JP5200895B2 - Performance data processing apparatus and program - Google Patents

Description

  The present invention relates to a performance data processing system that performs quantization processing of performance data including note information that defines timing in predetermined musical time units.

2. Description of the Related Art Conventionally, an electronic music apparatus such as a sequencer has a quantize function for correcting the timing of note information in original performance data when producing performance data such as MIDI. For example, in the automatic performance recording device disclosed in Patent Document 1, the note-on time of the melody sound (C, D, E,...) Is the highest among the specified beat timings (A, B, B, B, A,...). It is corrected at a close timing [FIG. 3 (a) → (c)]. Further, in the performance data processing apparatus of Patent Document 2, the sound generation timing of the quantization target note in the automatic performance data is selected by the user from a plurality (6) candidates created according to the positional relationship with the step timing. Rewritten.
JP-A-3-282595 JP-A-7-56564

  As described above, in the conventional quantization process, the note-on timing to be corrected is determined only by the temporal positional relationship between the original note-on event and the reference timing (beat timing or step timing). It is not a quantize that takes into account the sound length in combination with off. In other words, the position of the note-on event at the head of the note to be quantized has been corrected regardless of the position of the subsequent note-off event (that is, not depending on the sound generation period), so that the sound generation period continues for a long time. However, the presence of the continuous part is ignored and the sound is processed in the same way as a short sound, and even if performance data obtained by quantizing is reproduced, it is not musically expected.

  The prior art also has a disadvantage when analyzing performance data. For example, inadequate quantization when counting note events in musical time units such as beats and half beats and analyzing the rhythm pattern from the distribution of the frequency of note event occurrences for various uses (arrangement etc.) Will be made.

  In view of such inconveniences, the present invention can perform quantization processing utilizing the musical characteristics of the original performance data by considering the note length of the note information in the original performance data. An object of the present invention is to provide a performance data processing system capable of obtaining a practical quantize function.

  According to the main feature of the present invention, note information [N (Δ, Nn, Δf) that defines note-on timing (tn: tna,..., Tnc, tnd) and note-off timing (tf: tfa,..., Tfc, tfd). , Nf): Performance data acquisition means (S2 to S4) for acquiring performance data (PD) including Na,..., Nc, Nd], and note-on for the performance data (PD) acquired by the performance data acquisition means Reference timing (tg:..., Tgn−) in which a period (eg, Tc; Td) from timing (tn) to note-off timing (tf) appears in a predetermined musical time unit (Tg: one beat, half beat, etc.) , Tgn,..., Note detection means (S5 = YES) for detecting note information (eg, Nc; Nd), and notes detected by the note detection means For the information (Nc; Nd), the first time (Tng) from the note-on timing (tnc; tnd) to the next reference timing (for example, tgn) and the reference timing (tgn) to the note-off timing (tfc, tfd) Time calculating means (S7 to S8) for calculating the second time (Tgf) until time comparison means (S9) for comparing the first time (Tng) and the second time (Tgf) calculated by the time calculating means. And the reference timing (tgn−1; tgn) immediately before or immediately after the note-on timing (tnc; tnd) of the note information (Nc; Nd) detected by the note detection unit according to the comparison result by the time comparison unit. The selected reference timing (tgn−1; tgn) is used as the target time position for quantization, and the note on timing is selected. A performance data processing apparatus (Claim 1) and a note-on timing (tn) including quantizing execution means (S10 to S14) (tnc → tnc ′, tnc ″; tnd → tnd ′, tnd ″). : Tna,..., Tnc, tnd) and note information [N (Δ, Nn, Δ, Nf): Na,..., Nc, Nd] defining the note-off timing (tf: tfa,..., Tfc, tfd) Performance data acquisition steps (S2 to S4) for acquiring performance data (PD) including the period from note-on timing (tnc; tnd) to note-off timing for the performance data (PD) acquired in the performance data acquisition step Reference timing (tg:..., Tgn) in which (for example, Tc; Td) appears in a predetermined musical time unit (Tg: one beat, half beat, etc.) , Tgn,... For note information (for example, Nc; Nd), note on timing (S5 = YES) and note information (Nc; Nd) detected in the note detection step. Time to calculate a first time (Tng) from tnc; tnd) to the next reference timing (for example, tgn) and a second time (Tgf) from the reference timing (tgn) to the note-off timing (tgc, tgd) According to the comparison result in the time comparison step (S9) and the time comparison step (S9) for comparing the first time (Tng) and the second time (Tgf) calculated in the calculation step (S7 to S8) and the time calculation step , Immediately before the note-on timing (tnc; tnd) of the note information (Nc; Nd) detected in the note detection step or A later reference timing (tgn-1; tgn) is selected, and the note-on timing is changed using the selected reference timing (tgn-1; tgn) as a target time position for quantization (tnc → tnc ′, tnc ″; There is provided a performance data processing program (Claim 3) for causing a computer to execute a procedure consisting of tnd → tnd ′, tnd ″) and quantize execution steps (S10 to S14). Note that the parentheses indicate reference symbols, terms, parts, and the like of the examples added for convenience of understanding, and the same applies to the following.

