JPS61290495A - Automatic performer - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an improvement in an automatic performance device that can automatically perform a desired piece of music with rhythmic accompaniment.

[Summary of the invention]

This invention stores rhythm and V-turn selection information together with music information in a memory or the like, and selects a rhythm pattern '12r:a based on the rhythm pattern selection information when continuously outputting music information and performing automatic performance. By performing rhythm performance, it is possible to create a rhythmic accompaniment that is suitable for the performance piece and is rich in variation. Autorhythm device synchronized with the start of automatic performance? An automatic performance device is known that performs rhythm accompaniment by starting the device (for example, Japanese Patent Laid-Open No. 58-T3
(See Publication No. 0385).

[Problem that the invention seeks to solve]

With the above-mentioned conventional automatic performance devices, the rhythm to be played must be manually selected.
It is troublesome to select the appropriate rhythm for each song.
Moreover, since the autorhythm device only stores general rhythm patterns for each rhythm type such as march, waltz, swing, etc., the range of rhythm selection is limited. , the rhythm performance tended to become monotonous.

In order to vary the rhythm performance, it would be possible to memorize a large number of rhythm patterns, including each rhythm block UmK fill-in pattern, ending V-turn, etc., and select them as appropriate. It is inevitable that operations will become even more complicated.

[Means for solving problem χ]

This invention was made in order to solve the above-mentioned problems, and its purpose is to make it possible to create a rhythmic accompaniment that is suitable for a piece of music and is rich in variation, without any troublesome operations. It is something.

In the automatic performance device according to the present invention, a first storage section stores music information corresponding to the note progression of a desired song together with rhythm pattern selection information;

A second storage section storing a plurality of rhythm patterns is provided. From the first storage unit, music information and rhythm pattern selection information are read by the first successive output means, and based on the read music information, the musical sound generation means generates a musical sound signal corresponding to the note progression of the music. be done. Further, from the second storage section, one of the rhythm patterns is selected and read out by the second successive output means according to the rhythm pattern selection information, and the rhythm sound signal is output from the rhythm sound generation means according to the successive rhythm patterns. generated.

[Effect]

According to the configuration of the present invention, one of the rhythm patterns Ym is selected and read out from the second storage section based on the rhythm pattern selection information read out from the first storage section. Manual operation is not required. Further, by storing arbitrary rhythm pattern selection information in relation to the note progression of a song in advance in the first storage unit, it is possible to automatically control rhythm/turn pattern switching in the process of automatic performance of the song. This allows for a variety of rhythmic accompaniments.

〔Example〕

FIG. 1 shows an embodiment in which this invention is applied to a Kaoru Seiko musical instrument.The electronic musical instrument of this embodiment is capable of manual performance sound generation, automatic musical performance, automatic rhythm accompaniment, etc. with the help of a microcomputer. It's about to be controlled.

Circuit configuration (Fig. 1) - Each bus 10 includes a keyboard circuit 12 and a control switch circuit 14.
, central processing unit (CPU) 16, program memory 18
, working memory addition, reading device η, performance data memory tone, tempo excavator (O8C) 26, music tone generator (TG) 28, and rhythm tone generator (
TG) 30 is connected.

The keyboard circuit 12 is equipped with a manual performance keyboard having a large number of keys, and key operation information is detected by scanning a key switch provided for each key.

The control switch circuit 14 includes control switches for tone, volume, effects, etc., a performance data reading switch, and an automatic performance start/control switch.
It includes stop switches and the like, and switch operation information is detected by scanning these switches.

It executes various processes such as manual performance sound generation, automatic musical performance, automatic rhythm accompaniment, etc. according to the rhythm, and these processes will be described later with reference to FIGS. 7 to 12.

The working memory is comprised of RAM (Random Accessible Memory) and includes storage areas used as registers, counters, flags, etc. during processing by the CPU 16. Note that registers related to automatic performance of music and automatic rhythm accompaniment will be described later.

The reading device @, n is for reading performance data from an external storage medium 32 consisting of, for example, a ROM bank, and when the above-mentioned performance data reading switch is turned on with the external storage medium 32 inserted into the receiving slot, the reading operation starts. It is supposed to start.

The performance data memo IJ 24 consists of RAM.
This is for storing performance data read by the reading device η.

