JP4449370B2 - Automatic accompaniment generator and program - Google Patents

Description

  The present invention relates to an automatic accompaniment generation system that generates automatic accompaniment sound information according to a user's key press sound using an accompaniment pattern.

Conventionally, an arpeggiator or an accompaniment style is known as means for generating an automatic accompaniment using an accompaniment pattern. For example, there is an arpeggiator as shown in Patent Document 1.
JP 2001-22354 A

  The arpeggiator generates arpeggios (distributed chords) in response to key presses, and stores multiple key numbers (not note numbers corresponding to pitches) and their sounding timings as arpeggio patterns. ing. A number is assigned to a plurality of note numbers that are simultaneously pressed according to a predetermined rule (for example, in order of increasing pitches), and a note number assigned with a number corresponding to the key number in the arpeggio pattern is pronounced. The arpeggio is generated based on the keystroke sound so that it is generated at the timing.

  In automatic accompaniment generation by the arpeggiator, the name of the key pressed by the user is pronounced, but not all key pressing sounds are always reflected in the automatic accompaniment. For example, when a key is pressed more than the number of key presses assumed in the arpeggio pattern, there is a key press sound that is not pronounced. In addition, depending on the arpeggio pattern, there is a case where the sound is generated by being shifted by an octave, and the pitch of the key pressing sound is not always generated.

  In view of such circumstances, the present invention provides an input sound such as a key depression sound by a user's performance operation when generating automatic accompaniment sound information using accompaniment pattern data composed of a plurality of note information (note events). An object of the present invention is to provide an automatic accompaniment generating system capable of always reflecting all high accompaniment sound information.

According to the main feature of the present invention, accompaniment pattern data (Ap) composed of a plurality of note information (notes P1; P2;...) Associated in time series with respective timing information (Timing) representing sound generation timing is stored. Each of the note information (P1; P2;...) Includes storage means (2 to 4) including at least information (Gate) indicating the length of time, and pitch indication information (indicating the pitch of the accompaniment sound) Input means (5, PD) for inputting Ma, Mb,..., And each piece of note information (P1; P2;...) In accompaniment pattern data (Ap) stored in the storage means, with timing information (Timing). ) Among the reading means (S5 to S6) sequentially read out according to (1) and the note information (P1; P2;...) Read out by the reading means, the predetermined time length (for example, timing information (Timing) can be regarded as a chord (for example) For example, the selection means (S6: [0029]) that selects any one (for example, note P4) for a plurality of pieces of adjacent note information (for example, notes P4 and P5) that are close within a 32nd note length). To [0030], [0035]) and note information other than the adjacent note information (P4, P5) among the note information (P1; P2;...) Read by the reading means, and selected by the selecting means. Accompaniment tone information (N1a, N1b,...; N2a, N2b,...) Having pitches indicated by all the pitch indication information (Ma, Mb,...) Input from the note information (P4) by the input means. ...; ...) generating means (S6 to S18) and sounding means for generating accompaniment sounds based on accompaniment sound information (N1a, N1b, ...; N2a, N2b, ...; ...) generated by the generating means (7, S14) The apparatus (claim 1) and accompaniment pattern data (Ap) composed of a plurality of note information (P1; P2;...) Associated in time series with respective timing information (Timing) indicating the sound generation timing are stored. Storage means (2-4), each of note information (P1; P2;...) Including at least information (Gate) indicating time length, and pitch indication information for indicating the pitch of the accompaniment sound Each piece of note information (P1; in the accompaniment pattern data stored in the storage means is stored in a computer having an input means (5, PD) for inputting (Ma, Mb,...) And functioning as an automatic accompaniment generator. P2;... Are sequentially read out in accordance with the timing information (Timing), and the timing information among the reading information (S5 to S6) and the note information (P1; P2;...) Read in the reading step. For a plurality of adjacent note information (for example, notes P4 and P5) that are close within a predetermined time length (for example, 32nd note length) in which (Timing) can be regarded as a chord, any one (for example, note P4) ) In the selection step (S6: [0035]), note information other than the adjacent note information (P4, P5) among the note information (P1; P2;...) Read in the reading step, and the selection step Accompaniment sound information (N1a, N1b,...) Having pitches indicated by all the pitch instruction information (Ma, Mb,...) Input by the input means from the note information (P4) selected in (4). N2a, N2b,...), And accompaniment sounds based on the accompaniment sound information (N1a, N1b,...; N2a, N2b,...) Generated in the generation step. Steps to pronounce Automatic accompaniment generation program for executing the steps consisting the flop (S14) [Claim 4] is provided. Note that the parentheses are reference symbols, terms, parts, etc. of the examples added for convenience of understanding, and the same applies to the following.

