapped portion of the key-on area of the partial section immediately before the partial section to be converted and the key-on area of the partial section immediately after,The CPUGenerationStepReplacing the pitch information that was the basis for generating the fourth chord constituent sound with the pitch information of the generated fourth chord constituent soundStepAndPreparationThe conversionStepThen, when converting a plurality of pitch information included in the partial section, the replacementStepThe pitch information replaced by is converted into the pitch information after the replacement.
[0015]
  Further, according to claim 5 of the present inventionStorage mediaIsCPU,Supplying performance data containing multiple pitch information arranged in pronunciation ordermeansWhenA computer-readable storage medium storing a program for causing a computer to execute a control method for controlling a musical sound information conversion device comprising:,In the control method, the CPUSupplymeansDividing the whole section of the performance data supplied byStepWhen,The CPUSplitStepDetection for detecting chord information of a chord composed of a plurality of pitch information included in the partial section for each partial section divided byStepWhen,The CPU isFor each partial section, a plurality of pitch information included in the partial section is detected.StepChord information detected byEnteringPowermeansConversion based on the chord information input by, which converts into pitch information that forms a chord corresponding to the input chord informationStepAnd the transformationStepWhen the chord consisting of a plurality of pitch information included in the partial section to be converted by is a three-tone configuration, and the input chord information is a four-tone configuration,The CPU isAmong the plurality of pitch information included in the partial section to be converted, pitch information corresponding to a predetermined part of the three chord constituent sounds is generated to generate the fourth chord constituent sound. Extraction extracted as a pitch candidateStepWhen,The CPUExtractionStepGenerating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidates extracted byStepA preferentially generating fourth chord constituent sound that is included in the overlapped portion of the key-on area of the partial section immediately before the partial section to be converted and the key-on area of the partial section immediately after,The CPUGenerationStepAdds the pitch information of the fourth chord constituent sound generated by the step to the sounding position of the chord composed of a plurality of pitch information included in the partial sectionStepAndPreparationThe conversionStepThen, when converting a plurality of pitch information included in the partial section, the additionalStepThe pitch information after the pitch information is added is converted by.
[0016]
  Further, according to claim 7 of the present inventioncontrolThe method isCPU,Supplying performance data including multiple pitch information arranged in pronunciation ordermeansWhenA control method for controlling a musical sound information conversion device comprising:,The CPUSupplymeansA dividing step of dividing the entire section of the performance data supplied byThe CPUFor each partial section divided by the dividing step, detecting step for detecting chord information of a chord composed of a plurality of pitch information included in the partial section;The CPU isFor each partial section, a plurality of pitch information included in the partial section is converted into the chord information and the chord information detected by the detection step.EnteringPowermeansA conversion step for converting into pitch information that forms a chord corresponding to the input chord information, and a plurality of sounds included in the partial section to be converted by the conversion step based on the chord information input by When a chord consisting of high information has a three-tone configuration and the input chord information has a four-tone configuration,The CPU isAmong the plurality of pitch information included in the partial section to be converted, pitch information corresponding to a predetermined part of the three chord constituent sounds is generated to generate the fourth chord constituent sound. An extraction step for extracting as pitch candidates;The CPUA generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step, wherein the key-on area and the partial section immediately after the partial section immediately before the partial section to be converted Preferentially generating a fourth chord constituent sound that is included in the overlapped portion of the key-on area ofThe CPUReplacing the pitch information that is the basis for generating the fourth chord constituent sound by the generating step with the pitch information of the generated fourth chord constituent sound, and in the converting step, When converting a plurality of pitch information included in a section, the pitch information after the replacement is converted for the pitch information replaced by the replacement step.
[0017]
  Further, according to claim 8 of the present inventioncontrolThe method isCPU,Supplying performance data including multiple pitch information arranged in pronunciation ordermeansWhenA control method for controlling a musical sound information conversion device comprising:,The CPUSupplymeansA dividing step of dividing the entire section of the performance data supplied byThe CPUFor each partial section divided by the dividing step, detecting step for detecting chord information of a chord composed of a plurality of pitch information included in the partial section;The CPU isFor each partial section, a plurality of pitch information included in the partial section is converted into the chord information and the chord information detected by the detection step.EnteringPowermeansA conversion step for converting into pitch information that forms a chord corresponding to the input chord information, and a plurality of sounds included in the partial section to be converted by the conversion step based on the chord information input by When a chord consisting of high information has a three-tone configuration and the input chord information has a four-tone configuration,The CPU isAmong the plurality of pitch information included in the partial section to be converted, pitch information corresponding to a predetermined part of the three chord constituent sounds is generated to generate the fourth chord constituent sound. An extraction step for extracting as pitch candidates;The CPUA generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step, wherein the key-on area and the partial section immediately after the partial section immediately before the partial section to be converted Preferentially generating a fourth chord constituent sound that is included in the overlapped portion of the key-on area ofThe CPUAdding the pitch information of the fourth chord constituent sound generated by the generating step to the sounding position of the chord composed of a plurality of pitch information included in the partial section, and in the converting step, the partial step When converting a plurality of pitch information included in the section, the pitch information after the pitch information is added by the adding step is converted.
[0025]
DETAILED DESCRIPTION OF THE INVENTION
Next, a first embodiment in which the present invention is applied to an automatic performance device will be described with reference to the drawings.
[1] Hardware configuration of the first embodiment
FIG. 1 shows a schematic block diagram of the automatic performance apparatus.
The automatic performance device 100 includes a CPU 101 for controlling the automatic performance device 100 as a whole, a pointing device such as a mouse and a keyboard for inputting various data, and an input means 102 such as a keyboard, and between the CPU 101 and the input means 102. And an input means interface 103 for performing an interface operation.
[0026]
The automatic performance device 100 also includes a display unit 104 that performs various displays, a sound source unit 106 that generates music signals of a plurality of tone generation channels based on performance data supplied via the bus 105, and an output from the sound source unit 106. The effect means 107 for performing various effect processing (effect processing) on the music signal, and the sound device 108 for amplifying the sound signal input by the effect means 107 and outputting the sound signal as an acoustic signal. The pointing device has a switch or the like, and designates an arbitrary position on the display means 104 and performs an operation such as clicking (turning on the pointing device at a certain place and turning it off on the spot). The various functions can be specified.
[0027]
Further, the automatic performance device 100 includes a MIDI interface 110 that is an interface with an external MIDI device 109 and a communication interface 113 that is an interface with an external server computer 112 via a communication network 111. Has been.
[0028]
Further, an external storage device 114 such as a hard disk drive, a flexible disk drive, a CD-ROM drive, an MO (Magneto Optical) disk drive, a DVD (Digital Versatile Disc) drive, a semiconductor memory device such as a flash memory, a MIDI interface 110, A performance information memory (RAM) 115 for storing performance data supplied from the communication interface 113 and the external storage device 114, a working memory (RAM) 116 for temporarily storing various variables and data, a control program and data And the like, and a timer 118 for performing various timing controls.
[0029]
In this case, the CPU 101 includes an input means interface 103, display means 104, sound source means 106, effect means 107, MIDI interface 110, communication interface 113, external storage device 114, performance information memory (RAM) 115, working memory (RAM). 116 and a program memory (ROM) 117 are connected via a bus 105.
[0030]
The performance data supplied from the MIDI interface 110, the communication interface 113, or the external storage device 114 is composed of data of one or more channels, and the note data constituting each channel is a note representing the pitch. It has information such as number (0 to 127, C3 = 60). Note that the note code in this application is the remainder (0 to 11) obtained by dividing the note number by 12, and corresponds to the note name of the note data.
[0031]
FIG. 2 is a diagram showing a format example of the supplied performance data. The performance data supplied from the outside includes timing data and channels indicating the key-on timing by the number of clocks from the beginning of the song or the previous event data. Channel number (channel 1, channel 2,...), Event data indicating event information such as note number and velocity, and end data indicating the end of music.
[0032]
The performance data supplied from the MIDI interface 110, the communication interface 113, or the external storage device 114 is classified into the same track as the note data of the same channel as shown in FIG. It is separated and stored in a performance information memory (RAM) 115. At this time, the number of tracks is stored in a storage area tr (described later).
[0033]
Further, performance data supplied from the outside may be separated into a plurality of tracks in advance as shown in FIG.
[0034]
[2] Operation of the first embodiment
2-1 Configuration of processing screen
Next, an example of a processing screen displayed on the display unit 104 will be described with reference to FIG.
[0035]
The automatic performance device 100 has a track information display screen for displaying performance data of each track shown in FIG. 4A, and a chord input displayed when a chord designated by the operator shown in FIG. 4B is inputted. And a screen.
[0036]
The track information display screen shown in FIG. 4A includes a track number display area 400 for displaying a plurality of track numbers, and performance data stored for each track number and displaying performance data for the corresponding track for each track. An operator designates a display area 401, a chord detection process start instruction button 402 for instructing an operation for starting a chord detection process described later, a detection code display area 403 for displaying a chord detected by the chord detection process, and A designated chord display area 404 for displaying the chord, a designated chord conversion process start instruction button 405 for instructing a process for converting performance data into a chord designated by the operator, which will be described later, and a measure in which the operator is about to input the chord A bar position display cursor 406 for displaying the position of the scroll bar and a scroll bar 407 for changing the display area Equipped and are configured.
[0037]
In addition, the code input screen shown in FIG. 4B includes a code route instruction area 408 for indicating the code route, a code type instruction area 409 for specifying the code type, and the position of the bar where the operator is to input the code. A bar position display area 410 for displaying a code, a code display area 411 for displaying a currently specified code, a code input button 412 for instructing an operation for inputting a code specified by the operator to a specified position, and a code And a cancel button 413 for canceling the input. In the first embodiment, it is possible to specify chords for the first 1/2 bar and the second 1/2 bar of each bar.
[0038]
4, in the track information display screen and the code input screen, the first half of the twenty-first bar is designated as the chord input position as illustrated in the bar position display cursor 406 and the bar position display area 410.
[0039]
For example, when the operator designates and inputs the code “Cmaj” in the first half of the 21st bar, the operator clicks the first half of the 21st bar in the designated code display area 404 with the input means 102 such as a pointing device, thereby FIG. The bar position display cursor 406 is displayed at the position of the first half of the 21st bar shown in FIG. 4B, and the code input screen shown in FIG. 4B is displayed. The chord root instruction area 408 has buttons 414 for selecting each pitch name (C to B) and chromatic scale (#, ♭, etc.) corresponding to each pitch name, and corresponds to “C”. When the button is selected, “C” is designated as the root of the code. In the code type designation area 409, buttons 415 for selecting code types such as “maj” and “maj7” are arranged corresponding to each type, and when the button corresponding to “maj” is selected. “Maj” is designated as the code type. When the chord input button 412 is pressed, “Cmaj” is input in the first half of the 21st bar.