Tempo 08C26 generates a tempo clock pulse. If interrupts are enabled in the main routine of FIG. 7, the interrupt routine of FIG. 8 will be executed every time tempo 08C26 generates a tempo clock pulse.
As a result, automatic performance and automatic rhythm accompaniment of the music piece are performed. The tempo of automatic performance of the music and rhythm accompaniment is determined according to the repetition frequency of the tempo clock pulse generated from the ten yti'08c26.

The music TG 28 is a tone generator used for manual performance sound generation or automatic performance of music, and -
As an example, there are 16 time-sharing musical tone forming channels. Therefore, up to 16 sounds can be simultaneously produced as manual performance sounds or automatic performance sounds.

The rhythm TG 30 is a tone generator used for automatic rhythm accompaniment, and has, for example, eight time-sharing percussion sound formation channels Z.

Therefore, up to eight rhythm sounds can be produced simultaneously.

The sound system type includes an output amplifier, speakers, etc., and a musical tone signal from TG28 for music and a T for rhythm.
The rhythm sound signal from the G30 is converted into sound.

Data Format (FIGS. 2 to 6) FIG. 2 shows the format of performance data in the external storage medium 32.

The performance data is a header HD that shows the song number, song title, etc.7, and a C
Each channel from HI to CH16 includes control data CD for controlling pitch, volume, rhythm selection, etc., and rhythm data RD for representing rhythm patterns, etc. These data are read in the order described and stored in the performance data memory U. Then, automatic performance and automatic rhythm accompaniment of the music piece are performed based on the performance data stored in this way.

As shown in detail in Figure 3, the header HD includes data A of 2.might representing the song number, data B of 32.might representing the song title with 1 byte per character, and CH1 to CH1.
The first order of data for 16 channels of 6 is 2 per group.
It includes 16 sets of data C each expressed in bytes, 2-byte data D indicating the starting address of the control data CD, and 2-byte data E indicating the starting address of the rhythm data RD.

The music data MD consists of 4-byte data EVM for one event arranged in sequence, as shown in Fig. 4 for any one channel of CHI to CH16. R8 is placed as appropriate, and end code data MED is placed at the end of the array. l
The 4-might data EVM for I-Kento includes l-might key code data KC representing the pitch of the note to be sounded, l-byte key-on timing data KON representing the sound generation timing, and mute timing χ. Key-off timing data KOF of 'fl-ite' and 1-byte volume data VO representing the volume rail of the sound to be sounded.
The rest data R8 consisting of L is inserted in order to realize a long rest period in units of measures.
It consists of rest code data RC of 1.zeta. and bar number data BN of 1 byte representing the number of bars 7.

Timing values for timing data KON and KOF are divided into % if one measure is 4 beats and expressed as 0 to 95, if it is 3 beats it is divided into 72 and expressed as 0 to 71, and if it is 2 beats it is expressed as 0 to 71. For example, it is divided into 48 parts and expressed as any number from 0 to 47. Note that for rest periods shorter than one measure,
This can be achieved by providing a gap corresponding to a desired rest period between the key-off timing value of one event and the key-on timing value of the next event.

As shown in detail in FIG. 5, the control data CD is 3 bytes of data (gvcy) for 1 inte arranged in sequence, and end code data CED is placed at the end of the arrangement. 3 nodes data EVCV for 1 event
In the same way as described above for i, KON and KOF, control timing data CT of 1 d item representing the control timing and control timing data CT of 1 d item representing control types such as timbre, volume, effect, rhythm on/off, rhythm selection, etc. It consists of type data C8 and control value data dv representing the desired control state or control tyr.

As 3/5-ite data EVC regarding rhythm control,
Those related to rhythm on/off control and those related to rhythm selection are used. The 3rd and 5th data EVC regarding rhythm on/off control is control timing data C.
T represents timing 7 when the rhythm should be turned on or off, control type data C8 represents rhythm on/off control, and control value data Cv represents rhythm on (Ill) or rhythm off (rOJ) Y. Furthermore, the 3/5-item data EvC related to the rhythm selection pressure indicates the timing of rhythm selection using the control timing data CT, the rhythm selection using the control type data C8, and the timing of the rhythm data to be played using the control value data Cv. Represents a number (ie, rhythm pattern number).