  In the automatic accompaniment generating apparatus according to the present invention, the accompaniment pattern data (Ap) is included in the performance data in a predetermined format (Claim 2), and the generating means (S6 to S18) includes the note information (P1 , P2) when the time ranges overlap (Ta to Tb), the time length (Gate) of the accompaniment sound information generated from the previous note information (P1) is set according to the sound generation timing of the subsequent note information (P2). Limiting means (S9 to S10) for limiting may be provided.

  In the automatic accompaniment generating system according to the present invention (claims 1 and 4), accompaniment pattern data (Ap) comprising a plurality of note information (P1; P2;...) Is used, and these note information (P1; P2;...) Are associated with each timing information (Timing) in time series, and each includes at least information (Gate) indicating time length (length), and further includes, for example, pitch (Note), volume (Vel) Information. As the automatic accompaniment starts, the system sequentially reads each note information (P1; P2;...) In the accompaniment pattern data (Ap) according to the timing information (Timing), and the timing information (Timing) is regarded as a chord. For a plurality of adjacent note information (for example, notes P4 and P5) that are close within a predetermined time length (for example, 32nd note length) to be obtained, any one (for example, note P4) is selected (S6: [0029] to [0030], [0035]). Further, when the pitch instruction information (Ma, Mb,...) Of a desired pitch (for example, D3, F3) is input from the input means (5, PD) by the user's key pressing operation or the like, it is effectively input. A desired pitch indicated by all pitch indication information (Ma, Mb,...) Is acquired. Then, from each of the read note information (P1; P2;...), But for the adjacent note information (P4, P5), each pitch (Note) is selected from any one selected (P4). The accompaniment sound information (N1a, N1b,...; N2a, N2b,...) Having the desired pitch indicated by the whole pitch instruction information (Ma, Mb,...) Is used. Is generated (S13, S17), and the corresponding automatic accompaniment sound is generated through the sound source section (7) (S14).

  In this invention, in this way, when accompaniment sound information is automatically generated according to a user's performance operation using accompaniment pattern data including a plurality of note information including timing information, each accompaniment pattern data is generated according to each timing information. Read the note information in sequence, generate information such as length and volume of the read note information, and accompaniment sound information with all input pitches (key press tone pitches) based on user operation, and pronounce it Therefore, all the user input pitches (key press tone pitches) can be reflected in the automatically used accompaniment tone. Furthermore, since all the user input pitches (key press tone pitches) are immediately pronounced on the note information pattern as they are, the user plays the correspondence between his key press state and the generated automatic accompaniment sound. It can be easily recognized. In addition, note information in the accompaniment pattern data that is the source of the automatic accompaniment sound may be one note for the sound generation timing represented by each timing information, and therefore the amount of data is reduced compared to conventional various accompaniment pattern data. Can be saved. Furthermore, since the accompaniment sound information is generated by selecting any one of a plurality of adjacent note information that are close within a predetermined time length in which the timing information can be regarded as a chord, the chord state is mixed. Even with accompaniment pattern data, accompaniment sound information having a desired pitch can be generated and pronounced.

  The accompaniment pattern data used in the system of the present invention is included in performance data in a predetermined general format represented by MIDI, and the note information is described in the general format (claim 2). The creator of the pattern data can use a part of any piece of music data that is generally available for electronic musical instruments or the like, or an actual musical performance stored by an electronic musical instrument or the like.

In the present invention, in addition, in the case where temporal overlap (Ta to Tb) occurs in the note information (P1, P2) whose sound generation timings are temporally changed, the preceding accompaniment sound information generated according to the previous note information (P1). Is limited in accordance with the timing information (Timing = Ta) of the later note information (P2) (claim 3). For example, when new subsequent accompaniment sound information is generated according to the later note information (P2), the automatic accompaniment sound that is already sounding at that time is turned off (S9 to S10). As a result, the number of simultaneous sounds of the automatic accompaniment sound matches the number of all user input sounds (number of key press sounds) and does not exceed this. Therefore, problems such as exceeding the maximum number of simultaneous sounds or increasing the volume until the automatic accompaniment sound that is already sounding is turned off by the preceding accompaniment sound information do not occur.