[0040]
In the processing screen of FIG. 4, the track information screen shown in FIG. 4 (A) displays only a part of a large number of tracks and a part of a plurality of measures. By changing the display area using the scroll bar 407 displayed to the right or below the display area 401, other tracks or other measures can be displayed.
[0041]
2-2 Overall operation
Next, the operation of the first embodiment configured as described above will be described. First, the flow of code conversion processing executed by the CPU 101 of the first embodiment will be described with reference to FIG.
[0042]
The CPU 101 executes various programs such as the code conversion processing program of FIG. 5 in accordance with an event generated by an operator's operation or the like. A main routine (not shown) constantly monitors the occurrence of an event, calls a program corresponding to the generated event, and executes a corresponding process. FIG. 5 shows a flow of processing executed in the CPU 101 when a chord detection instruction event, a chord information supply event, a pitch conversion instruction event, or the like is generated by an operation of the operator.
[0043]
In the following description and each flowchart, each storage area of the RAM 116 used for control is indicated by the following label, and the contents of each storage area indicated by the same label are the same unless otherwise specified.
[0044]
Imax: Storage area for counting the number of 1/2 bar intervals
I: Storage area indicating the current position of the 1/2 bar section
CHRD: a storage area for temporarily storing whether or not there is a pronunciation corresponding to each pitch name in a chord detection section (1/2 measure section), and a 12-bit storage area corresponding to a note code (0 to 11)
RPT: Storage area for counting the number of cyclic shifts when matched
RES: storage area for storing the type of chord detected when matching
SRT_I: Storage area for storing chord root of detected chord in chord detection section (1/2 measure section)
STP_I: Storage area for storing the type of chord detected in the chord detection section (1/2 measure section)
DRT_I: Storage area for storing the chord root of the specified chord in the chord detection section (1/2 measure section)
DTP_I: Storage area for storing the type of specified chord in the chord detection section (1/2 measure section)
tr: Storage area for counting the number of tracks
n: Storage area indicating the current track number
CA: a storage area for temporarily storing all note data corresponding to a pitch of 1 degree and a pitch of 5 degrees in a chord detection section (1/2 measure section)
[0045]
In the chord conversion process of FIG. 5, in response to a chord conversion process instruction event generated by an operator's operation or the like, first, the input means 500 performs a performance from the MIDI interface 110, the communication interface 113, and the external storage device 114 in advance. As shown in the block display area 401 of FIGS. 3 and 4, the supplied performance data is classified for each track and stored in the performance information memory (RAM) 115. At this time, the track information display screen shown in FIG. 4A is displayed on the display screen. At this time, both the detection code display area 403 and the specified code display area 404 are blank. Next, when a chord detection instruction event 501 occurs in response to the operation of the chord detection process start instruction button 402, the chord information extraction unit 502 starts the chord detection process (described later) in FIG. Store information in memory. (This chord detection process is performed based on all the note data of the performance data excluding the drum part. Hereinafter, the drum part refers to the track including the note data of the drum tone.) At this point, the track information display screen The code name is displayed in the upper detection code display area 403 as shown in FIG.
[0046]
The chord information supply means 503 allows the operator to input a code using the code input screen shown in FIG. 4B, so that the input code is displayed in a specified measure in the specified code display area 404. Next, when a pitch conversion instruction event 504 occurs in response to pressing of the designated code conversion process start instruction button 405, the detected chord is inversely converted 505, and the fourth chord constituent sound generation means 506 is informed. If the detected chord has a three-tone structure and the specified chord has a four-tone structure, a seventh-degree sound generation process is performed, and the conversion means 507 performs the chord conversion process (described later) in FIG. Convert performance data into chords. (This reverse conversion process is performed for all the note data of the performance data excluding the drum part, and the chord conversion process is performed for each track of the performance data.) Then, the performance data converted by the output means 508 is stored in the performance information memory. 115.
[0047]
Here, the inverse conversion process 505 is a four-tone structure with a pitch corresponding to the case where the note numbers of all the note data are the chord “Cmaj” if the detected chord is a three-tone structure in the chord detection process. If so, the note numbers of all the note data are converted into pitches corresponding to the chord “Cmaj7”. FIG. 6 shows an example of a reverse pitch conversion table. When reverse conversion is performed to “Cmaj” or “Cmaj7” based on this table, each note data is corrected so as to have an appropriate pitch relationship.
[0048]
For example, the note number of some note data is “58” (A₂#), If the detected chord is “Gm” (3-tone configuration), the chord root pitch name of the detected chord is “G” when shifting to “Cmaj”. Note code “7” (see FIG. 6) is subtracted from note number “58” to obtain note number “51” (D₂Shift to #). Next, since the type of the detected chord is “min” (minor), the correction data “1” is read from “D #” (note code “3”) and “min” in the reverse pitch conversion table of FIG. This correction data “1” is added to the shifted note number “51” to obtain the note number “52” (E₂). As a result, the note number “58” in the chord “Gm” (A₂Note data of #) is note number "52" (E in chord "Cmaj")₂), And the pitch of the chord “Gm” is converted back to the pitch of “Cmaj”.
[0049]
In this embodiment, prior to the 7th tone generation process and the chord conversion process, the above-described inverse conversion process is performed, and all note data (note data) excluding the drum part of the performance data is detected as the chord root of the detected chord. Based on the type, the pitch is converted to a pitch corresponding to “Cmaj” or “Cmaj7”. Such an inverse conversion processing method is disclosed in, for example, Japanese Patent Laid-Open No. 8-160946.
[0050]
2-3 Chord detection processing
Next, with reference to FIG. 7, a chord detection process performed in the chord information extraction unit 502 in response to occurrence of a chord detection instruction event 501 by pressing the chord detection process start instruction button 402 will be described.
[0051]
In the chord detection process of FIG. 7, first, in step 700, the performance data is divided into ½ measures, and the total number of divided ½ measures is stored in Imax. In step 701, I = 1 is set. Then, the process proceeds to step 702. In the following processing, chords are detected for each section using the divided sections as chord detection sections.
[0052]
In step 702, the logical sum of all note names (note codes) of all parts except the drum part in the chord detection section I is stored in CHRD. Here, the bits corresponding to all pitch names that have been on events at least once in the chord detection section I are stored in the CHRD as “1”.
[0053]
KCD (CHRD) in step 703 is a function for performing chord matching processing (not shown) described below, and performs matching with the chord type table while cyclically shifting CHRD, and stores the detected chord type in RES. At the same time, the number of cyclic shifts is stored in the RPT as a note code corresponding to a chord chord root. RPT stores 0 to 11 if detection is successful, and -1 if it fails.
[0054]
FIG. 8 shows an example of the chord type table described above. Data in which each bit of 12 bits is associated with 12 pitch names and the bits of the chord component are set to “1” and other bits are set to “0”. This is stored for each type of chord. In the function KCD (CHRD) in step 703, chord matching processing is performed using this table. Such chord matching processing is disclosed in, for example, Japanese Patent Publication No. 57-16359.
[0055]
Next, in step 704, it is determined whether chord detection has failed. If a chord is detected, the RPT is 0 or more (NO). In step 705, the chord root RPT of the detected chord is changed to SRT._IAnd store chord type RES in STP_IAnd proceed to step 709.
[0056]
If chord detection fails in step 704, RPT is -1 (YES), and in step 706, it is determined whether the current chord detection section is the head section.
[0057]
If it is determined in step 706 that I ≠ 1, that is, if it is not the first section (NO), the process proceeds to step 707 and the chord chord route SRT in the immediately preceding section is reached._I-1SRT_IAnd chord type STP in the immediately preceding section_I-1STP_IAnd proceed to Step 709. In other words, the previous section_I-1Do not change the chords determined in step.
[0058]
If it is determined in step 706 that I = 1, that is, if it is the first section (YES), the process proceeds to step 708 and SRT₁, STP₁A default value (for example, Cmaj) is stored in step 709 and the process proceeds to step 709.
[0059]
In step 709, it is determined whether or not I = Imax. If I ≠ Imax (NO), the pointer I of the chord detection section is incremented by 1 in step 710, and step 702 and subsequent steps are repeated. The processing in steps 702 to 710 is repeated for all the sections divided in step 700. If I = Imax is determined in step 709 (YES), the chord detection processing is terminated.
[0060]
2-4 Chord conversion processing
Next, referring to FIG. 9, an operation performed by the fourth chord constituent sound generation means 506 and the conversion means 507 after the reverse conversion processing 505 of the chord detected in response to the occurrence of the pitch conversion instruction event 504 is performed. A chord conversion process for converting performance data into chords designated by the user will be described.
[0061]
In the chord conversion process of FIG. 9, the process is performed for each section divided in step 700 of FIG. 7. First, 1 is set to I indicating the section in step 900, and 1 is set to n indicating the track number in step 901. To do.
[0062]
Chord chord root SRT of chord detection section I detected by the process of FIG._IAnd type of detected chord STP_IThe chord root DRT of the chord designated by the operator at 504 in FIG._IAnd specified chord type DTP_IIs read. Next, in step 903, performance data for 1/2 bar is read for each track. However, drum parts are excluded because there is no need to convert the pitch of drum parts.
[0063]
In step 904, it is determined whether or not the detected chord is composed of four tones. If the detected chord is composed of four tones (YES), the process proceeds to step 907, and pitch conversion processing is performed based on the designated chord as it is. If the detected chord is composed of three tones in step 904 (NO), the process proceeds to step 905.
[0064]
In step 905, it is determined whether or not the user-specified chord is composed of four tones. If the user's designated chord is composed of 3 tones in step 905 (NO), the detected chord is composed of 3 tones and the designated chord is composed of 3 tones, so that the process proceeds to step 907 and the pitch conversion is performed based on the designated chord as it is. Processing is performed.
[0065]
If it is determined in step 905 that the user-specified chord is composed of four notes (YES), the process proceeds to the fourth chord constituent sound generating means in step 906. In this case, since the detected chord has a three-tone configuration and the specified chord has a four-tone configuration, in step 906, the fourth chord component tone (7 degrees) of the specified chord is generated to be a four-tone chord. After performing the seventh tone generation process (described later), the process proceeds to step 907, where the pitch conversion process is performed based on the specified chord.