As shown in detail in FIG. 6, the rhythm data RD includes 1-byte data PN representing the number N of rhythm patterns, attribute data PS representing the attributes (time signature, etc.) for each rhythm pattern, and N different rhythm data PN. First to Nth rhythm pattern data PT1 to PTN representing patterns χ, respectively
Contains.

The attribute data PS consists of NI bit data from Sl to SN, and each 1-byte data has a time signature of 81 in the upper 2 bits.
7, and the lower 3 bits represent the instrument group number ``IGN''.The 2 bits representing the time signature BT are ``If it is a 4 time signature,''
00", "Ol" for triple time, l tO for double time.
You can take a set of each in J.

In this example, eight rhythm patterns'? : Can be stored, and the maximum value of N is 8. In addition, these eight rhythm patterns (i.e. rhythm pattern numbers 1 to 8)
) corresponding to 0 to 7 musical instruments Glucinan-s
r G N is determined. Each instrument group Nan J
IGN is used to allocate a group of percussion sounds necessary for playing the corresponding rhythm pattern to the eight percussion sound forming channels of the rhythm TG30,
For example, the sounds assigned to each channel for IGN=0 and IGN=7 are as follows.

Channel IGN=0 1GN,=71
Cymbal■ Caravelle I2 Bybutt Bybutt 3 Brush Shot High Conga 4 Brush Roll
Low Conga 5 Cymbal ■ Shinnoru 6 Snare Drum I Snare Drum 7
・Strum Pass Drum 8
Snare Drum ■ Cowl ■ Each rhythm pattern data is representatively shown in Fig. 6 for PTI. Here, the sound control data for one measure is 4.
If it is a meter, it is 48 sets (96・tito), if it is a triple meter, it is 36.
set (72··ito), and if it is in double time, it is 24M (48··ito).

Each sound generation control data 5DFi, for each sound generation timing, the sound or non-sound of percussion sound forming channels 1 to 8 is expressed as @l'' or '''0'', and accent ACC.
(sound t) is expressed in one of four stages from "OO" to "11". In this case, each sound generation control data S
Regarding the 1st to 4th bits of D, 2 bits are assigned to each channel of 1 to 4, but this means that even if the hitting sound of the same channel (for example, high note) is [lO
] This is to make different tones be produced when the sound is "11" and when the sound is "11".

Rhythm pattern No. 8 ~ 8 is instrument group No. 1
0 to 7, respectively, and the rhythm pattern data PT1-p'rB, so the above-mentioned 3-byte data EVC can be used to write any rhythm pattern number.
Rhythm accompaniment can be performed with any rhythm pattern by specifying the following. For example, the rhythm pattern data p
The rhythm accompaniment used for 'r1' becomes possible. In this case, the percussion sound forming channels 1 to 8 are assigned the percussion sound of 8 toku corresponding to the musical instrument Grunonan -5oVc as described above, so for example, channel 3 has a brush shot sound, and channel 7 has a q strum sound. Working memory 20 (FIG. 1) formed Among the registers added to the working memory, those related to automatic performance of music and automatic rhythm accompaniment are as follows.

(1) Run flag RUN This flag stores "1" or V1'"01' based on the operation of the automatic performance start/stop switch mentioned above. If it is 1", it means that the performance is in progress; 0
” indicates that the playback is stopped 7.

(2) Rhythm run flag RHYRUN This flag is associated with automatic rhythm accompaniment and stores 61' or addition. 11" indicates that accompaniment is in progress, and 10'
If, it indicates that the accompaniment has stopped.

(3) Tempo counter CLK This counter increments by 1 each time the interrupt routine shown in Figure 8 is executed during automatic performance.
It is reset at the timing when the count value χ takes 0 and becomes %, and if it is a triple beat, it is reset when the count value χ is taken from O~71 and becomes 72, and when it is a double beat, it is reset when it becomes 72.
It is reset when the count value reaches 48.

f4)! JIA/V turn number register
RHYNO This is for storing the 37th item of the 3-byte data EVC regarding rhythm selection, that is, the rhythm pattern number and sound data.

(5) Beat register BEAT This was selected from the attribute data PS mentioned above! The data of the upper two bits of the byte data (any of Ss to SN), that is, the beat BT is stored.