〔System configuration〕
FIG. 1 shows a hardware configuration example of an automatic accompaniment generation system according to an embodiment of the present invention. In this example, an automatic accompaniment generating / generating apparatus uses a music information processing apparatus dedicated to music such as an electronic musical instrument, but a general-purpose information processing apparatus provided with a performance input and musical tone generating function in a personal computer may be used. . This automatic accompaniment generator includes a central processing unit (CPU) 1, a random access memory (RAM) 2, a read only memory (ROM) 3, an external storage device 4, an input operation unit 5, a display unit 6, a sound source unit 7, a communication Elements such as an interface (communication I / F) 8 are provided, and these elements 1 to 8 are connected to the bus 9.

  The CPU 1 that controls the entire apparatus executes various processes using a timer clock in accordance with a predetermined control program, and the RAM 2 is a process for temporarily storing various information used in these processes. Functions as a buffer. For example, at the time of processing by the automatic accompaniment generation program, key-press information (Ma, Mb,...), Accompaniment note information (N1a, N1b,...; N2a, N2b,...), Pattern note information (P1; P2;... ) Are stored on the RAM 2 (KeyOnList, DataList, PatternNote). The ROM 3 stores various data such as various control programs including an automatic accompaniment generation program and accompaniment pattern data (Ap) including pattern note information in the CPU 1.

  The external storage device 4 is a hard disk (HD), a compact disk read only memory (CD-ROM), a flexible disk (FD), a magneto-optical (MO) disk, a digital multipurpose disk (DVD), a memory card, etc. Storage means using a storage medium. Various control data such as an automatic accompaniment generation program, various data such as song data and accompaniment pattern collections can be stored not only in the ROM 3 but also in the external storage device 4.

  For example, when the control program is not stored in the ROM 3, the control program is stored in the ROM 3 by storing the control program in the external storage device 4 such as an HD or a CD-ROM and reading it into the RAM 2. The CPU 1 can be operated in the same manner as described above, and the control program can be easily added or upgraded. For example, a desired automatic accompaniment generating apparatus can be realized by installing programs and various data used for automatic accompaniment generation processing and the like.

  The input operation unit 5 includes a panel operation element such as a mouse and a key, a performance operation element such as a performance keyboard, and a detection circuit. The detection circuit detects the contents of the setting / performance input operation by the operation element, and setting information and a key-press sound. Introduce information into the system itself. For example, you can use the panel controls to set the operating status of this system, such as automatic accompaniment start / stop, input various information necessary for automatic accompaniment generation, or perform performance data according to the operation of the performance controls. Can be entered. Here, in the performance keyboard of the performance operator, a plurality of keys set as accompaniment keys or accompaniment keys (hereinafter referred to as “accompaniment keys”) are key press sound information (indicating the pitch of the accompaniment sound) ( Used to input pitch indication information). The display unit 6 controls display contents of the display (display such as CRT and LCD) 10 and lighting states of various indicators in accordance with commands from the CPU 1, and performs display assistance for the operation of the input operation unit 5.

  The sound source unit 7 includes a sound source (including software) and an effect imparting DSP, generates a musical sound signal corresponding to input performance data and automatic accompaniment data processed by the CPU 1, and is connected to the sound source unit 7. 11 includes a D / A converter, an amplifier, and a speaker, and generates a musical sound based on a musical sound signal from the sound source unit 7. That is, the sound source unit 7 and the sound system 11 form a musical sound generation unit, and, for example, in an automatic accompaniment generation process, an accompaniment sound can be generated according to automatic accompaniment data generated according to the key depression sound information.

  The communication I / F 8 shown in the figure collectively represents various interfaces connected to a local area network (LAN), a general communication network such as the Internet and a telephone line, or a MIDI network. Various necessary information can be exchanged with various external devices PD such as other computers and MIDI devices. For example, when the control program and various data are not stored in this apparatus, the control program or the like can be downloaded from another computer PD via the communication I / F 8. Further, the performance data and performance sound based on the performance operation can be taken into the external device PD from the communication I / F 8 using another keyboard, a MIDI performance operation device including a stringed instrument and a wind instrument.

[Example of accompaniment pattern data and automatic accompaniment generation]
In the automatic accompaniment generation system according to one embodiment of the present invention, when an accompaniment key 5 is operated (key depression), the accompaniment pattern data Ap stored in the ROM 3 or the external storage device 4 is used in accordance with the automatic accompaniment generation program. Accompaniment data reflecting all key depression pitches can be automatically generated, and accompaniment sounds can be emitted from the tone generation units 7 and 11. 2 and 3 show accompaniment pattern data and automatic accompaniment generation examples according to one embodiment of the present invention in numerical values and time charts (only for the first period).