[0066]
Here, the pitch conversion process in step 907 corresponds to the specified chord, the note number of the note data that has been converted back to the pitch corresponding to “Cmaj” or “Cmaj7” by the inverse conversion process 505. This is a process of converting into a note number and storing it in the working memory 116.
[0067]
For example, it is assumed that “Fm” is designated and the detected chord has a four-tone structure. At this time, each note data is inversely converted into a note number in the case of “Cmaj7” by the inverse conversion process, and the pitch of these note data is converted into a note number in the case of “Fm”. FIG. 10 shows an example of the pitch conversion table, and the converted pitch is obtained based on the pitch conversion table.
[0068]
For example, in the case described above, the note number of a certain note data is “52” (E₂). In this case, since the designated chord type is “min” (minor), the conversion data “−1” is read from the chord type “min” designated by the note code “E” corresponding to the note number “52”. The converted data “−1” is added to the note number “52” to be corrected to the note number “51”. Since the chord root of the designated chord is “F”, the note code “5” (see FIG. 10) of “F” is added to the corrected note number “51”, and the note number “52” (E₂) Note number “56” (G₂#). Such pitch conversion processing is disclosed in, for example, Japanese Patent Laid-Open No. 7-219534.
[0069]
By performing the above processing and converting all the data, “Cmaj7” is converted to a chord “Fm”.
[0070]
When the detected chord has a three-tone structure, the performance data reversely converted by the reverse conversion process is “Cmaj”, but the pitch conversion is performed in the same manner as described above based on the pitch conversion table of FIG. It can be performed.
[0071]
Performing the above-described process, the pitch corresponding to “Cmaj” or “Cmaj7” of the performance data inversely converted by the inverse conversion process is converted into a pitch corresponding to the specified chord, and the working memory 116.
[0072]
In step 908 of FIG. 9, it is determined whether n = tr. If n ≠ tr (NO), the track number n is incremented by 1 in step 909, and steps 904 and thereafter are repeated. If n = tr (YES), the process proceeds to step 910.
[0073]
In step 910, it is determined whether I = Imax. If I ≠ Imax (NO), the pointer I of the chord detection section is incremented by 1 in step 911, and step 901 and subsequent steps are repeated. If I = Imax (YES), the process returns to the original routine.
[0074]
As a result of this routine, the original performance data is converted into chords into a chord designated by the operator without changing the style.
[0075]
2-5 7th sound generation processing
The seventh-degree sound generation process in step 906 will be described. The 7th tone generation process is a process in which appropriate note data is converted to a chord “Cmaj7” from the note data reversely converted to the reference chord “Cmaj” when the detected chord has a three-tone configuration and the specified chord has a four-tone configuration. The fourth chord constituent sound (7 degrees) “B” is converted into a four-tone chord.
[0076]
When generating the seven-degree sound, the seventh-degree sound is generated based on the following rules so that it can be heard as a chord of the designated route and type and can be heard as a chord without a sense of incongruity.
(1) The lowest pitch is not subject to conversion to a pitch of 7 degrees in order to prevent misunderstanding of chords of different routes and chords of different types.
(2) The generated 7th sound is not the lowest sound.
(3) Of the 1st, 3rd, and 5th pitches, the 3rd pitch is farthest from the 7th pitch, so it generates a 7th pitch from the 3rd pitch. In this case, the pitch difference of each note data may be inappropriate as a chord, so it is not subject to conversion to 7 degrees.
(4) If there is a pitch of 1 degree other than the lowest pitch, generating a pitch of 7 degrees higher than this pitch from this pitch will change the octave and make the connection with the previous and next sections unnatural. Therefore, a pitch of 7 degrees (7 degrees lower) that is in the immediate vicinity is generated.
(5) If there is no pitch other than the lowest tone, the pitch is generated by appropriately changing the pitch from 5 ° to 7 ° higher than this sound or 7 ° lower than this sound. Also in this case, on the basis of generating a pitch of 7 degrees lower than the pitch of 5 degrees, when a pitch of 7 degrees lower than this sound is generated, the pitch of 7 degrees becomes the lowest sound, etc. Only when inconvenience occurs, a pitch of 7 degrees higher than this sound is generated.
(6) Other than the above (1) to (5) (when the pitch of the chord component is 4 or more and the number of pronunciations is 5 or more), the pitch to be generated is 7 degrees (in the first embodiment, A pitch of 7 degrees is generated from the chord constituent sound closest to the pitch of the 7 degrees within the region of the chord constituent sound.
[0077]
FIGS. 11A to 11G are examples of the seventh-degree sound generation process.
FIG. 11A shows that the pronunciation in the chord detection section I is C.₃This is a case where only the above-mentioned condition is satisfied. The above condition (1) is satisfied, and the pitch is not converted to 7 degrees.
FIG. 11B shows that the pronunciation in the chord detection section I is E.₃, G₃, C₄This is the case where the pitch of the highest pitch (C₄) To a lower 7 degree pitch (B₃).
FIG. 11C shows a chord detection section I with C₃, E₃, G₃, C₄In the same way as (b), the above note (4) is applied, and the highest pitch (C₄) To a lower 7 degree pitch (B₃).
FIG. 11D shows that the pronunciation in the chord detection section I is E.₂, E₃, G₃This is the case where the pitch of 5 degrees is the highest (G₃) To a lower 7 degree pitch (B₂).
FIG. 11 (e) shows that the pronunciation in the chord detection section I is C.₃, E₃, G₃This is the case where the pitch of 5 degrees is the highest (G₃) Is 7 degrees higher than this sound (B₃).
FIG. 11 (f) shows that the pronunciation in the chord detection section I is C.₃, E₃The above condition (3) is satisfied, and the pitch is not converted to 7 degrees.
In FIG. 11 (g), the pronunciation in the chord detection section I is G.₂, C₃, E₃, G₃, E₄The pitch of 7 degrees to be generated (B₃) Closest chord component (G₃) Is converted to a pitch of 7 degrees.
[0078]
In the example described here, FIG. 11 (a) is an example of performance data often found in a bass performance part, and FIGS. 11 (b) to 11 (f) are often seen in a piano performance part. FIG. 11G shows an example of performance data, and FIG. 11G shows an example of performance data often found in guitar-type performance parts.
[0079]
Hereinafter, with reference to the example of FIG. 11 and FIG. 12, the seventh-degree sound generation process executed in step 906 of FIG. 9 will be described.
[0080]
First, in step 1200, note data (hereinafter referred to as chord constituent sound) from which only the chord constituent sound is excluded from the chord detection section I, that is, the non-chord constituent sound is read out, and the process proceeds to step 1201.
[0081]
In step 1201, note data having a pitch of 1 degree and a pitch of 5 degrees of the chord constituent sound are extracted as generation candidates from the chord detection section I and stored in CA, and the process proceeds to step 1202.
[0082]
In step 1202, it is determined whether or not the pitch of the chord constituent sound read in step 1200 is one (the number of pronunciations is arbitrary). If the pitch of the chord constituent sound is one (YES), step 1202 is performed. In 1203, the routine returns to the original routine even if the pitch is not changed (the pitch of 7 degrees is not generated). This corresponds to the above condition (1).
[0083]
If it is determined in step 1202 that the pitch of the chord constituent sound is not one (NO), the process proceeds to step 1204, where the pitch of the chord constituent sound is two or three (the number of pronunciations is arbitrary), or the pitch is four. Thus, it is determined whether or not the number of pronunciations of the chord constituent sounds is four (that is, they are sounded once at each pitch).
[0084]
If it is determined in step 1204 that the pitch of the chord constituent sound is two or three, or if the pitch is four and the number of pronunciations of the chord constituent sound is four (YES), the process proceeds to step 1205, and in step 1201 above It is determined whether or not note data having one pitch other than the lowest tone of the chord constituent sound exists in the stored CA.
[0085]
If there is a pitch other than the lowest tone of the chord component in the determination in step 1205 (YES), in step 1206, the highest pitch is converted to a lower pitch of 7 °. Return to the original routine. This corresponds to the above condition (3).
[0086]
If it is determined in step 1205 that there is no pitch other than the lowest tone of the chord constituent sound (NO), the process proceeds to step 1207, and the CA stored in step 1201 contains 5 other than the lowest tone of the chord constituent sound. It is determined whether or not there is note data with a certain pitch.
[0087]
If it is determined in step 1207 that there is a pitch of 5 degrees other than the lowest tone of the chord constituent sound (YES), the process proceeds to step 1208, and there is no pitch of 5 degrees other than the lowest tone of the chord constituent sound. (NO), step 1203 returns to the original routine even if the pitch is not changed (a pitch of 7 degrees is not generated). This corresponds to the above condition (2).
[0088]
In step 1208, it is determined whether or not a pitch of 7 degrees, which is lower than the pitch of 5 degrees of the highest tone, is the lowest tone. If the pitch of 7 degrees lower than the highest pitch of 5 degrees is the lowest pitch in this determination (YES), the process proceeds to step 1209, and the highest 5 degrees pitch is 7 degrees higher than this pitch. Return to the original routine after conversion to. This corresponds to the above conditions (1) and (4).
[0089]
If it is determined in step 1208 that the pitch of 7 degrees lower than the highest pitch of 5 degrees is not the lowest pitch (NO), the process proceeds to step 1210, and the highest 5 degrees pitch is set to 7 degrees lower than this pitch. It returns to the original routine even if it is converted to the pitch. This corresponds to the condition (4) above.
[0090]
If it is determined in step 1204 that the pitch of the chord constituent sound is two or three, or if the pitch is four and the number of sound of the chord constituent sound is not four (NO), it is generated in step 1211 and step 1212. A chord constituent sound closest to a power of 7 degrees (in the first embodiment, the highest pitch among the 7 degrees within the chord constituent sound area) is converted to a 7 degree pitch. Return to the original routine. This corresponds to the above condition (5).
[0091]
2-6 Effects of the first embodiment
As described above, according to the first embodiment, when the performance data is converted to a pitch corresponding to the chord having the four-tone structure, even if the performance data is based on the three-tone chord, It can be correctly converted to a 4-tone chord. Furthermore, since the fourth chord component can be changed and generated as appropriate according to the characteristics of the performance part (bass, piano, guitar, etc.), it is possible to create natural performance data with no musical discomfort. It becomes.