  The performance data processing apparatus according to the present invention further includes strength setting means (S1) for setting the quantize strength (ST), and the quantize execution means (S10 to S14) are detected by the note detection means (S5 = YES). Time (tnc to tgn-1; tnd to tgn) until the reference timing (tgn-1; tgn) selected as the target time position is set to the note-on timing (tnc; tnd) of the note information (for example, Nc; Nd). ) Is changed (tnc → tnc ″; tnd → tnd ″) by a ratio corresponding to the quantized intensity (for example, ST = 50%) set by the intensity setting means (step 12). Item 2] can be configured as follows.

  In the performance data processing system according to the present invention (claims 1 and 3), note information defining note-on timing (tn: tna,..., Tnc, tnd) and note-off timing (tf: tfa,..., Tfc, tfd) Performance data (PD) including [N (Δ, Nn, Δ, Nf): Na,..., Nc, Nd] is converted into predetermined musical time units (Tg:..., Tgn−1,. When quantizing at tgn,..., first, for the original (original) performance data (PD), the period from the note-on timing (tnc; tnd) to the note-off timing (for example, Tc; Td) is the reference timing (tg). :..., Tgn−1, tgn,...) Is detected (for example, Nc; Nd) (S5 = YES). This reference timing (tg) is called a grid (Grid:..., Grid n-1, Grid n,...) And appears in a predetermined musical time unit (Tg) according to a predetermined tempo. Next, for the detected note information (Nc; Nd), the first time (Tng) from the note-on timing (tnc; tnd) to the next reference timing (tgn) and the reference timing (tgn) to the note-off timing (tgn) The second time (Tgf) until tgc, tgd) is calculated (S7 to S8), and the calculated first time (Tng) and second time (Tgf) are compared (S9). Then, the reference timing (tgn-1; tgn) immediately before or immediately after the note-on timing (tnc; tnd) of the detected note information (Nc; Nd) is selected according to the comparison result, and the selected reference is selected. The note-on timing is changed (tnc → tnc ′, tnc ″; tnd → tnd ′, tnd ″) with the timing (tgn−1; tgn) as the target time position for quantization, and the quantization is executed (S10 to S14).

  As is well known, the principle of quantize maps time points of an event such as note-on timing to discrete time points such as a grid (reference timing) to time the event. When creating new performance data by quantizing the original performance data, as a basic mapping operation, the on-timing of notes (notes) in the original performance data immediately before the note-on timing is used. Map to the grid (notes are not moved if the note-on timing matches the grid). Therefore, if the basic mapping operation is applied as it is to a note across the grid, the note-on point is mapped to the grid in front of the grid even if the note tip slightly protrudes in front of the grid, and the original performance data The musical features of will be lost. In the present invention, for notes straddling the grid in this way, it is determined in consideration of the sound length which of the discrete points (grids) before and after the note-on point should be mapped. In other words, when the note length spans the grid, compare the note length from the note-on to the first grid with the remaining length after that, and if the note-on side is long, The note-on point is mapped to the grid in front of it, but otherwise it operates to map to the grid immediately after it. Therefore, according to the present invention, by considering the note length of the note information in the original performance data, it is possible to perform the quantization process utilizing the musical characteristics of the original performance data, which is practical in music production such as MIDI. Quantize function can be obtained.