(6) Control data successive pointer PNT is an address pointer used when reading the control data CD described above.

(7) Rhythm data read pointer PNTR This is
This is an address pointer used when reading the rhythm data RD mentioned above.

(8) Music data successive pointer PNT1 to PNT1
Their pointers PNT1 to PNT is are the second
This is an address pointer for each channel used when successively outputting music data for each channel of CH1 to CH3 in the figure.

(9) Key-on register KEYON This is CH1
This is a 2-ite register having bit positions fullY for 16 channels of CH15, and 'l' is stored in the bit position corresponding to the channel in which sound is being generated.

00) End register END This is a 2-signature register that has bit positions for 16 channels of CH1 to C) (16), and end code data MID has been detected (data has ended).
11" is stored in the bit position corresponding to the channel.

(l]) Control end register END stores I'1'' when end code data CED is detected for control data CD (data ends).

(12) Instrument group number 5 register IGNRE
G This is 17 selected from the attribute data PS mentioned above.
The lower 3 bits of data of 1 byte (any of 81 to SN), i.e. instrument group number 5IG
N is the store name.

(13) Rest registers REST1 to REST1 are these registers RE S Tl to REST16 are CH
This register corresponds to each of the 16 channels from CH1 to CH16, and is provided to realize a long rest period in bar units for each channel. REST,
~Each register of RE8T16 has a 2-byte configuration, and when rest data R8g is stored for a certain channel as shown in Figure 3, the l-qth ite of the rest register corresponding to that channel is stored. Then, the measure number data BN in the rest data R8 is stored in the 2nd and 5th bytes of the same register, and the key-off timing data KOF in the 3rd and 5th byte data EVM immediately before the rest data R8 are stored, respectively.

(14) Maximum timing value register MAX This is
This is to store the maximum timing + i'v for each beat, "%" for 4 beats, "72" for 3 beats.
, if it is a double time signature, "48" is stored respectively.

(15) Load flag LOAD When performance data is loaded from the external storage medium 32 to the internal memo IJ 24, @ and H are sent to this flag.

(16) Sound generation control data register RHYDAY This is a 2-byte register for storing the sound generation control data 5D7i shown in FIG.

Main Routine (FIG. 7) Next, the processing flow of the main routine will be explained with reference to FIG.

First, when the power switch (not shown) is turned on, seven initial setting processes are performed in step 40, and the run flag RU
N and the load flag LOADχ are both set to 1110N.

Next, in step 42, key operation information is fetched from the keyboard circuit 12, and if there is a pressed key, music T028'ft is controlled in accordance with the corresponding pitch data to generate manual performance sound.

Next, in step 44, the interrupt is stopped, and then in step 4
6, it is determined whether the performance data reading switch is on.

If this switch is on (Y), step 48
Move to.

In step 48, the performance data is read from the external storage medium 32 by the reading device ρ and loaded into the memory. In this case, the sound generation control data that represents the rhythm pattern of one measure is 48・sheets (in the case of 2 beats) or 72・4 sheets (in the case of 3 beats).
(in the case of beat), data of any number of bytes is converted to 96 bytes and loaded separately into memory. After this, in step (capital), LOAD) is set to 1"l", and then the process moves to step 52. Note that as a result of the determination in step 46,
If not switched on (N), steps 48 and 5
The process moves to step 52 without passing through 0.

In step 52, automatic performance start/stop switch processing 7 is performed. In this process, each time the automatic performance start/stop switch is pressed, the RUNK “1” and “o”
v are set alternately.

Next, in the step diagram, it is determined whether RUN has changed from 0'' to 0!''. As a result of this judgment, if there is a change from "()" to 1' (Y), move to the step diagram and LOAD
Determine if is H, tt. If the performance data loading process has been completed as described above, LOAD is 1'', so the determination result in step 560 is affirmative (Y), and the process moves to step edict.

In the step simple, initial setting processing for automatic performance is performed, and the rhythm run flag RHYRUNJ mark register RES is set.
T1-REST 15, key-on register KEYON,
End register FJND, tlj rage 1 end register E
Clear both NDC'Y to rOJ.