  Here, the outline of automatic accompaniment generation in this system will be described very simply with reference to FIG. In this automatic accompaniment generation system, accompaniment pattern data Ap composed of a plurality of note events P1 to P5 is used, and these note events P1 to P5 are related in time series by timing information Timing that each has. When key-pressing sound information Ma, Mb,... That designates an arbitrary pitch (for example, D3, F3) is input from the accompaniment key 5, each piece is sequentially read out from the accompaniment pattern data Ap according to the respective timing information Timing. The pitches of note events P1; P2;... Are changed to pitches (D3, F3) indicated by all key input tone information Ma and Mb that are effectively input, and note events N1a representing automatic accompaniment sounds. , N1b,..., N2a, N2b,... Are generated one after another, and accompaniment sounds corresponding to these note events N1a, N1b,.

  Hereinafter, FIG. 2 and FIG. 3 will be described in more detail. FIG. 2 (1) shows an example of accompaniment pattern data Ap for one part. As shown in FIG. 2 (1), the accompaniment pattern data Ap includes a plurality of accompaniment patterns. Note events (hereinafter referred to as “pattern notes”) P1, P2,... Are arranged in a time series over a predetermined time (hereinafter referred to as “pattern time”) Tp. Each pattern note P1, P2,... (Each row) includes information such as timing information Timing, gate time information Gate, pitch information Note, velocity information Vel (hereinafter, the term “information” may be omitted). When these pitches are developed over time, the result is as shown in FIG.

  Such accompaniment pattern data Ap is described in a predetermined general-purpose format such as MIDI, and may be quoted from existing music data or recorded actual performance. . 2 (1) and 3 (1) exemplify the accompaniment pattern data for the pattern time Tp = 1 bar for the sake of simplicity, but of course it may be longer, of course, A shorter pattern may be used.

  Here, the timing Timing and the gate time Gate are time information indicating the sound generation start time (timing) and the length of the note-on time of each pattern note P1, P2,. In the example shown in the drawing, it is represented by the number of clocks in which one beat is 480 clocks, and the pattern time Tp = 1920 clocks. The gate time Gate (P1) = “0490” of the first pattern note P1 exceeds the timing Timing (P2) = “0480” of the second pattern note P2, and a legato-like performance expression is contemplated. .

  Next, the pitch note and the velocity Vel are information representing the pitch and volume (key press speed or intensity) of the pattern notes P1, P2,. In the example shown in the figure, the pitch name “C4”, “D4”,... Or the note number (number) value is used for the pitch information Note, and the velocity value is used for the volume information Vel. Accordingly, the minimum value = “0” or “1” to the maximum value = “127” is set.

  In this system, when a user operates (presses) an arbitrary number of accompaniment keys of arbitrary pitches and key depression sound information Ma, Mb,... The key-pressing sound information Ma, Mb,... Input to the are all used as pitch instruction information for instructing the pitch of the accompaniment sound. For example, if two keys corresponding to the pitches “D3” and “F3” are pressed against the accompaniment pattern data Ap in FIGS. 2 (1) and 3 (1), the automatic accompaniment generation program .. (2) as the pattern notes P1; P2;... Associated with each timing Timing are sequentially read from the accompaniment pattern data Ap according to the timing Timing of the pattern notes P1; As shown in FIG. 3 (2), the accompaniment note events N1a, N1b; N2a, N2b;.. Automatically generated as An.

  That is, every time the read timing (Tr) based on the timing Timing of each pattern note P1; P2;... Arrives, each pattern note P1; Vel, and automatic accompaniment note events (hereinafter referred to as “accompaniment notes”) N1a, N1b; N2a, N2b having key depression pitches “D3”, “F3” (note numbers “62”, “65”) ; Is generated. Such generation of accompaniment notes is continuously performed while the key pressing operation of the same accompaniment key is continued, and is repeated cyclically with the pattern time Tp of the accompaniment pattern data Ap as one cycle.