[0092]
Further, since the lowest pitch is not subject to conversion to the fourth chord component, it is possible to prevent misidentification of a chord of a different route or a chord of a different type. In addition, since the musical tone information that generated the fourth chord constituent tone is selectively generated from among a plurality of musical tone information that does not have the lowest pitch, it does not sound like a different chord. In addition, since the third pitch is not subject to conversion to the fourth chord constituent sound, the pitch difference of each note data does not become inappropriate as a chord. In addition, when there is a pitch other than the lowest tone, the fourth chord constituent sound (7 degrees lower in the embodiment) that is immediately close to this one pitch is generated, so that there is a connection with the preceding and following sections. There is no unnaturalness.
[0093]
Further, since the performance data may be composed of three sounds, the memory size can be reduced. Also, performance data can be created with three-tone chords, making creation easy.
[0094]
[3] Second embodiment
3-1 Hardware configuration
Next, a second embodiment in which the present invention is applied to an automatic accompaniment apparatus will be described with reference to FIG.
The automatic accompaniment apparatus 1300 includes a CPU 1301 for controlling the automatic accompaniment apparatus 1300 as a whole, sound source means 1303 for generating a music signal based on performance data supplied via the bus 1302, and music output by the sound source means 1303. An effect unit 1304 that performs various effect processing (effect processing) on the signal, and a sound device 1305 that amplifies the audio signal input by the effect unit 1304 and outputs it as an acoustic signal are configured.
[0095]
Also, a working memory (RAM) 1306 that temporarily stores various data, a program memory (ROM) 1307 that stores a control program and control fixed data, an accompaniment pattern memory (ROM) 1308 that stores accompaniment patterns, A panel switch 1310 having a keyboard 1309 for performing operations such as keyboard performance and chord designation, a pattern selection switch for selecting an accompaniment pattern for automatic accompaniment, a start / stop switch for instructing start and stop of automatic accompaniment, and other switches And a timer 1311 for performing various timing controls.
[0096]
The CPU 1301 controls the entire electronic musical instrument using the working area of the working memory 1306 based on the control program stored in the program memory 1307, plays the keyboard by operating the keyboard 1309, and sets the mode by operating the panel switch 1310. Switching, automatic accompaniment based on the accompaniment pattern ROM 1308 is performed.
[0097]
The accompaniment pattern ROM 1308 stores a plurality of accompaniment patterns, and each accompaniment pattern is composed of a plurality of tracks such as a bass and a drum. Of these, the accompaniment pattern selected according to the operation of the panel switch 1310 is read by the CPU 1301 into the working memory (RAM) 1306 and controlled according to the automatic accompaniment start / stop instruction. In this embodiment, all the accompaniment patterns stored in the accompaniment pattern ROM 1308 are stored as key codes corresponding to the chord “Cmaj”, which has three types of chord constituent sounds.
[0098]
The keyboard 1309 is virtually divided into a low-pitched left key range and a high-pitched right key range. During automatic accompaniment, the CPU 1301 performs note codes corresponding to the operated keys for key events in the right key range. The tone signal is generated and silenced, and a chord is detected based on the detected note code for the key event in the left key range. At the time of this automatic accompaniment, note data of the currently selected accompaniment pattern is read out, and the note data is converted into a pitch based on the chord detected from the keyboard 1309 and output to the sound source 1303 together with the key-on signal or key-off signal. Further, for the rhythm part, the timbre number and the key-on signal are output to the sound source 1303, and the sound generation process and the mute process for the automatic accompaniment are performed.
[0099]
3-2 Operation of the second embodiment
Next, the operation of the second embodiment configured as described above will be described.
First, the flow of code conversion processing executed by the CPU 1301 of the second embodiment will be described with reference to FIG.
[0100]
Similar to the first embodiment, the CPU 1301 executes various programs such as the code conversion processing program of FIG. 14 according to events generated by the operation of the operator or the like. A main routine (not shown) constantly monitors the occurrence of an event, calls a program corresponding to the generated event, and executes a corresponding process.
[0101]
FIG. 14 shows the flow of processing executed by the CPU 1301 when an accompaniment pattern selection event, an automatic accompaniment start event, a chord information supply event, or the like is generated by the operation of the operator.
[0102]
First, when an accompaniment pattern selection event occurs in response to an operation of the panel switch 1310 or the like by the operator, an accompaniment pattern is selected by the input means 1400. Next, when an automatic accompaniment start event occurs in response to the start switch button 1401 being operated, automatic accompaniment is started. At this time, the accompaniment pattern selected by the operator is output to the sound source 1303 and reproduced as the key code corresponding to the chord “Cmaj”. Alternatively, only the rhythm part may be played until a chord is designated by the keyboard 1309 or the like, and the parts other than the rhythm may be played only after the chord is designated.
[0103]
Next, when a chord information supply event occurs in response to a chord specified by the keyboard 1309 or the like in the chord information supply means 1402, the fourth chord constituent sound generation means 1404 performs a seventh-degree sound generation process on the conversion means. In 1405, the chord conversion process is started in real time, and the accompaniment pattern stored with the key code corresponding to the chord “Cmaj” is pitch-converted to a chord designated by the operator. In the seventh tone generation processing performed by the fourth chord constituent sound generation means 1404 and the chord conversion processing performed by the conversion means 1405, the fourth chord constituent sound generation means 506 and the conversion means 507 in FIG. 5 of the first embodiment. The processes described with reference to FIGS. 9 to 12 are applied as they are.
[0104]
Next, the accompaniment pattern after the pitch conversion is output to the sound source 1303 by the output means 1406 and reproduced.
[0105]
By performing the above processing, the accompaniment pattern composed of three types of chord constituent sounds can be converted into a four-tone chord as described in the chord conversion processing described above, and there is no musical discomfort. Natural performance data can be created.
[0106]
3-3 Effects of the second embodiment
As described above, the chords stored in the accompaniment pattern memory 1308 conventionally have to be composed of four notes, but in the automatic accompaniment apparatus of the second embodiment, the chords stored in the accompaniment pattern memory 1308 have a three-tone structure. Even if it exists, it can be correctly converted into a chord composed of four tones. Furthermore, since the fourth chord component can be changed and generated as appropriate according to the characteristics of the performance part (bass, piano, guitar, etc.), it is possible to create natural performance data with no musical discomfort. It becomes.
[0107]
Further, since the lowest pitch is not subject to conversion to the fourth chord component, it is possible to prevent misidentification of a chord of a different route or a chord of a different type. In addition, since the musical tone information that generated the fourth chord constituent tone is selectively generated from among a plurality of musical tone information that does not have the lowest pitch, it does not sound like a different chord. In addition, since the third pitch is not subject to conversion to the fourth chord constituent sound, the pitch difference of each note data does not become inappropriate as a chord. In addition, when there is a pitch other than the lowest tone, the fourth chord constituent sound (7 degrees lower in the embodiment) that is immediately close to this one pitch is generated, so that there is a connection with the preceding and following sections. There is no unnaturalness.
[0108]
Further, the accompaniment pattern stored in the accompaniment pattern memory may have a three-tone structure, so that the memory size can be reduced. In addition, since the accompaniment pattern may be based on a chord having a three-tone structure, the accompaniment pattern can also be created with a chord having a three-tone structure, and the creation becomes easy.
[0109]
[4] Modification
The embodiment described above can be variously modified as described below.
4-1 Modification of chord detection
In the first embodiment, the chord detection section is performed as a fixed section every 1/2 bar as shown in FIG. 7, but the chord detection section may be changed as appropriate as shown in FIG. Good.
[0110]
In the chord detection process of FIG. 15, the performance data is divided into ½ measures, the chord detection target is one measure, and chord detection is performed at the first half measure and the second half measure of the target measure. When a chord is detected only in one of the first and second half, the detected chord is applied to all the one measure. If the chord is unknown in both the first and second half, the chord detection is performed with the target measure being the chord detection section. This makes it possible to compensate for a section where chords cannot be detected.
[0111]
First, in step 1500, the performance data is divided into ½ measures, and the total number of divided ½ measures is stored in Imax. In step 1501, n = 1 is set, and the flow proceeds to step 1502. In step 1502, the first half of the target 1 bar, the second half of the target 1 bar, and the target 1 bar are all chord detection sections S1, S2, and S3. A process of storing the logical sum of the names in CHRD1, CHRD2, and CHRD3 is performed. CHRD1 to CHRD3 are storage areas for temporarily storing whether or not there is a pronunciation corresponding to each note name in the chord detection section, and are 12-bit storage areas corresponding to note codes (0 to 11). Here, the bits corresponding to all pitch names that are on events at least once in the chord detection section are stored as “1” in CHRD1, CHRD2, and CHRD3.
[0112]
The function KCD (CHRD1) in step 1503 is a function for performing chord matching processing as described above, and performs chord matching processing for the first half bars. Next, in step 1504, it is determined whether or not chord detection for the first half of the bar has failed. If the chord detection is successful, the RPT becomes 0 or more (NO), and in step 1505, the chord root RPT of the detected chord is changed to SRT._2n-1And store chord type RES in STP_2n-1And proceed to Step 1506.
[0113]
In step 1506, the chord matching process of the latter half 1/2 bar is performed by the function KCD (CHRD2). Next, in step 1507, it is determined whether or not chord detection for the latter half 1/2 bar has failed. If the chord detection is successful, the RPT becomes 0 or more (NO), and in step 1508, the chord root RPT of the detected chord is changed to SRT._2nAnd store chord type RES in STP_2nAnd proceed to Step 1518.
[0114]
If chord detection fails in step 1507, RPT indicates a negative value (YES), and chord chord root SRT detected in the first half of bar in step 1517._2n-1The second half of SRT_2nAnd chord type STP_2n-1STP_2nAnd proceed to Step 1518.
[0115]
If chord detection fails in step 1504, RPT indicates a negative value (YES), and the chord matching process for the second half of the bar is performed using the function KCD (CHRD2) in step 1509.
[0116]
Next, in step 1510, it is determined whether or not the chord detection of the latter half 1/2 bar has failed. If the chord detection is successful, the RPT becomes 0 or more (NO), and in step 1511 the chord root RPT of the detected chord is changed to SRT._2n-1And store chord type RES in STP_2n-1The process proceeds to step 1517, the same processing as described above is performed, and the process proceeds to step 1518. Thereby, the chord detected in the latter half 1/2 bar is also applied to the first half 1/2 bar in which the chord could not be detected.
[0117]
If chord detection fails in step 1510, RPT indicates a negative value (YES), and the chord matching process for the target 1 bar is performed using the function KCD (CHRD3) in step 1512.