Next, the process moves to step (a) to process the start address set. That is, since headers are loaded in the performance data memo I724 in the format shown in FIG. 3, the header addresses of the data CHI to CH16 are set to the music data successive pointers PNT1 to PNT16, and the header is loaded to the header in the format shown in FIG. street address? Control data read pointer PNT
At C, the first address of the rhythm data RD is transferred to the rhythm data read pointer PNTH and stored.

After this, in step 62, the previously prohibited interruption request is permitted. In addition, in determining the step diagram, from 0'' to '1
If the result is N) without any change to ``, then step I ~
6o? : Skip to step 62.

Further, in the determination of the step diagram, if LOAD is not "]" (N), the automatic performance cannot be started because the performance data loading process has not been completed. Therefore, the RUN flag is set to "0" according to the step diagram, and then the process moves to step 62.

After step 62, the process returns to step 42 and the above-described process is repeated.

Interrupt Routine (FIG. 8) Next, the processing flow of the interrupt routine will be explained with reference to FIG. This interrupt routine is executed every time a tempo clock pulse is generated from tempo 08C26 when interrupts are permitted in the main routine described above.

First, in step 70, it is determined whether RUN is 1" (or during automatic performance). If the result of this determination is "l" (Y), the process moves to control data processing in step 72. This process The data memo +724 is used to output control data CDs to control tone, volume, effects, rhythm on/off, and rhythm selection ring Y!-1IIJ.
The figure will be described later.

Next, the process moves to step 74, and music data processing is executed. This process is performed from performance data memo IJ24 to CHl-
This is for reading music data for each channel of CH16 and automatically playing the music, and will be described later with reference to FIGS. 10 and 11.

Next, the process moves to step 76, and rhythm generation processing is executed. This process is for performing automatic rhythm accompaniment by successively extracting rhythm pattern data selected according to the rhythm pattern number from the performance data memory 24, and will be described in detail later with reference to FIG. 12.

After this, in step 78, the control end register END
Check if C is 11" (end of data) and not 'l" (
If N), move on to step (capital).

In step (capital), the count value 'a'l of the tempo counter CLK is increased. Then, proceed to step 82,
It is determined whether the count value of CLK is smaller than the maximum timing value of register MAX (before the end of one measure). As a result, if it is determined that the value is small (Y), the process returns to the main routine of FIG. 7 (RET). Also, it is not small (
If it is determined (No), the CLK count value is set to 0 by step polishing, and then the process returns to the main routine shown in FIG.

By the way, in the determination at step 78, if ENDC is '
If the result is "1" (Y), it is determined in step 86 whether all bits of the end register END for 16 channels are "1" (whether data is complete for all channels CHI to CH16). As a result of this judgment, all bits “l”
If not (N), move to step leap to continue automatic performance.Also, if all bits are "l" (Y),
In order to stop the automatic performance, step A sets RUN4 to ``0'', and then returns to the main routine shown in FIG.

If an interrupt occurs after RUN becomes ``O'' as in the case of Jin, the determination result at step 70 becomes negative (N), and the process moves to step 90 tic . This step (a)
Then, the CLK count value "fOK" is returned to the main routine shown in Figure 7. Therefore, RUN is "0".
After that, automatic performance stops and CLK also stops advancing.

Control Data Processing (FIG. 9) Next, the control data processing (step 72 in FIG. 8) will be explained in detail with reference to FIG.

First, in step lOO, it is determined whether the end register ENDC of ft1lJ is 1" (data end), and
If not (N), the process moves to step 102. In this step 102, 1flIJ# data read pointer PN
It is determined whether the data (PNTCI) is end code data CED to the address indicated by TC. If the result of this determination is that it is not end code data CED (N), the process moves to step 104. In this case, [PNTCI is end code data CED] This means that 3.item data E
'/C, that is, the control timing data CT.

In step 104, the value of the control timing data CT is directly compared with the value of the tempo counter CLK to determine whether the timings match. As a result of this determination, if the timing does not match (N), the process returns to the routine of FIG. 8, but if the timing matches (Y), step 106
Move to.

In step 106, it is determined whether the control type corresponds to "rhythm on/off J, r+) rhythm selection" or "others" by reading out the control type data CS from the second byte of the third seat data EVC.

If the result of this judgment is 1'Other, step summer is 0.
8, control processing for tone, sound, etc. is executed.