  In the performance data processing apparatus according to the present invention, the quantize strength (ST) is set by the user (S1), and when the quantize is executed (S10 to S14), the set quantize strength (for example, ST = 50%). The amount of quantization can be controlled according to (Claim 2). That is, the note-on timing (tnc; tnd) of the note information (for example, Nc; Nd) is the time (tnc to tgn; tnd to tgn) until the reference timing (tgn-1; tgn) selected as the target time position. On the other hand, the temporal position can be changed (tnc → tnc ″; tnd → tnd ″) by the ratio indicated by the quantize strength (ST = 50%) set by the strength setting means (S1). For example, when the quantization intensity (ST) setting is 100%, the note-on timing is changed to the reference timing grid, but when it is 0%, the timing is not changed, and the setting intensity (ST) is intermediate between these. In the case of a value, the note-on timing is changed to an internal dividing point of 100% and 0%. Therefore, according to the present invention, it is possible to control the strength of the quantize according to the setting when performing the quantize utilizing the musical characteristics of the original performance data in consideration of the note length of the note information. In production, a more practical quantization function can be obtained.

[System Overview]
The electronic music apparatus incorporating the function of the performance data processing system according to one embodiment of the present invention is a kind of computer having a music information processing function including performance data processing, typically like a sequencer or an electronic musical instrument. A device dedicated to music information processing is used, but a music data processing device in which a music information processing function is added to a general-purpose information processing device such as a personal computer (PC) can also be used. FIG. 1 is a block diagram showing the hardware configuration of an electronic music apparatus according to an embodiment of the present invention.

  The electronic music apparatus in which the performance data processing system is constructed includes a central processing unit (CPU) 1, a random access memory (RAM) 2, a read only memory (ROM) 3, an external storage device 4, a performance operation detection circuit 5, a panel ( Setting) An operation detection circuit 6, a display circuit 7, a sound source circuit 8, an effect circuit 9, a MIDI interface (I / F) 10, a communication interface (I / F) 11 and the like are provided. 12 to each other.

  The CPU 1 constitutes a data processing unit together with the RAM 2 and the ROM 3 and executes various music information processing including performance data processing using a clock by the timer 13 according to a predetermined control program. The RAM 2 is used as a work area for temporarily storing various data necessary for these processes. The ROM 3 stores various control programs, control data, preset performance data, and the like necessary for executing these processes.

  The external storage device 4 includes a compact disk read only memory (CD-ROM), a flexible disk in addition to a built-in storage medium such as a hard disk (HD) and a rewritable nonvolatile semiconductor memory such as a flash memory. Includes various portable external storage media such as (FD), magneto-optical (MO) disk, digital multipurpose disk (DVD), smart media (registered trademark), etc. The produced original performance data can be stored in any storage medium.

  The performance operation detection circuit 5 constitutes a performance operation section together with a performance operation element 14 such as a keyboard, detects the operation of the performance operation element 14, and introduces performance data corresponding to the detected performance operation to the data processing section. Can do. The panel operation detection circuit 6 detects the operation content for a panel operation element (setting operation element) 15 such as a switch, and introduces various corresponding setting information into the data processing unit. The display circuit 7 controls lighting / display contents of various display elements (not shown) and a display (display device) 16 such as an LCD in accordance with commands from the CPU 1 and performs display assistance for various operations. For example, in the performance data processing mode, the unit (Tg) and intensity (ST) of quantization can be set by the user operation of the panel operator with display assistance from the display circuit 7.

  The tone generator circuit 8 constitutes a musical sound signal generation unit together with the effect circuit 9 having a DSP, and the performance data based on the performance operation of the performance operator 14 and the performance data read out from the storage means 3 and 4 and stored in the RAM 2. Based on the above, a musical tone signal in an audio format representing a musical tone waveform is generated. The sound system 17 connected to the effect circuit 9 includes a D / A converter, an amplifier, and a speaker, and generates a musical sound based on a musical sound signal from the effect circuit 9.

  An external MIDI device 50 having a MIDI music information processing function is connected to the MIDII / F10 in the same manner as this electronic music device, and MIDI performance data is exchanged between the electronic music device and the external MIDI device 50 through the MIDII / F10. can do. A communication network 60 such as the Internet or a local area network (LAN) is connected to the communication I / F 11. A control program or performance data is downloaded from an external server computer 70 or the like, stored in the external storage device 4, and stored in this electronic music. Can be used in the device.

[Structure of performance data]
In the performance data processing system according to one embodiment of the present invention, performance data produced by the electronic music device or the external MIDI device 50 or performance data stored in the external storage device 4 is read into the RAM 2 as original performance data. Can be quantized. FIG. 2 shows an example of performance data used as original performance data in the performance data processing system according to one embodiment of the present invention.