  Here, timing Timing (Pn) = Ta of a certain pattern note Pn is consumption time t = Tb = Timing (Pn-1) + Gate (Pn−) when the gate time Gate (Pn−1) of the previous pattern note Pn−1. If the accompaniment note Nn-1 is generated as it is in accordance with the pattern note Pn-1 in the case before 1), the sound of the accompaniment note Nn-1 is continued during the timings Ta to Tb, and it corresponds to the pattern note Pn. The next accompaniment note Nn is generated and the number of pronunciations is doubled. In such a case, the gate time Gate (Nn-1) of the preceding accompaniment note Nn-1 is limited so that the accompaniment notes Nn-1, Nn do not overlap. Therefore, for example, at the generation timing Ta of the next accompaniment note Nn, all the accompaniment notes Nn-1 that are already sounding are muted by the tone generator 7 according to the gate time Gate (Pn-1) of the pattern note Pn-1. Alternatively, the gate time Gate (Nn-1) of the accompaniment note Nn-1 itself to be sent to the sound source unit 7 is set to a length suitable for the timing Ta of the next pattern note Pn [Gate (Nn-1) = Timing (Pn) − Timing (Pn-1)].

  In the illustrated example, as indicated by “(Note 1)” in the time chart of FIG. 3, the consumption time Tb = 490 (clock) of the gate time Gate (P1) of the first pattern note P1 is the second pattern note P2. Timing Timing (P2) = 480 = Ta is exceeded, and both pattern notes P1, P2 overlap during the time Ta-Tb = 480-490. Therefore, as shown by the arrows, the first accompaniment notes N1a and N1b generated according to the first pattern note P1 do not overlap the second accompaniment notes N2a and N2b generated next according to the second pattern note P2. The sound is muted at the timing Ta, and the gate time Gate is shortened to a length [Gate (N1a, N1b) = Timing (P2) −Timing (P1) = 480] that matches the timing of the second pattern note P2.

  In the accompaniment pattern data Ap, a plurality of pattern notes having substantially the same timing Timing (for example, close to each other within a 32nd note length (60 clocks)) are regarded as chords and selected according to an appropriate predetermined method. Accompaniment notes are generated only for any one pattern note, and other pattern notes are not reflected in automatic accompaniment generation.

  In the illustrated example, as shown in “(Note 2)” in the time chart of FIG. 3, the fourth and fifth pattern notes P4 and P5 have the same start timing Tc [Tc = Timing (P4) = Timing (P5) = 1440], accompaniment notes N4a and N4b are generated according to the fourth pattern note P4 [Gate (P4)> Gate (P5)] having a large time length, which is regarded as a chord, and the fifth pattern note P5 is used for automatic accompaniment generation. Not reflected. That is, in this example, the method of prioritizing the pattern note with the longest time is adopted, and for the fifth pattern note P5, the time range Tc to Td [Td = Timing (P5) + Gate (P5) = 1660] is the fourth pattern. Since it is included in the time range Tc to Te [Te = Timing (P4) + Gate (P4) = 1670] of the note P4, no automatic accompaniment sound is generated.

[Operation flow for automatic accompaniment generation]
4 and 5 are flowcharts showing an operation example of automatic accompaniment generation according to one embodiment of the present invention. When an automatic accompaniment start is instructed by a user operation of the input operation unit (panel operation unit) 5, and the operation flow of automatic accompaniment generation starts, the CPU 1 first performs initial setting according to the user operation or default (step S1: Note that In the following, the symbol “S...” Alone represents the corresponding step). Initial settings include, for example, setting of accompaniment pattern data Ap used for automatic accompaniment generation, setting of accompaniment keys (for example, setting of split points for defining accompaniment key ranges, specification of accompaniment keyboards, etc.), automatic accompaniment There is a setting of the maximum number of simultaneous pronunciations (for example, 16). The accompaniment pattern data Ap can be set by selecting an accompaniment pattern portion of a desired track from song data prepared in advance in the ROM 3 or the external storage device 4, selecting a desired accompaniment pattern from a collection of accompaniment patterns, A method of registering use after recording an accompaniment pattern by an actual performance is used.

  Next, the CPU 1 sets the keying sound information storage area KeyOnList, the accompaniment note information storage area DataList, and the pattern note information storage area PatternNote (hereinafter, each area is represented by a reference symbol alone) on the RAM 2 and initializes them. (S2) and a timer is started (S3). Note that KeyOnList and DataList have an element number of “0” in the initial state, and the maximum number of elements is the maximum number of simultaneous pronunciations. The data structure of each note event (pattern note) of the accompaniment pattern data Ap stored in PatternNote may be the same description format as normal note event information, and an invalid value is set in the initial state. Then, the following processing steps (S4 to S18) are entered by a timer interruption at a predetermined time interval.