[0118]
Next, in step 1513, it is determined whether or not the chord detection of the target 1 bar has failed. If the chord detection is successful, the RPT becomes 0 or more (NO), the same processing as described above is performed in step 1511, and the same flow as in steps 1517 and 1518 is followed. Thereby, the chord detected in the target 1 bar is applied to the first half 1/2 bar and the second half 1/2 bar where the chord could not be detected.
[0119]
If chord detection fails in step 1513, RPT indicates a negative value (YES), and in step 1514, it is determined whether n = 1.
[0120]
If it is determined in step 1514 that n = 1 is not satisfied (NO), the process proceeds to step 1515 and the chord chord route SRT in the immediately preceding section is reached._{2 (n-1)}SRT_2n-1And chord type STP in the immediately preceding section_{2 (n-1)}STP_2n-1The process proceeds to step 1517, the same processing as described above is performed, and the process proceeds to step 1518. In other words, the previous section_{2 (n-1)}The chord determined in (1) is applied to the first half 1/2 bar and the second half 1/2 bar where the chord of the target 1 bar could not be detected.
[0121]
If it is determined in step 1514 that n = 1 (YES), the process proceeds to step 1516, and SRT₁, STP₁A default value (for example, Cmaj) is stored in step 1517, and the process proceeds to step 1517. The same processing as described above is performed, and the process proceeds to step 1518. That is, the default value is applied to the first half 1/2 bar and the second half 1/2 bar where the chord of the target 1 bar could not be detected.
[0122]
In step 1518, it is determined whether or not n = Imax / 2. If n ≠ Imax / 2 (NO), n is incremented by 1 in step 1519, and step 1502 and subsequent steps are repeated. The processing in steps 1502 to 1517 is repeated for all sections divided in step 1500. If n = Imax / 2 is determined in step 1518 (YES), the chord detection processing is terminated.
[0123]
In this modification, the performance data is divided into 1/2 bars, the chord detection target is 1 bar, the chord detection is performed at the first 1/2 bar and the latter 1/2 bar of the target 1 bar, and the first half. When a chord is detected on only one side, the detected chord is applied to all the measures of the target. If the chord is unknown in both the first and second half, the chord detection is performed with the target measure being the chord detection section, so that it is possible to compensate for the chord undetectable section.
[0124]
4-2 Modification of chord conversion process
In the chord conversion process, the process is performed every 1/2 bar as shown in FIG. 9, but when there is no chord change in one bar as shown in FIG. You may make it perform.
[0125]
In the chord conversion process of FIG. 16, first, the chord detected in the target 1 bar and the specified chord are read in step 1600. That is, the detected chords and designated chords of the first half 1/2 bar and the second half 1/2 bar are read out. In the following, the same processing as in FIG.
[0126]
In step 1601, it is determined whether or not there is a change in the designated chords of the first half 1/2 bar and the second half 1/2 bar read in step 1600. If there is a change in the specified chords of the first half 1/2 bar and the second half 1/2 bar in step 1601 (YES), the process proceeds to step 1603.
[0127]
If there is no change in the specified chords of the first half 1/2 bar and the second half 1/2 bar in step 1601 (NO), the process proceeds to step 1602 where the detected chords of the first half 1/2 bar and the second half 1/2 bar read in step 1600 are detected. It is determined whether or not there has been a change.
[0128]
If there is a change in the detected chords in the first half 1/2 bar and the second half 1/2 in step 1602 (YES), the process proceeds to step 1603, and there is no change in the detected chords in the first half 1/2 bar and the second half 1/2 bar. (NO), go to Step 1604.
[0129]
In step 1603, the performance data for the first half of the bar is read for each track, the number of tracks is stored in tr, and the process proceeds to step 1605. However, drum parts are excluded because there is no need to convert the pitch of drum parts.
[0130]
In step 1604, performance data for one measure of the target measure is read for each track, the number of tracks is stored in tr, and the process proceeds to step 1605. However, drum parts are excluded because there is no need to convert the pitch of drum parts.
[0131]
In step 1605, 1 is set to n indicating the track number, and the processing from step 904 is executed. The processing from step 904 to step 909 is the same as the processing from step 904 to step 909 of the chord conversion processing shown in FIG.
[0132]
In step 1606, it is determined whether or not the performance data read in step 1603 or 1604 is one measure. If the read performance data is one measure (YES), the process proceeds to step 1609, and the read performance data is one measure. Otherwise (NO), the process proceeds to step 1607. In this case, the chord conversion process for the first half 1/2 bar has only been completed, and the chord conversion process for the second half 1/2 bar has not yet been performed.
[0133]
In step 1607, it is determined whether or not the chord conversion process for the remaining half measures has been completed. If the chord conversion process for the remaining 1/2 bar has been completed (YES), the process proceeds to step 1609. If the chord conversion process for the remaining 1/2 bar has not been completed (NO), the second half 1 is performed in step 1608. The performance data for two bars is read for each track, the number of tracks is stored in tr, and the process proceeds to step 1605. However, drum parts are excluded because there is no need to convert the pitch of drum parts.
[0134]
In step 1609, it is determined whether or not the chord conversion process has been completed for all the bars. If the chord conversion process has not been completed (NO), the target bar is moved to the next bar in step 1610, and step 1600 and subsequent steps are repeated. If the chord conversion process is completed for all measures (YES), the process returns to the original routine.
[0135]
As a result of this routine, if there is no chord change in the target 1 bar, chord conversion and 7th tone generation processing are performed using 1 bar as the determination section. The effect obtained by performing such processing will be described with reference to the example of FIG.
[0136]
In the example of FIGS. 17A and 17B, C in the first half of the target 1 bar of a certain part₃, E₃, G₃Note data, C in the last half 1/2 bar₄, G₃, C₃Note data is present, and the detected chord is Cmaj in both the first 1/2 bar and the second 1/2 bar. Assume that Cmaj7 is designated as the specified chord in both the first half 1/2 bar and the second half 1/2 bar.
[0137]
FIG. 17A shows a case where the chord conversion process is performed based on the chord conversion process shown in FIG. In the first half of the bar, the above condition (5) is satisfied, and in FIG. 12, step 1202 → step 1204 → step 1205 → step 1207 → step 1208 → step 1210,₃B₂Convert to In the second half of the bar, the above condition (4) is satisfied, and in FIG. 12, the process proceeds from step 1202 to step 1204 to step 1205 to step 1206.₄B₃Convert to
[0138]
FIG. 17B shows a case in which the chord conversion process shown in FIG. 16 is performed, and a seventh-degree sound generation process is performed in units of one measure. The chord conversion process satisfies the above condition (6), and in FIG. 12, the process proceeds from step 1202 to step 1204 to step 1211 to step 1212.₄B₃Convert to
[0139]
In this way, when the chord conversion process shown in FIG. 9 is performed as shown in FIG. 17A, a lot of pitches of 7 degrees may be generated, and as shown in FIG. When the chord conversion process shown in FIG. 16 is performed, it is possible to create natural performance data that is more musically comfortable.
[0140]
4-3 Modification 1 of 7th-degree sound generation processing
In addition, in the seventh-degree sound generation processing, when the condition (5) is satisfied, the processing is performed from step 1208 to step 1210 in FIG. 12 to convert the fifth-degree sound into a seventh-degree pitch. As described above, the seventh-degree sound generation may be performed based on the overlapped portion of the key-on area of the preceding and following sections (the key-on area indicates a range in which all note data sounded in the section exists).
[0141]
At this time, the seventh-degree sound is generated based on the following conditions. First, the key-on area in the section before and after the conversion target section from which the note data is read out in step 1200 of FIG. 12 is determined, and the overlapped portion of the key-on area before and after that is determined.
[0142]
(5) -1 If a pitch of 7 degrees higher than a pitch of 5 degrees and a pitch of 7 degrees lower than the pitch of 5 degrees are present in the superimposed portion, a pitch of 7 degrees lower is generated.
(5) -2 If only a low 7 degree pitch exists in the overlapped portion, a low 7 degree pitch is generated.
(5) -3 If only a high 7 degree pitch exists in the superposed portion, a high 7 degree pitch is generated.
(5) -4 If there is neither a high 7 degree pitch nor a low 7 degree pitch in the overlapped portion, a low 7 degree pitch is generated.
[0143]
FIGS. 18H to 18J are examples in which the seventh sound generation process is performed under the above conditions.
The chord component sound in the conversion target section is G₃, E₄, G₄Consists of
In FIG. 18 (h), the overlapping portion is A.₃~ D₃And the pitch of 5 degrees (G₄) Lower than 7 degrees (B₃) Only exists in the overlapped portion, the above condition (5) -2 is satisfied, and the highest pitch of 5 degrees (G₄) Is a 7 degree pitch (B₃).
In FIG. 18 (i), the overlapping portion is C.₄# ~ F₄# Is the pitch of 5 degrees (G₄) Lower 7 degree pitch (B₃) Is a high 7 degree pitch (B₄) Also does not exist in the overlapped portion, the above condition (5) -4 is satisfied, and the highest pitch of 5 degrees (G₄) Is a 7 degree pitch (B₃).
In FIG. 18 (j), the overlapping portion is F.₄# ~ B₄And the pitch of 5 degrees (G₄) Higher than 7 degrees (B₄) Only exists in the overlapped portion, the above condition (5) -3 is satisfied, and the highest pitch of 5 degrees (G₄) Is 7 degrees higher than this sound (B₄).
[0144]
Hereinafter, with reference to the example of FIG. 18 and FIG. 19, a case where the seventh-degree sound generation process is performed under the above-described conditions will be described.
[0145]
In FIG. 19, first, in step 1900, the key-on area in the section before and after the conversion target section is determined, and the overlapping part of the key-on area before and after that is determined. Next, in step 1901, it is determined whether or not there is a 7 degree pitch lower than the highest 5 degree pitch in the overlapped part determined in step 1900.
[0146]
If it is determined in step 1901 that there is a pitch of 7 degrees lower than the highest pitch of 5 degrees in the determined overlapped portion (YES), the process proceeds to step 1902 and the highest pitch of 5 degrees is set to this sound. Convert to a lower 7 degree pitch. This corresponds to the above conditions (5) -1 and (5) -2.
[0147]
If it is determined in step 1901 that there is no 7 degree pitch lower than the highest 5 degree pitch in the determined overlapped portion (NO), the process proceeds to step 1903, and the overlapped portion determined in step 1900 is the highest. It is determined whether or not a pitch of 7 degrees higher than a high pitch of 5 degrees exists.