  The performance data PD that can be used as a quantization target in this system is, for example, MIDI performance data produced in a normal SMF (Standard MIDI File) format as shown on the left side of FIG. It consists of a header chunk HC describing guidance information for the entire performance data and a plurality (16) of track chunks TC1 to TC16. In each track chunk TC1 to TC16, as shown on the right side of FIG. 2, meta events or MIDI events are sequentially arranged together with a delta time Δ indicating an elapsed time from the immediately preceding event, and the smallest track chunk TC1 = “ The track chunk 1 is called a conductor track and includes meta events St and Ts representing time signatures and tempos, and MIDI events Nn and Nf representing notes in the second to sixteenth track chunks TC2 to TC16. Is included. In each track chunk TC1 to TC16, the value of the delta time Δ for the first event at the track start time is “0”.

  That is, the first track chunk TC1 indicates the initial time signature and tempo by the first time signature event St and the set tempo event Ts (not shown) described before the track data performance start (SMPTE offset) event. By using the time signature event St and the set tempo event Ts described together with the delta time Δ, the time signature and tempo instructions from that time can be changed. Further, the TC2 to TC16 below the second track chunk indicate notes (notes) to be played at that time by the note-on event Nn and the note-off event Nf sequentially arranged with the delta time Δ as the performance progresses.

[Quantization of performance data]
In the performance data processing system according to the embodiment of the present invention, the note information is taken into account when the note information included in the original performance data is quantized in predetermined musical time units to produce new performance data. By doing so, a practical quantization function can be realized. 3 and 4 are diagrams for explaining an example of the quantization process according to one embodiment of the present invention. FIG. 3 (1) shows the note information of the original performance data according to the difference in appearance timing and sound length. The four cases are divided into (a) to (d) and expressed in a piano roll score style. The note information in each case can be obtained from the delta time Δ and the note-on and note-off events in the track data of the second track chunk TC2, using the performance data PD shown in FIG. 2 as the original performance data for quantization.

  Here, “Grid n−1”, “Grid n”,... Are referred to as “Grid”, and a quantizing process reference timing tg: tgn−1 that appears every predetermined musical time unit Tg from the start of performance. , Tgn, tgn + 1,. The time unit Tg is called a quantize unit, and is defined by an interval between adjacent grid appearance timings (for example, tgn−1, tgn). One beat or half beat is used with a quarter note length as one beat. . The value of the time unit Tg is determined from the time and tempo values indicated by the time event St and the set tempo event Ts of the first (conductor) track chunk TC1 of the performance data PD. Therefore, the performance time unit is represented by the same reference symbol “Tg” in FIG. 3A, but the value is not necessarily the same, and the time and tempo values indicated at that time are not the same. Dependent.

  In FIG. 3 (1), note Na shown in the uppermost row represents the case (a) that is located between adjacent grids, and the note length indicating the note duration from note-on timing tna to note-off timing tfa. Ta is shorter than the musical time unit represented by the grid interval Tg between the grids Grid n and Grid n + 1, and does not straddle any grid. The four notes Nb1 to Nb4 shown in the next stage also show a case (b) in which different notes are accommodated between adjacent grids. These notes Nb1 to Nb4 are shifted and continuous, and one note Group Nb is formed. On the other hand, the notes Nc and Nd shown in the third row and the bottom row represent cases (c) and (d) in which the note durations (sound lengths Tc and Td) extend over the grid. The sound lengths Tc, Td indicating the time from the on timings tnc, tnd to the note off timings tfc, tfd are determined by the grid over which the notes are straddled, that is, the grid Grid n immediately after the note on, It can be divided into the rear part Tgf. In this case, the note length front part Tng of each note Nc, Nd is represented by the time (period) tnc to tgn, tnd to tgn from the note-on to the appearance timing tgn of the grid Grid n, and the note length rear part Tgf is The time (period) from the appearance timing tgn of the grid Gridn to the note-off is expressed as tgn to tfc, tgn to tfd.