  In this processing stage, first, it is determined whether or not an automatic accompaniment stop instruction has been given (S4). When there is no automatic accompaniment stop instruction (S4 → NO), the initial setting (S1) is currently set. It is determined whether or not it is the read timing Tr to read the pattern note of the accompaniment pattern data Ap (S5). Here, the read timing Tr is a value obtained by adding the timing Timing of each pattern note to n times the pattern period Tp of the accompaniment pattern data Ap, and is represented by Tr = n × Tp + Timing. Note that the value n is initially 0, but is a natural number that increases by “1” every time the time progression of the automatic accompaniment exceeds the pattern time Tp.

  If it is the read timing Tr (S5 → YES), the pattern note is read from the accompaniment pattern data Ap and stored in PatternNote (S6). In the example of FIGS. 2 and 3, the processing is started at the performance start time t = 0 (clock), and the first pattern note P1 is immediately read out at the read timing Tr = Timing (P1) = “0000” equal to this time point. Then, the pattern notes P2, P3,... Are sequentially read at every read timing Tr = n × 1920 + Timing (n = 0 or natural number).

  Note that the determination of the read timing Tr (S5) is performed with a certain time width (the time width may be set). Also, in the pattern note storage step (S6), when all the pattern notes that can be regarded as chords are read out at the same time or within a predetermined margin time (for example, 60 clocks) and a plurality of events are read out, Any one pattern note selected by an appropriate method is stored in PatternNote. For example, as shown in FIGS. 2 and 3, when the fourth and fifth pattern notes P4 and P5 are read at the timing t = Tc = 1440, the fourth pattern note P4 is stored in the PatternNote.

  Subsequently, the KeyOnList is initialized (S7), and the current key pressing sound information based on the key pressing of the accompaniment key 5 is acquired and stored in the KeyOnList (FIG. 5: S8). In this case, the key depression exceeding the maximum simultaneous number of sounds set in the initial setting step (S1) is ignored, and the key depression equal to or less than this number of sounds is validated. In the example of FIGS. 2 and 3, the first read timing Tr = Timing (P1) = 0, and the key press sound pitch “D3” (62), “F3” is stored in the key press sound storing step (S8). (65) is stored in the KeyOnList.

  Next, it is determined whether or not there is an automatic accompaniment sound being pronounced (S9). Here, when the automatic accompaniment sound by the accompaniment note in the DataList according to the previous pattern note is continuously sounding (S9 → YES), the sound source unit 7 is turned off so as to turn off the automatic accompaniment sound being sounded. Command (S10) and initialize DataList (S11). 2 and 3, the second pattern note P2 is stored in the PatternNote at the second read timing Tr = Timing (P2) = “0480” = Ta (S6). An automatic accompaniment sound is being generated by the accompaniment notes N1a and N1b of the DataList based on one pattern note P1 (S9 → YES). Therefore, the CPU 1 immediately turns off the fully automatic accompaniment sound to the sound source unit 7 at this time Ta (S10) and initializes the DataList (S11). Accordingly, the gate time Gate of the accompaniment notes N1a and N1b is consequently shortened to the value Ta.

  Further, it is determined whether or not the key depression timing of the accompaniment key 5 is later than the time Ts indicated by the timing Timing of the pattern note stored in PatternNote (S12). This time Ts is equal to the read timing Tr. In other words, the time Ts may be slightly delayed from the read timing Tr when physically speaking, but it is a negligible delay in handling music data and is substantially the same as the read timing Tr.

  Here, when the key is pressed before the start time Ts (S12 → NO), the start time Ts corresponding to the timing Timing of the pattern note, the length Gate, and the volume for each key press sound information in the KeyOnList Accompaniment notes having Vel and key depression pitches are generated (in short, only the pitch information Note of the pattern note is rewritten with the changed pitch), and each is stored in the DataList (S13). Then, the accompaniment notes in the DataList are sent to the sound source unit 7 and a reproduction process for generating the corresponding automatic accompaniment sound from the sound system 11 is performed (S14), and then the process returns to the stop instruction determination step (FIG. 4: S5).

  Next, returning to the read timing determination step and determining that it is not the read timing Tr (S4 · NO → S5 · NO), subsequently, it is determined whether or not the current key press state is the same as the key press tone of the KeyOnList. Judgment is made (S15). Here, when the same key pressing state is maintained (S15 → YES), the process returns to the stop instruction determination step (S4), and the above-described operation (S6 to S14) every time the read timing Tr arrives (S5, YES). repeat.