[0148]
If it is determined in step 1903 that there is a pitch of 7 degrees higher than the highest pitch of 5 degrees in the determined overlapped portion (YES), the process proceeds to step 1904, and the highest pitch of 5 degrees is set to this sound. Convert to a higher pitch of 7 degrees. This corresponds to the above condition (5) -3.
[0149]
If it is determined in step 1903 that there is no 7 degree pitch higher than the highest 5 degree pitch in the determined overlapped portion (NO), the process proceeds to step 1902 and the highest 5 degree pitch is set to this pitch. Convert to a pitch of 7 degrees lower than the sound. This corresponds to the above condition (5) -4.
[0150]
As described above, by generating the seventh sound based on the overlapped portions of the key-on areas in the preceding and following sections, it is possible to create more natural performance data with no musical discomfort.
[0151]
4-4 Second Modification of Seven-degree Sound Generation Processing
Also, in the 7th tone generation process of FIG. 12, the note data closest to the 7th pitch to be generated in step 1212 is converted to a 7th pitch, but a chord other than the lowest tone in the obsessive or weak beat is used. The note data of the constituent sounds may be converted into a pitch of 7 degrees. Alternatively, note data of chord constituent sounds other than the lowest note having the longest note length may be converted to a pitch of 7 degrees. Alternatively, note data of chord constituent sounds other than the lowest sound that is pronounced most often may be converted to a pitch of 7 degrees.
[0152]
4-5 Modification 7 of 7th-degree sound generation processing
In addition, although the chord constituent sound that is the target in the seventh-degree sound generation process of FIG. 12 is converted to the pitch of seven degrees, the pitch of the seventh degree is added without conversion, and the original chord constituent sound that is the target You may pronounce it at the same time.
[0153]
For example, “E₃"," G₃"," C₄When a chord consisting of “(Cmaj”) is converted to “Cmaj7”, the conventional technique uses “Cmaj”.₄"B", which is the 7th component sound₃Note that the converted note code is “E₃"," G₃"," B₃There is a problem that a chord “Em” is heard. If processing is performed to add a pitch of 7 degrees as in this modification, “B₃"Is added, so the converted note code is" E₃"," G₃"," B₃"," C₄"And can be accurately converted as a chord" Cmaj7 ".
[0154]
20A to 20G show examples in which 7-degree pitches are added using the example of the 7-degree sound generation process shown in FIG.
In FIG. 20A, as in FIG. 11A, the pronunciation in the chord detection section I is C.₃This is a case where only the above-mentioned condition is satisfied. The above condition (1) is satisfied, and the pitch is not converted to 7 degrees.
In FIG. 20B, as in FIG. 11B, the pronunciation in the chord detection section I is E.₃, G₃, C₄This is the case where the pitch of the highest pitch (C₄) To a lower 7 degree pitch (B₃Generate note data C)₄When pronunciation₃Is pronounced at the same time.
FIG. 20 (c) is similar to FIG.₃, E₃, G₃, C₄Note that there is one note data each, and the highest pitch (C₄) To a lower 7 degree pitch (B₃Generate note data C)₄When pronunciation₃Is pronounced at the same time.
In FIG. 20D, as in FIG. 11D, the pronunciation in the chord detection section I is E.₂, E₃, G₃This is the case where the pitch of 5 degrees is the highest (G₃) To a lower 7 degree pitch (B₂) Generates note data G₃When pronunciation₂Is pronounced at the same time.
In FIG. 20 (e), as in FIG. 11 (e), the pronunciation in the chord detection section I is C.₃, E₃, G₃This is the case where the pitch of 5 degrees is the highest (G₃) To 7 degrees higher than this sound (B₃) Generates note data G₃When pronunciation₃Is pronounced at the same time.
In FIG. 20 (f), as in FIG. 11 (f), the pronunciation in the chord detection section I is C.₃, E₃The above condition (3) is satisfied, and the pitch is not converted to 7 degrees.
In FIG. 20 (g), as in FIG. 11 (g), the pronunciation in the chord detection section I is G.₂, C₃, E₃, G₃, E₄The pitch of 7 degrees to be generated (B₃) Closest chord component (G₃) To 7 degree pitch (B₃) Generates note data G₃When pronunciation₃Is pronounced at the same time.
[0155]
4-6 Modification 4 of 7th-degree sound generation processing
In addition, when there are a plurality of note data to be converted to a pitch of 7 degrees in the conversion target section in the 7-degree sound generation process of FIG. 12, all the note data has been converted to 7 degrees. Only part of the note data may be converted to a pitch of 7 degrees.
[0156]
Furthermore, in the 7th sound generation process of FIG. 12, the same 7th sound generation process is performed for the performance data of all measures, but the above-described modifications such as A melody, B melody, chorus, ending, etc. The example processing may be selectively performed.
[0157]
4-7 Detection of repeat count
Further, when performing the chord conversion process, it is detected whether or not the note data sounded in the n track of the conversion target section is constituted by repetition of some pattern, and if the repetition of the pattern is recognized, the conversion target section is selected. It is also possible to divide and perform a seventh-degree sound generation process for each of the divided areas.
[0158]
For example, in the previous stage of step 906 in FIG. 9, the number of repetitions is detected, the conversion target section of the chord conversion process is divided into a plurality of areas based on the number of repetitions, and the seventh-degree sound generation process is performed for each area. What should I do?
[0159]
The number of repetitions is detected as shown in FIG. First, in step 2100, ptn1 to ptn1 are created, and the process proceeds to step 2101.
[0160]
This ptn is a storage area for storing a repetitive area pattern, and is created based on the note data of the target track n among the music data for 1/2 bar read in FIG. This ptn is created as follows by equally dividing a 1/2 bar into 16 blocks.
ptn1 = first 16 blocks (first 1/2 bar)
ptn2 = first 8 blocks (first quarter bar)
ptn3 = first 4 blocks (first 1/8 bar)
ptn4 = first 2 blocks (first 1/16 bar)
[0161]
In step 2101, it is determined whether ptn2 + ptn2 = ptn1. If ptn2 + ptn2 ≠ ptn1 (NO), it means that the 2nd repetition of the first 1/4 bar is different from the first 1/2 bar, so that it is determined that there is no repetition, and the process proceeds to Step 2102.
[0162]
In step 2102, it is determined that there is no repetition at ptn1 (1/2 measure), and the routine returns to the original routine even if 1 is stored in R indicating the number of repetitions.
[0163]
If it is determined in step 2101 that ptn2 + ptn2 = ptn1 (YES), the process proceeds to step 2103 to determine whether ptn3 + ptn3 = ptn2. If ptn3 + ptn3 ≠ ptn2 (NO), the second repetition of the first 1/8 bar is different from the first 1/4 bar, so the target 1/2 bar is determined to be the second repetition of 1/4 bar. Then, the process proceeds to Step 2104.
[0164]
In step 2104, ptn2 (1/4 bar) is determined as a repeated region, and the process returns to the original routine even if R = 2 is stored.
[0165]
If ptn3 + ptn3 = ptn2 (YES) in step 2103, the process proceeds to step 2105 to determine whether ptn4 + ptn4 = ptn3. If ptn4 + ptn4 ≠ ptn3 (NO), it means that the 2nd repetition of the first 1/16 bar is different from the first 1/8 bar, so the target 1/2 bar is determined to be 4 repetitions of the 1/8 bar. Then, the process proceeds to Step 2106.
[0166]
In step 2106, ptn3 (1/8 measure) is determined as a repeated area, and the process returns to the original routine even if R = 4 is stored.
[0167]
If ptn4 + ptn4 = ptn3 (YES) in step 2105, it means that the 2nd repetition of the first 1/16 bar coincides with the first 1/8 bar, so the target 1/2 bar is 1/16 bar. It is determined that the repetition is 8 times, and the process proceeds to Step 2107.
[0168]
In step 2107, ptn4 (1/16 measure) is determined as a repeated area, and the process returns to the original routine even if R = 8 is stored.
[0169]
Based on the number of repetitions detected as described above, the conversion target section is divided into 1 to 8 areas, and the 7th sound generation process of FIG. 12 is performed for each of the divided areas.
[0170]
Note that the number of repetitions may be different depending on whether the note number in the note data or the note code is referenced.
[0171]
FIG. 22A and FIG. 22B show the results of performing the seventh-degree sound generation process shown in FIG. 12 when the repetition area is detected based on the note number and when the number of repetitions is detected based on the note code. Indicates.
[0172]
Here, in the example of FIG. 22, E is included in the first 1/4 bar of the target 1/2 bar of a certain part.₃, G₃Note data, E in the last 1/4 bar₄, G₄Note data is present, and the detected chord in the target 1/2 measure is Cmaj. Assume that Cmaj7 is designated as the specified chord.
[0173]
FIG. 22A shows the result of the seventh-degree sound generation process shown in FIG. 12 when a repetitive region is detected based on the note number. First 1/2 bars (E₃, G₃) And the latter half of the bar (E₄, G₄) Are different note numbers, it is determined that there is no repetition, and the above condition (5) is met, and the process proceeds from step 1202 → step 1204 → step 1205 → step 1207 → step 1208 → step 1210.₄B₃Convert to
[0174]
FIG. 22B shows the result of performing the seventh-degree sound generation process shown in FIG. 12 when a repetitive region is detected based on the note code. First 1/2 bars (E₃), G₃)) And the latter half 1/2 bar (E₄), G₄)) Is the same note code (E, G), so it is confirmed that the 1/4 measure is repeated twice, and the first 1/4 measure and the second 1/4 measure are set to the conditions (2) and (5) above. If so, go to step 1202 → step 1204 → step 1205 → step 1207 → step 1208 → step 1209,₃) B₃), G₄) B₄).
[0175]
Accordingly, as illustrated in FIGS. 22A and 22B, different results are obtained when the note number is referred to and when the note code is referred to.
[0176]
Whether to refer to the note number or note code in the detection of the number of repetitions can be selected as a whole or in units of 1/2 measure or 1 measure, creating natural performance data that does not feel uncomfortable musically What can be used may be used selectively.
[0177]
Furthermore, it is of course possible to combine the above-described seventh-degree sound generation processing modifications 1 to 4 as appropriate. For example, as illustrated in FIG. 19 as illustrated in FIGS. 22C and 22D, the seventh-degree sound generation processing may be combined with the first modification.