  Quantization of performance data is performed by discretizing reference points (grids) obtained by discretizing points (note on timing) tn: tna,..., Tnc, tnd of certain note events N: Na,. It is based on the principle that the note event N is moved in time by mapping to the Grid. As a basic mapping operation, the note-on timing tn is mapped to the immediately preceding grid Grid, and the note-on coincides with the grid Grid. The timing tn is not changed. In this system, when the notes fit between the grids, the notes Na and Nb of (a) and (b) follow the basic mapping operation and are mapped to the grid Grid n immediately before the note-on. In the case of straddling n, the notes Nc and Nd of (c) and (d) are mapped to the grids Grid n-1 and Grid n immediately before or after the note-on depending on the magnitude relationship between the front part and the rear part Tng and Tgf. When Tng> Tgf, (c) is mapped to the grid Grid n−1 immediately before the note-on, and when Tng ≦ Tgf (d), it is mapped to the grid Grid n immediately after the note-on.

Further, in this system, when such mapping is performed, the quantization amount ST can be set by the user operation (15) to control the quantization amount. That is, the quantize strength ST is the ratio of the amount of time (quantize [change] amount) that is actually moved to the time from the original note-on timing tn to the appearance timing (reference timing) tg of the grid Grid that is the target position. When the strength ST is set, the note-on timing tn is changed by the ratio indicated by the set strength ST. For example,
Quantize change amount = (original note-on timing−target reference timing) × ST
It is represented by As described above, the note-on timing tn is changed to the target reference timing tg when the intensity ST is 100%, is not changed when the intensity ST is 0%, and when the intensity ST is an intermediate value thereof. The internal reference point is changed between the target reference timing tg at 100% and the original note-on timing tn at 0%.

  FIGS. 3 (2) and 4 are mappings when the quantizing intensity ST is set to 100% and 50% for the cases (a) to (d) in FIG. 3 (1), respectively. The operation is represented in a piano roll score like FIG. In each figure, the broken lines indicate the notes Na to Nd of the original performance data, and the hatched lines indicate the notes Na ′ to Nd ′, Na ″ to Nd ″ after quantizing these notes at 100% and 50% strength. Show.

  When the notebook N fits between the grids, like the notebooks Na and Nb shown in the uppermost stage and the second stage of FIG. 3 (1), (a) and (b) follow the conventional basic quantization procedure. As shown in the uppermost stage and the second stage of FIG. 3B or FIG. 4, the appearance timing (reference timing) of the front grid Grid n of the inter-grid range Grid n to Grid n + 1 in which the notes Na and Nb are accommodated. ) Let tgn be the target position for quantization. When the quantize strength ST = 100%, the notes Na and Nb are moved to the target position tgn as shown in FIG. 3 (2). When the quantize strength ST = 50%, the target position as shown in FIG. The notes Na and Nb are moved by 50% up to tgn.

  On the other hand, when the note N extends over the grid Grid as in the case of the notes Nc and Nd shown in the third level and the lowest level in FIG. 3A, in the cases (c) and (d), the notes Nc and Nd In consideration of the sound lengths Tc and Td, as shown in the third and bottom levels of FIG. 3 (2) or FIG. 4, the sound length front part Tng up to the grid Grid n where the notes straddle is the grid grid. When the sound length rear part Tgf after n is exceeded (c), the appearance timing tgn-1 of the grid Grid n-1 ahead of the grid Grid n is set as the target position for quantization, and the sound length front part Tng is the sound length rear part. When the time is equal to or less than Tgf (d), the appearance timing tgn of the grid Gridn itself over which the note straddles is set as a target position for quantization. Then, when the quantize strength ST = 100%, the notes Nc and Nd are moved to the target positions tgn−1 and tgn as shown in FIG. 3B, and when the quantize strength ST = 50%, as shown in FIG. In addition, the notes Nc and Nd are moved by 50% from the target positions tgn−1 and tgn. In any of the cases (a) to (d), the sound lengths Ta to Td of the original notes Na to Nd are maintained after quantization.

  Specifically, when quantizing with the intensity ST = 100%, as shown in FIG. 3 (2), the note Na in the case (a) maintains the tone length Ta while the original on-timing tna becomes the grid appearance timing ( The reference timing is moved to tgn, and converted to a new note Na ′ having an on timing tna ′ = tgn and an off timing tfa ′ = tna ′ + Ta. For each of the notes Nb1 to Nb4 of the note group Nb in the case (b), a new note group Nb composed of new notes Nb1 ′ to Nb4 ′ in which the respective on timings are moved to the reference timing tgn while maintaining the respective sound lengths. 'Is obtained. In the case Nc of the note Nc, the original on-timing tna moves to the reference timing tgn−1 while maintaining the sound length Tc, and the on-timing tnc ′ = tgn−1 and the off-timing tfc ′ = tnc ′. It is converted into a new note Nc ′ of + Tc. Further, for the note Nd in the case (d), the original on-timing tnd moves to the reference timing tgn while maintaining the sound length Td, and the new on-timing tnd ′ = tgn and off-timing tfd ′ = tnd ′ + Td. Is converted into a note Nd ′.