  In the examples of FIGS. 2 and 3, since the accompaniment keys D3 and F3 are pressed at the start time t = Ts = 0 (clock) of the first pattern note P1 (S12 → NO), only the pitch information Note is pressed. After storing the first accompaniment notes N1a and N1b rewritten to the key pitches D3 and F3 in the DataList (S13), the pattern notes P2, P3,. It is stored in a manner (S13). However, the timing Timing stored in the DataList is the start time Ts = n × 1920 + Timing (n = 0 or natural number) of each accompaniment note.

  When it is determined in the key pressing state determination step (S15) that a key is pressed that is different from the key pressing pitch of the KeyOnList (S15 → NO), the PatternNote pattern is also displayed when the key is changed. It is checked whether or not the note is valid, that is, whether or not the pattern note stored in PatternNote is during the pronunciation duration (S16). If the gate time Gate of the pattern note is not consumed and the sound generation period of the pattern note is maintained (S16 → YES) at the time of key change, the above-described KeyOnList initialization to DataList initialization are performed. The process proceeds to the key pressing timing determination step (S12) through the conversion step (S7 to S11). The timing comparison in the pattern note validity determination step (S16) and the key pressing timing determination step (S12) may be performed with a certain time width. Each time width may be settable.

  In this case, in the key pressing timing determination step (S12), it is determined that the key has been pressed after the pattern note start time Ts stored in PatternNote, that is, while the corresponding accompaniment note is sounded (S12 → YES). The timing and length of the pattern note are adjusted according to the key pressing timing, and for each key pressing sound information in the KeyOnList, the adjusted timing Timing and length Gate, the volume Vel of the pattern note, and the pressing Accompaniment notes having key pitches are generated and stored in the DataList (S17). And after performing the reproduction | regeneration processing of the said accompaniment note (S14), it returns to a stop instruction | indication determination step (S4).

  For example, in FIG. 3, the start time Ts of the second pattern note P2 = Timing (P2) + 1920 × 0 = 480 = after Ta or before the end time of the pattern P2 = 960 (the second accompaniment notes N2a and N2b are sounded) When the key is pressed at time t = Tf (middle), it is determined that the key has been changed in the middle of the accompaniment note (S16 → YES), the timing Timing of the pattern note P2 is changed to the key pressing timing Tf, and the gate Adjustment is made to change the time Gate to Gate (P2)-(Tf-Ta), and accompaniment notes with the velocity Vel set to Vel (P2) = 72 and the note Note set as the new key-pressed pitch are stored in the DataList ( S14).

  If another accompaniment key is pressed before the pattern note start time Ts stored in PatternNote [S15 (NO) → S16 (NO)], the process proceeds to the pattern note storage step. [S4 (NO) → S5 (YES) → S6], and through KeyOnList initialization to DataList initialization steps (S7 to S11), it is determined that the key is pressed before the start of sounding of the accompaniment note (S12 → NO). Then, only the pitch Note of the pattern note is rewritten to the key depression pitch (timing Timing is the start time Ts) and stored in the DataList (S13).

  2 and 3, after the initial key depression (D3, F3), another accompaniment key is depressed before the start time Ts of any pattern note P1 to P5 stored in PatternNote. In this case, for example, in the blank time range in which there is no note event between the preceding and following pattern notes P3 to P4, for example, between the times Tg and Tc [Tg = Timing (P3) + Gate (P3) = 1210] in FIG. If the key is locked, the above process [S15, S16, S4 (NO) → S5 (YES) → S6 to S12 (NO)] is performed, and the pattern note P4 start time Ts = Timing (P3) = Tc = 1440, accompaniment notes corresponding to the same pattern notes are stored in the DataList.

  If there is no key depression at the start of the automatic accompaniment, the pattern note is read and the PatternNote is updated at every read timing of the accompaniment pattern data Ap. During this time, KeyOnList remains “empty” at initialization (S7), and DataList is also in an “empty” state in which accompaniment note events are not substantially stored.

  As described above, after performing the reproduction process of a predetermined accompaniment note (S14), when returning to the stop instruction determination step (S4), automatic accompaniment due to the user operation of the panel operator 5 or the end of progression of song data, etc. When a stop instruction is detected (S4 → YES), if there is an automatic accompaniment sound that is sounding, the sound source unit 7 is instructed to turn it off (S18), and this automatic accompaniment generation flow ends.