[0178]
FIG. 22 (c) shows the result of the seventh-degree sound generation process shown in FIG. 19 in the case where the repeated area is detected based on the note number. First 1/2 bars (E₃, G₃) And the latter half of the bar (E₄, G₄) Is different note data, it is determined that there is no repetition, and the above condition (5) -3 is met, and the process proceeds from step 1900 to step 1901 to step 1903 to step 1904.₄B₄Convert to
[0179]
FIG. 22D shows the result of performing the seventh-degree sound generation process shown in FIG. 19 when a repetitive region is detected based on the note code. First 1/2 bars (E₃, G₃) And the latter half of the bar (E₄, G₄) Is the same note code (E, G), so it is confirmed that the quarter bar is repeated twice, and the latter quarter bar section becomes the rear section key-on area with respect to the former quarter bar section, and the overlapping part (E₄~ G₄) Does not have a pitch of 7 degrees.₃B₂Convert to Similarly, in the latter half 1/4 measure, the above condition (5) -4 is applied.₄B₃Convert to
[0180]
As described above, the musical performance-free natural performance data can be created by appropriately combining the means for detecting the number of repetitions and the seventh-degree sound generation processing means shown in FIGS. Can also be used selectively.
[0181]
Further, the seventh-degree sound generation process may be repeated for each area, or the processes of the first to fourth modifications of the seventh-degree sound generation process may be selectively performed for each area. . In addition, instead of performing the seventh-degree sound generation process in all the regions, the seventh-degree sound generation may be performed by selecting an area in which the seventh-degree sound generation process is performed.
[0182]
4-8 Other variations
Moreover, although the detection and designation of the chord are possible every 1/2 bar, the present invention is not limited to this, and it may be performed every beat or every bar.
[0183]
In addition, the chord detection is performed based on the note data of all the parts except the drum part. However, it may be a part of the part. The part may be fixed, or the user may May be specified.
[0184]
In addition, although “Cmaj7” is used as the reference chord for performing the inverse transformation process, other chords may be used. In that case, the shift amount shown in the reverse pitch conversion table of FIG. 6 or the pitch conversion table of FIG. 10 may be appropriately changed.
[0185]
Further, chord progression data may be stored in advance, and the chord detection process may be omitted.
[0186]
Also, the part to be converted is based on the note data of all parts except the drum part. However, it may be a part of the part, and this part may be fixed or used. The person may specify.
[0187]
In the first embodiment, the chord conversion process is performed for each track, but may be performed for each part.
[0188]
The performance data may be composed of automatic performance data of the melody part and automatic performance data of the accompaniment part constituting the music, and the automatic performance data of only the accompaniment part may be subjected to chord conversion processing.
[0189]
In the first embodiment, when the specified chord conversion process start instruction button 405 is pressed by the operator, the chord conversion process is performed on the performance data of all measures. However, the operator performs chord detection and chord conversion process. May be designated, and chord detection or chord conversion processing may be performed only in that section.
[0190]
In the first embodiment, the designated code is blank, and the operator inputs the code using the code input screen shown in FIG. 4B. However, it is detected when the chord is detected. The chord may be copied in advance as a specified code. Further, when the designation code conversion process start instruction button 405 is pressed by the operator while the designation code remains blank, the chord conversion process may be performed by copying the detection code as the designation code.
[0191]
The note data is not limited to reverse conversion based on the detected chord, but may be directly converted into a specified chord. In this case, the attribute (1 degree, 3 degree, 5 degree, etc.) of each note data may be obtained based on the detected chord and applied to the present invention. In this case, in the fourth chord constituent sound generation process, a conversion table corresponding to each note name is prepared, and the fourth chord constituent sound may be generated based on the obtained attribute of each note data and the conversion table. . Note that the method for obtaining the attribute of each piece of note data is described in detail, for example, in JP-A-6-337677.
[0192]
Further, the fourth chord constituent sound to be generated is not limited to 7 degrees long, but may be 7 degrees short, 6 degrees long, 6 degrees short, or the like.
[0193]
Further, the fourth chord constituent sound is generated by the processing flow based on the fourth chord constituent sound generation rule, but the fourth chord constituent sound generation rule is tabulated (for example, the combination of the original note and the generated note combination is tabulated). The fourth chord constituent sound may be generated based on this.
[0194]
Further, the fourth chord constituent sound may be generated from the non-chord constituent sound, not limited to the fourth chord constituent sound generated from the chord constituent sound.
[0195]
Further, in the second embodiment, all accompaniment patterns are created based on “Cmaj”, but those created based on “Cmaj7” may also be included. In that case, it is discriminated whether it was created based on “Cmaj” or “Cmaj7” (for example, original chord information is included in the accompaniment pattern), and what is said to be “Cmaj” Only the present invention has to be applied. In addition, all accompaniment patterns may be created based on chords of another three-tone structure. In this case, it may be converted back to “Cmaj” or a corresponding table may be prepared. . In addition, the accompaniment pattern may be created by mixing other roots and other types of chords. In that case, the accompaniment pattern may be converted back to “Cmaj” or “Cmaj7” once.
[0196]
In the second embodiment, the chord may be specified by a performance operator other than the keyboard (for example, a stringed instrument type).
[0197]
Further, music data that is input in real time by a performer may be converted instead of performance data on a recording medium.
[0198]
Further, in the performance data editing apparatus for creating and editing performance data, for example, performance data arbitrarily created and edited by a user may be appropriately converted.
[0199]
Further, the input performed by the performance data input means may be supplied from a keyboard, a panel switch or the like, or may be supplied from communication or various memories.
[0200]
The output performed by the output means may be generated as a musical sound or may be written in a memory.
[0201]
Moreover, although the example which applied this invention to the automatic performance apparatus and the automatic accompaniment apparatus was demonstrated, embodiment is not limited to this.
[0202]
For example, the present invention may be an application program that operates on a general-purpose computer such as a personal computer. In this case, the program may be supplied by the above-mentioned various recording media, or may be executed on a network and downloaded from a remote server or the server. There may be.
[0203]
Moreover, the form integrated as a part of function of a mobile telephone, a game device, and game software may be sufficient.
[0204]
Further, the performance information memory and the accompaniment pattern memory may be placed in a remote server computer or the like.
[0205]
A software sound source may be used as the sound source of the present invention.
[0206]
In the second embodiment, the keyboard used in the automatic accompaniment apparatus of the present invention may be a software keyboard, a controller of a game machine, or an operation panel of a mobile phone.
[0207]
【The invention's effect】
As described above, according to the present invention, when the performance data is converted into a pitch corresponding to the chord having the four-tone structure, even if the performance data is based on the three-tone chord, Since the four-chord constituent sounds can be generated by appropriately changing, it is possible to create natural performance data that is musically uncomfortable.
[0208]
Further, according to the present invention, the chord stored in the accompaniment pattern memory or the like needs to have a four-tone configuration, but even if the chord stored in the accompaniment pattern memory or the like has a three-tone configuration, Since it can be correctly converted, it is possible to create natural performance data that is musically comfortable.
[0209]
Further, according to the present invention, when generating the fourth chord constituent sound, the musical tone information as the conversion candidate is determined and only those that can be converted are converted into the fourth chord constituent sound. There is no conversion.
[0210]
Further, according to the present invention, musical tone information of a plurality of pitches corresponding to the fourth chord constituent tone is selectively added, so that it can be accurately converted into the supplied chord.
[0211]
According to the present invention, when the fourth chord constituent sound is generated, an appropriate one of the plurality of rules for generating the fourth chord constituent sound is selected, and the fourth chord constituent sound is selected based on the selected rule. Since musical tone information corresponding to chord constituent sounds is generated, it becomes possible to create natural musical performance data that does not feel strange according to the characteristics of the performance part (bass, piano, guitar, etc.). .
[0212]
In addition, according to the present invention, when the fourth chord constituent sound is generated, the tone information that does not have the lowest pitch is selectively generated, so that the chord having the four-tone structure can be obtained without changing the root sound. It can be converted correctly.
[0213]
Further, according to the present invention, the pitch of the third degree is not subject to conversion to the fourth chord constituent sound, so that the pitch difference of each note data does not become inappropriate as a chord.
[0214]
Further, according to the present invention, when there is a pitch of one degree other than the lowest pitch, the fourth chord constituent sound (7 degrees lower in the embodiment) that is close to this one pitch is generated. There is no unnatural connection with the section.
[0215]
Further, according to the present invention, the performance data or the accompaniment pattern may have a three-tone structure, so that the memory size can be reduced.
[0216]
In addition, according to the present invention, performance data or accompaniment patterns can be created with three-tone chords, which facilitates creation.
[0217]
Further, according to the present invention, in the chord detection process, the performance data is divided every 1/2 bar, the chord detection target is one bar, and the first half bar and the second half bar of the target bar are one. Each chord is detected, and when a chord is detected in only one of the first and second half, the detected chord is applied to all the one measure. If the chord is unknown in both the first and second half, the chord detection is performed with the target measure being the chord detection section, so that it is possible to compensate for the chord undetectable section.
[0218]
Further, according to the present invention, in the chord conversion process, when there is no chord change in one measure, since the entire one measure is converted, natural performance data that is more musically comfortable is created. Can do.
[0219]
Further, according to the present invention, since the fourth chord constituent sound generation processing is performed based on the overlapped portions of the key-on areas of the preceding and following sections, it is possible to create more natural performance data that is not musically uncomfortable. it can.
[0220]
Further, according to the present invention, in the chord conversion process, it is detected whether or not the note data sounded in the n track of the conversion target section is constituted by some pattern repetition, and when the pattern repetition is recognized. Since the conversion target section is divided and the fourth chord constituent sound generation processing is performed for each of the divided areas, more natural performance data without musical discomfort can be created.
[Brief description of the drawings]
FIG. 1 is a block diagram of an automatic performance device according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a format of performance data supplied in the first embodiment.
FIG. 3 is a diagram showing track division in the first embodiment.
FIG. 4 is an explanatory diagram of an example of a processing window in the first embodiment.
FIG. 5 is a main processing block diagram in the first embodiment.
FIG. 6 is a diagram showing an example of a reverse pitch conversion table in the first embodiment.
FIG. 7 is a flowchart of chord detection processing in the first embodiment.
FIG. 8 is a diagram showing an example of a chord type table in the first embodiment.
FIG. 9 is a flowchart of chord conversion processing in the first embodiment.
FIG. 10 is a diagram showing an example of a pitch conversion table in the first embodiment.
FIG. 11 is a diagram showing an example of seventh-degree sound conversion in the first embodiment.
FIG. 12 is a flowchart of 7-degree generation processing in the first embodiment.
FIG. 13 is a block diagram of an automatic accompaniment apparatus according to a second embodiment of the present invention.