  Further, when quantizing with intensity ST = 50%, as shown in FIG. 4, the original note Na in the case (a) maintains the tone length Ta, and the original on timing tna is only 50% up to the reference timing tgn. It is converted to a new note Na ″ with an on timing tna ″ = tna− (tna−tgn) / 2 and an off timing tfa ″ = tna ″ + Ta. For each of the notes Nb1 to Nb4 of the note group Nb in the case (b), a new note Nb1 ″ to Nb4 ″ in which each on-timing is moved by 50% up to the reference timing tgn while maintaining each tone length. The note group Nb ″ is obtained. In the case (c), the note Nc maintains the tone length Tc while the original on-timing tna moves by 50% up to the reference timing tgn−1, and the on-timing tnc “= Tnc− {tnc− (tgn−1)} / 2, off timing tfc ′ = tnc” + new note Nc ”of Tc”. Further, for the note Nd in the case (d), the original on-timing tnd moves by 50% from the reference timing tgn−1 while maintaining the sound length Td, and the on-timing tnd ″ = tnd + {(tgn−1 ) -Tnd} / 2, converted to a new note Nd ′ of off timing tfd ″ = tnd ″ + Td.

  For example, when using the result of quantization processing to analyze how many notes exist in which quantization unit Tg, as in the notes Na and Nb in (a) and (b), When a note in a certain quantize unit Tg is quantized so as to be attributed to the quantize unit Tg, a note that slightly protrudes forward from a certain quantize unit Tg as in the note Nd in (d). However, it is more musical if it is judged that the note belongs to the quantizing unit Tg on the back side, ignoring the protrusion, and the musicality of the original performance data is often not impaired. If it does, as shown in FIG. 3 (1) and FIG. 4, the result according to such a case will be obtained.

  As described above, in this performance data processing system, note information N: Na,..., Nc, which defines note-on timing tn: tna,..., Tnc, tnd and note-off timing tf: tfa,. When the original performance data PD including Nd is quantized in a predetermined musical time unit Tg such as one beat or half beat, the period (note length Ta˜) of the performance data PD from the note on timing tn to the note off timing tf. When the note information N: Nc, Nd is detected over the reference timings tg:..., Tgn−1, tgn,... Td) appearing in a predetermined musical time unit Tg, such note information Nc, Nd is detected. Is from note-on tn: tnc, tnd to the next reference timing (appearance timing of grid Grid) tgn Reference timing tgn−1 immediately before or after note-on, depending on the comparison result between the sound length front (first time) Tng and the sound length rear (second time) Tgf from the reference timing tgn to note-off tf, tgn is selected, and the note-on timing is changed with the selected reference timing tg as the target time position for quantization. That is, when Tng> Tgf, the target is set to the immediately preceding reference timing tgn−1 to change the ON time (tnc → tnc ′, tnc ″), and when Tng ≦ Tgf, the target is set to the immediately following reference timing. The on-time is changed at the timing tgn (tnd → tnd ′, tnd ″).

[Example of processing flow]
5 and 6 are flowcharts showing an example of performance data processing according to one embodiment of the present invention. The performance data processing is started by instructing the performance data processing mode as the operation mode of the electronic music apparatus by a user operation in the standby state of various processing modes in which the mode selection screen (initial screen) is displayed on the display 16. To do.

  In the first step S1, the CPU 1 loads desired performance data PD from the storage means 4 onto the RAM 2 in accordance with a user operation, selects a track chunk (eg, TC2) to be quantized, and responds to this. Various parameters necessary for performance data processing (quantization processing) such as the quantization unit Tg and the quantization strength ST are set. In the next step, the tempo information Ts and the time signature information St for generating the grid are read from the first track chunk TC1 of the original performance data PD, and the grid information Grid representing the reference timing tg appearing in the set quantization unit Tg. : ..., Grid n-1, Grid n, Grid n + 1, ... are generated and stored in the grid information storage area of the RAM 2, and the note information to be processed first is designated from the track chunk to be quantized. Proceed to step S3.