[Various Embodiments]
Although a preferred embodiment of the present invention has been described above, this is merely an example, and the present invention can be variously modified without departing from the spirit thereof and can be implemented in various modes. it can. For example, the accompaniment pattern data assumes one part (track), that is, one tone color or instrument sound in the embodiment, but may be a plurality of parts (tracks).

  The note information string in the accompaniment pattern data may be a single note state or a chord state. When there is a chord in the accompaniment pattern data, the length and volume information of one sound in the chord may be used for automatic accompaniment generation. At this time, various appropriate methods can be applied to the selection of one sound, for example, the first read in the chord or the longest time length. As a method for determining the chord state, for example, a plurality of note information whose timing information is within a predetermined time width of about 32nd notes may be determined as one chord.

  Although the upper limit of the number of simultaneous pronunciations is 16 in the embodiment, it is not limited to this. Or it may be user-configurable. In addition to the specific keyboard (accompaniment keyboard) or part of the keyboard range (accompaniment keyboard range), the pitch instruction information can be input using non-keyboard type performance operations such as stringed instruments and wind instruments. A child or a device may be used. The pitch instruction information may be played in real time, or may be received sequentially in the MIDI format or the like.

  In addition, regarding the input of the pitch instruction information, in the embodiment, the presence or absence of a key pressing sound is determined in the actual key pressing state. For example, once the key is pressed, the input state of the pitch instruction even if the key is released May be configured to hold (hold) the key and keep the key pressed until a different key pressing or release command is issued due to a stop operation.

It is a block diagram which shows the hardware structural example of the automatic accompaniment production | generation system by one Example of this invention. It is a figure explaining numerically the example of the accompaniment pattern data by one Example of this invention, and automatic accompaniment production | generation. It is a figure explaining the example of the accompaniment pattern data by one Example of this invention and the automatic accompaniment production | generation for 1 pattern time (1 bar) with a time chart. It is a part of flowchart which represents the operation example of the automatic accompaniment production | generation by one Example of this invention. It is the other part of the flowchart showing the operation example of the automatic accompaniment production | generation by one Example of this invention.

Explanation of symbols

P1; P2; ... Pattern notes (note information) constituting the accompaniment pattern data Ap,
Tp pattern time (repetition period),
Ma, Mb, ... Key pressing sound information (pitch instruction information),
N1a, N1b; N2a, N2b; ... Accompaniment notes (accompaniment sound information) constituting the automatic accompaniment data An.

Claims (4)

Storage means for storing accompaniment pattern data composed of a plurality of note information associated in time series with respective timing information representing pronunciation timing, each of the note information including at least information representing a time length;
Input means for inputting pitch instruction information for instructing the pitch of the accompaniment sound;
Reading means for sequentially reading each note information in the accompaniment pattern data stored in the storage means according to the timing information;
Among the note information read out by the reading means, for a plurality of adjacent note information that is close within a predetermined time length in which the timing information can be regarded as a chord, selection means for selecting any one of them,
Instructed by all pitch indication information input by the input means from the note information other than the proximity note information of the note information read by the reading means and the note information selected by the selection means Generating means for generating accompaniment sound information having a pitch to be played,
An automatic accompaniment generating apparatus comprising: sound generation means for generating an accompaniment sound based on the accompaniment sound information generated by the generation means.   2. The automatic accompaniment generating apparatus according to claim 1, wherein the accompaniment pattern data is included in performance data in a predetermined format.   The generating means includes a restricting means for restricting a time length of the accompaniment sound information generated from the previous note information according to a sound generation timing of the subsequent note information when time ranges of the note information whose sound generation timings precede and follow overlap. The automatic accompaniment generator according to claim 1 or 2, further comprising: Storage means for storing accompaniment pattern data composed of a plurality of note information time-sequentially associated with respective timing information representing pronunciation timing, each of the note information including at least information representing a time length; and In addition, a computer functioning as an automatic accompaniment generating apparatus, comprising input means for inputting pitch instruction information for instructing the pitch of an accompaniment sound,
A step of sequentially reading out each note information in the accompaniment pattern data stored in the storage means in accordance with the timing information;
Among the note information read out in the reading step, for a plurality of adjacent note information within a predetermined time length in which the timing information can be regarded as a chord, a selection step for selecting any one of them,
Instructed by all the pitch indication information inputted by the input means from the note information other than the proximity note information of the note information read in the reading step and the note information selected in the selection step Generating step for generating accompaniment sound information having a pitch to be played;
An automatic accompaniment generation program for executing a procedure comprising a sound generation step for generating an accompaniment sound based on the accompaniment sound information generated in the generation step.

Images