FIG. 14 is a block diagram of main processing in the second embodiment.
FIG. 15 is a flowchart of a modification of the chord detection process.
FIG. 16 is a flowchart of a modification of the chord conversion process.
FIG. 17 is a diagram illustrating the reason for performing the seventh-degree sound generation processing in units of one measure when there is no chord change within one measure.
FIG. 18 is a diagram showing a modified example of the seventh-degree sound generation process.
FIG. 19 is a flowchart of a modification of the seventh-degree sound generation process.
FIG. 20 is a diagram showing an example of adding a seventh-degree sound.
FIG. 21 is a flowchart of the repeat count detection process.
FIG. 22 is a diagram showing another modified example of the seventh-degree sound generation process.
[Explanation of symbols]
101 CPU, 102 input means, 104 display means, 106 sound source means, 115 performance information memory (RAM), 116 working memory (RAM), 117 program memory (ROM),
401 performance data display area 402 chord detection process start instruction button 403 detection code display area 404 specified code display area 405 specified chord conversion process start instruction button 408 chord route instruction area 409 chord type instruction area 411 chord display Area, 412 code input button, 414 code route selection button, 415 code type selection button,
1308 Accompaniment pattern memory (ROM), 1309 keyboard, 1310 Panel switch

Claims (9)

Supply means for supplying performance data including a plurality of pitch information arranged in pronunciation order;
Dividing means for dividing the entire section of the performance data supplied by the supplying means into partial sections of a predetermined length;
Detecting means for detecting chord information of a chord composed of a plurality of pitch information included in the partial section for each partial section divided by the dividing means;
Input means for inputting chord information corresponding to the partial section;
For each partial section, a plurality of pitch information included in the partial section is associated with the input chord information based on the chord information detected by the detecting means and the chord information input by the input means. Conversion means for converting into pitch information that constitutes a chord to be performed;
When a chord composed of a plurality of pitch information included in the partial section to be converted by the converting means has a three-tone configuration and the input chord information has a four-tone configuration, the partial section to be converted Extraction that extracts pitch information corresponding to a predetermined part of the three chord constituent sounds from among the plurality of pitch information included as a pitch candidate for generating the fourth chord constituent sound Means,
A generating means for generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extracting means, wherein the key-on area and the immediately following part of the partial section immediately before the partial section to be converted Preferentially generating the fourth chord constituent sound as included in the overlapped portion of the key-on area of the section;
Replacement means for replacing the pitch information which is the basis for generating the fourth chord constituent sound by the generating means with the pitch information of the generated fourth chord constituent sound,
When converting a plurality of pitch information included in the partial section, the converting means converts the pitch information after the replacement for the pitch information replaced by the replacing means. Music information converter. Supply means for supplying performance data including a plurality of pitch information arranged in pronunciation order;
Dividing means for dividing the entire section of the performance data supplied by the supplying means into partial sections of a predetermined length;
Detecting means for detecting chord information of a chord composed of a plurality of pitch information included in the partial section for each partial section divided by the dividing means;
Input means for inputting chord information corresponding to the partial section;
For each partial section, a plurality of pitch information included in the partial section is associated with the input chord information based on the chord information detected by the detecting means and the chord information input by the input means. Conversion means for converting into pitch information that constitutes a chord to be performed;
When a chord composed of a plurality of pitch information included in the partial section to be converted by the converting means has a three-tone configuration and the input chord information has a four-tone configuration, the partial section to be converted Extraction that extracts pitch information corresponding to a predetermined part of the three chord constituent sounds from among the plurality of pitch information included as a pitch candidate for generating the fourth chord constituent sound Means,
A generating means for generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extracting means, wherein the key-on area and the immediately following part of the partial section immediately before the partial section to be converted Preferentially generating the fourth chord constituent sound as included in the overlapped portion of the key-on area of the section;
Adding means for adding the pitch information of the fourth chord constituent sound generated by the generating means to the sounding position of the chord composed of a plurality of pitch information included in the partial section,
The converting means converts the pitch information after the pitch information has been added by the adding means when converting a plurality of pitch information included in the partial section. .   3. The musical tone information conversion apparatus according to claim 1, wherein a predetermined part of the three chord constituent sounds has a pitch of 5 degrees. A computer-readable memory storing a program for causing a computer to execute a control method for controlling a musical tone information conversion apparatus comprising a CPU and a supply means for supplying performance data including a plurality of pitch information arranged in the order of pronunciation A medium ,
The control method is:
A division step in which the CPU divides the entire section of the performance data supplied by the supply means into partial sections of a predetermined length;
The CPU is, the divided every divided part section by step, a detection step of detecting the chord information chord comprising a plurality of tone pitch information included in the subinterval,
The CPU is, for each of the subinterval, a plurality of tone pitch information contained in the part section, on the basis of the chord information input by said detected by the detecting step chord information and input means, said input A conversion step for converting into pitch information that constitutes a chord corresponding to the chord information performed;
When the chord composed of a plurality of pitch information included in the partial section to be converted by the conversion step has a three-tone configuration and the input chord information has a four-tone configuration, the CPU Pitch candidates for generating the fourth chord constituent sound from the pitch information corresponding to a predetermined part of the three chord constituent sounds among the plurality of pitch information included in the partial section of An extraction step to extract as
The CPU is a generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step , and the key-on area of the partial section immediately before the partial section to be converted And preferentially generating the fourth chord constituent sound as included in the overlapped portion of the key-on area of the partial section immediately after,
Wherein the CPU, the pitch information is the basis in generating a fourth chord constituent notes by said generating step, and a replacement step of replacing the pitch information of the fourth chord component notes that are the product,
In the converting step , when converting a plurality of pitch information included in the partial section, the pitch information after the replacement is converted for the pitch information replaced by the replacing step . Storage medium . A computer-readable memory storing a program for causing a computer to execute a control method for controlling a musical tone information conversion apparatus comprising a CPU and a supply means for supplying performance data including a plurality of pitch information arranged in the order of pronunciation A medium ,
The control method is:
A division step in which the CPU divides the entire section of the performance data supplied by the supply means into partial sections of a predetermined length;
The CPU is, the divided every divided part section by step, a detection step of detecting the chord information chord comprising a plurality of tone pitch information included in the subinterval,
The CPU is, for each of the subinterval, a plurality of tone pitch information contained in the part section, on the basis of the chord information input by said detected by the detecting step chord information and input means, said input A conversion step for converting into pitch information that constitutes a chord corresponding to the chord information performed;
When the chord composed of a plurality of pitch information included in the partial section to be converted by the conversion step has a three-tone configuration and the input chord information has a four-tone configuration, the CPU Pitch candidates for generating the fourth chord constituent sound from the pitch information corresponding to a predetermined part of the three chord constituent sounds among the plurality of pitch information included in the partial section of An extraction step to extract as
The CPU is a generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step , and the key-on area of the partial section immediately before the partial section to be converted And preferentially generating the fourth chord constituent sound as included in the overlapped portion of the key-on area of the partial section immediately after,
Wherein the CPU, and a additional step of adding a pitch information of the fourth chord component tones generated by the generating step, the sound producing position of a chord consisting of a plurality of tone pitch information included in the subinterval,
In the conversion step, a storage medium in converting a plurality of tone pitch information included in the subinterval, and converting the pitch information after the pitch information is added by said adding step. 6. The storage medium according to claim 4, wherein a predetermined part of the three chord constituent sounds has a pitch of 5 degrees. A control method for controlling a musical tone information conversion apparatus comprising a CPU and a supply means for supplying performance data including a plurality of pitch information arranged in the order of pronunciation,
A division step in which the CPU divides the entire section of the performance data supplied by the supply means into partial sections of a predetermined length;
The CPU detects, for each partial section divided by the dividing step, chord information of a chord composed of a plurality of pitch information included in the partial section;
The CPU is, for each of the subinterval, a plurality of tone pitch information contained in the part section, on the basis of the chord information input by said detected by the detecting step chord information and input means, said input A conversion step for converting into pitch information that constitutes a chord corresponding to the chord information performed;
When the chord composed of a plurality of pitch information included in the partial section to be converted by the conversion step has a three-tone configuration and the input chord information has a four-tone configuration, the CPU Pitch candidates for generating a fourth chord constituent sound from pitch information corresponding to a predetermined part of the three chord constituent sounds among a plurality of pitch information included in the partial section of An extraction step to extract as
The CPU is a generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step, and the key-on area of the partial section immediately before the partial section to be converted And preferentially generating a fourth chord constituent sound that is included in the overlapped portion of the key-on area of the partial section immediately after,
A step of replacing the pitch information, which is the basis for generating the fourth chord constituent sound by the generating step, with the pitch information of the generated fourth chord constituent sound;
In the converting step, when converting a plurality of pitch information included in the partial section, the pitch information after the replacement is converted for the pitch information replaced by the replacing step. Control method. A control method for controlling a musical tone information conversion apparatus comprising a CPU and a supply means for supplying performance data including a plurality of pitch information arranged in the order of pronunciation,
A division step in which the CPU divides the entire section of the performance data supplied by the supply means into partial sections of a predetermined length;
The CPU detects, for each partial section divided by the dividing step, chord information of a chord composed of a plurality of pitch information included in the partial section;
The CPU is, for each of the subinterval, a plurality of tone pitch information contained in the part section, on the basis of the chord information input by said detected by the detecting step chord information and input means, said input A conversion step for converting into pitch information that constitutes a chord corresponding to the chord information performed;
When the chord composed of a plurality of pitch information included in the partial section to be converted by the conversion step has a three-tone configuration and the input chord information has a four-tone configuration, the CPU Pitch candidates for generating a fourth chord constituent sound from pitch information corresponding to a predetermined part of the three chord constituent sounds among a plurality of pitch information included in the partial section of An extraction step to extract as
The CPU is a generation step of generating a fourth chord constituent sound corresponding to the pitch information belonging to the pitch candidate extracted by the extraction step, and the key-on area of the partial section immediately before the partial section to be converted And preferentially generating the fourth chord constituent sound as included in the overlapped portion of the key-on area of the partial section immediately after,
The CPU includes an adding step of adding the pitch information of the fourth chord constituent sound generated by the generating step to a sounding position of a chord composed of a plurality of pitch information included in the partial section,
Wherein in the converting step, the control method characterized in that when converting a plurality of tone pitch information contained in the part section converts the pitch information after the pitch information is added by said adding step. 9. The control method according to claim 7, wherein a predetermined part of the three chord constituent sounds has a pitch of 5 degrees.

Images