  In step S3, the note-on event Nn of the instructed note information is acquired. In the next step S5, the note-off event Nf corresponding to the note-on event is acquired, and the timing information tn and tf of these events is obtained. In addition, it is held in the work area of the RAM 2. At this time, in step S5, the note-on event that appears temporally after the note-on event acquired in step S3 and that is the earliest note-off event of the same pitch as that of the note-on event (same note-off event immediately after the same pitch) ) To get. In this way, when the information of note-on and note-off events Nn and Nf is acquired for one piece of note information to be processed, the process proceeds to step S5.

  In step S5, whether or not the period (sound length) tn to tf of the note-on event Nn and the note-off event Nf of the processing target note information acquired in steps S3 and S4 crosses the grid Grid of the quantization unit Tg indicated by the grid information. Determine whether. Here, for example, as shown in (a) and (b) of FIG. 3A, if it is determined that the period from the note-on event Nn to the note-off event Nf does not cross the grid Grid (S5 = NO) In step S6, the quantize target position is set in the grid Grid immediately before the note-on event Nn.

  On the other hand, when it is determined that the period from the note-on event Nn to the note-off event Nf straddles the grid Grid as shown in (c) and (d) in FIG. 3 (1) (S5 = YES), In step S7, a first time (front part of sound length) Tng from the note-on tn to the immediately following grid Grid is obtained. Here, when a plurality of grids Grid n and Grid n + 1 are straddled as shown in (d), the grid Grid n that is the earliest in time is set as a calculation target of the first time Tng. Next, in step S8, a second time (back part of the sound length) Tgf from the grid Grid to the note-off tf is obtained, and in step S9, the note-off from the grid is performed more than the first time Tng from the note-on to the grid. It is determined whether or not the second time Tgf is long.

  Here, as in the case (c), when the second time Tgf is longer than the first time Tng (S9 = YES), in step S10, the quantized target position of the note-on tn is set to the immediately following grid position (for example, Grid Set to n). On the other hand, as in the case (d), when the second time Tgf is less than the first time Tng (S9 = NO), in step S11, the quantized target position of the note-on tn is set to the grid position one before (for example, Grid n-1).

  After the target position is set in step S6, step S10 or step S11, the process proceeds to steps S12 to S14 in order, and note information change processing is performed. First, in step S12, a new note-on position (tn ′, tn ″) is determined according to the quantize strength ST set in step S1, and in the next step S13, the current note-on position tn is determined. In step 14, the note-off timing tf is changed by the same amount as the note-on movement amount (the tone length, that is, the sound generation period is maintained). When the change is made, the process proceeds to step S15, and when it is determined in step S5 that the grid grid is not straddled (S5 = NO), if the note on timing tn coincides with the reference timing tg at which the grid grid appears, the corresponding note Immediately proceed to step S15 without changing the on and note-off timings tn and tf.

  In step S15, it is determined whether or not the quantizing process of steps S3 to S14 has been performed for all the note information. If all the note information has not been processed (S15 = NO), the note information to be processed next is determined. Is returned to step S3, the processing of steps S3 to S14 is performed for the next note information, and the processing of steps S3 to S15 is repeated until the processing of all the note information is completed. When the quantization process for all the note information is completed (S15 = YES), the process proceeds to step S16, and new performance information obtained by the quantization process is used. For example, a storage process for storing new performance information in the storage device 4, an automatic performance using the new performance information for automatic performance, or an analysis process for analyzing the new performance information and using it for arrangement or the like is performed. After the performance information use processing in step S16, the performance data processing is terminated and the original standby state is restored.

[Various Embodiments]
The preferred embodiment of the present invention has been described in detail with reference to the drawings. However, this is merely an example, and various modifications can be made without departing from the spirit of the present invention. For example, in the embodiment, the unit of the grid is exemplified by beats and half beats, but is a musical time unit generally used in quantizing, and is a measure, a quarter note (beat), an eighth note (half beat). ), Sixteenth notes (quarter beat), and so on. In addition, setting with absolute time, such as 1 second, 0.5 second, ..., is also considered.

  In the embodiment, the performance data defining the note-on and note-off timing is shown using the delta time Δ, but not only the note-on / off format but also the note-on and note-time format. Performance data that defines the off timing may be used.

In the embodiment, (1) note-on (tn) is converted and note-off (tf) is adjusted so that the sound length does not change, but the following cases (2) and (3) are further described. In addition, if it is possible to set whether to convert each of note-on (tn) and note-off (tf) in a combination of these cases, the scene where the performance data after quantization is used can be expanded:
(2) When note-on (tn) is converted but note-off (tf) is not changed.
(3) When note-on (tn) is converted and note-off (tf) is also converted independently.

  In the embodiment, the determination of the grid selection when straddling the grid is the first time from the grid (tg) to the note-off (tf) than the first time (Tng) from the note-on (tn) to the grid (tg). Is 2 hours (Tgf) long? (S9) is it larger than twice? You may weight as follows.

  In the embodiment, the setting range of the quantize strength ST is set to 0 to 100%, and the note information is moved to a point that is internally divided to the target position. A special effect in editing performance data may be obtained by moving the note information to the outer dividing point.

1 is a system configuration explanatory diagram of an electronic music apparatus according to an embodiment of the present invention. FIG. It is a figure showing the example of the performance data used for the quantization process in the performance data processing system by one Example of this invention. FIG. 3 is a part of a diagram showing an example of quantization according to an embodiment of the present invention. It is the other part of the figure showing the example of quantization according to one Example of this invention. It is a part of flowchart showing the example of performance data processing by one Example of this invention. It is the other part of the flowchart showing the example of performance data processing by one Example of this invention.

Explanation of symbols

Tg Quantize unit (musical time unit),
Grid: ..., Grid n-1, Grid n, ... Grid,
tg: ..., tgn-1, tgn, ... Grid appearance timing (reference timing),
N: Na to Nd Note information or note (note)
Ta to Td Notes (notes) Na to Nd,
tn: tna, tnc, tnd original note-on timing,
tf: tfa, tfc, tfd original note-off timing,
Tng, Tgf Sound length front (first time) and sound length rear (second time)
ST Quantize strength (quantize strength),
tn ′: tna ′, tnc ′, tnd ′, note-on timing after 100% intensity processing,
tf ′: tfa ′, tfc ′, tfd ′ note-off timing after 100% intensity processing,
tn ″: tna ″, tnc ″, tnd ″, note-on timing after 50% intensity processing,
tf ″: tfa ″, tfc ″, tfd ″ Note-off timing after 50% strength processing.

Claims (3)

Performance data acquisition means for acquiring performance data including note information defining note-on timing and note-off timing;
About the performance data acquired by the performance data acquisition means, note detection means for detecting note information that crosses a reference timing in which a period from a note-on timing to a note-off timing appears in a predetermined musical time unit;
Time calculation means for calculating a first time from a note-on timing to a reference timing immediately after the note-on timing and a second time from the reference timing to the note-off timing for the note information detected by the note detection means;
Time comparison means for comparing the first time and the second time calculated by the time calculation means;
According to the comparison result by the time comparison means, the reference timing immediately before or immediately after the note-on timing of the note information detected by the note detection means is selected, and the note-on timing is selected with the selected reference timing as the target time position for quantization. A performance data processing apparatus comprising: a quantize executing means for changing the sound. Furthermore, it comprises strength setting means for setting the quantize strength,
The quantize executing means is a ratio of the note on timing of the note information detected by the note detecting means corresponding to the quantize intensity set by the intensity setting means with respect to the time to the reference timing selected as the target time position. The performance data processing apparatus according to claim 1, further comprising intensity control means for changing only the intensity control means. A performance data acquisition step for acquiring performance data including note information defining note-on timing and note-off timing;
For the performance data acquired in the performance data acquisition step, a note detection step for detecting note information that exceeds a reference timing in which a period from a note-on timing to a note-off timing appears in a predetermined musical time unit;
A time calculating step for calculating a first time from a note-on timing to a reference timing immediately after the note-on timing and a second time from the reference timing to the note-off timing for the note information detected in the note detection step;
A time comparison step for comparing the first time and the second time calculated in the time calculation step;
Depending on the comparison result in the time comparison step, the reference timing immediately before or after the note-on timing of the note information detected in the note detection step is selected, and the selected reference timing is set as the target time position for quantization. A performance data processing program for causing a computer to execute a procedure comprising a quantize execution step for changing timing.

Images