This process is to set and control the tone, volume, effect, etc. of the music TG 28 in accordance with the 3rd byte city 1'' data CV of the J-ite data EVC regarding the tone, volume, effect, etc.

After this, in step 110, the add value of PNTC is incremented by 3 to make it possible to read the 3.q-ite data EvC of the next point. Then, the process returns to step lOO, and the same process as described above is repeated.

By the way, when it is determined in step 106 that the rhythm is "rhythm on/off", the process moves to step 112. In this step 112, the rhythm run flag RH
To store the rhythm on/off signal in YRUN, the control value data Cv in the third byte of the 3-byte data EVC related to rhythm on/off control is successively stored, and if this data indicates rhythm on, RHYRUN["
L'n, if you instructed rhythm off, RHYRUNK"
0” Store each.

After this, step 1ioy is executed in the same manner as described above.

Further, if it is determined in step 106 that it is "rhythm selection", the process moves to step 114. In step 114, the rhythm pattern number is stored in the rhythm pattern number register RHYNO. That is, the control value data Cv of the 3rd turn of the 3rd turn data EVC relating to rhythm selection is read out and stored in the rhythm number (turn number 4RHYNO) indicated by this data.Then, the process moves to step 116.

In step 116, the instrument group number register t
GNREG's musical instrument group number IGNY strike ends.

That is, RHYNO's rhythm pattern number 1 (1
-8) and specifying the address, RHY N O (7) IJ rhythm pattern number d corresponding to the attribute data PS (1-4 bit data (S□ to S
N), the lower three pit instrument group numbers dIGN4 in this 1-'ite data are read out and stored in IGNREG. Then, the process moves to step 118.

In step 118, the musical instrument group number IGN) stored in the IGNRF 2G is sent to the TG 30 for i rhythm. As a result, the eight percussion sound forming channels of the rhythm TG 300 are assigned the percussion sounds to be formed according to the musical instrument group No. 5 GN.

Next, the process moves to step 120, and the beat BTY is stored in the beat register BEAT. i.e. Sten, l'l16
In the same way as in the case of RI (YNO's rhythm A turn number 5), select from the 1-'ite data required attribute data PS, and select the time signature BTy!- of the upper 2 bits of this lI-ite data one after another to B EAT. Then, the process moves to step 122.

In step 122, the contents of BEAT are checked and roOJ
(double time signature), rolJ (3 time signature), "lO" (double time signature)
Determine which of the following. If it is "00", the maximum timing value is stored in step 124, and the register MAXK value 96 is stored. Similarly, if the value is "01", 72 numerical values are stored in MAX in step 126, and if it is "IO", 48 numerical values are stored in MAX in step 128. In this way, step 124.126.1
After completing either step 28, execute step 110z' in the same way as described above, and repeat the above processing every time an interrupt occurs. Eventually, the data at the address specified by PNTC [PNTC]
becomes the end code data CED. Therefore, the determination result in step 1020 is affirmative (Y), and step 1
Move to 30. In this step 130, ENDC is set to "t".
”, and then returns to the routine of FIG. 8. In the subsequent interrupt, the determination result at step 1000 becomes affirmative (Y), so the routine returns to the routine of FIG. 8, and no actual control processing is performed. Next, the 10th song data processing (Figures 10 and 11)
The song data processing (step 74 in FIG. 8) will be explained in detail with reference to FIGS.

In FIG. 10, in step 140, the channel number i Y l is set. Then, moving to step 142,
Execute processing by channel. This processing performs key on/off processing, rest processing, etc. for the channel number l that has been set, and will be described in detail later with reference to FIG. 11.

Next, in step 144, the channel number H' is incremented by 1, and then in step 146, it is determined whether l is greater than 16. As a result of this determination, if the amount is not greater than 16 (N), the process returns to step 142 and the same process is repeated until 1) 16 is reached.

In this way, when the process of FIG. 1J is executed for 16 channels, l becomes 17 in the subsequent process of step 144. Therefore, the determination result in step 1460 is positive (
Y) and returns to the routine of FIG.

1. In the figure, in step 150, the END signal corresponding to the channel number 5 of the end register END is
(i) is ``1'' (is the data of the first channel complete?)
If the value is not '1'' (N), the process moves to step 152.

In step 152, it is determined whether the 1st to 2nd byte (measure number data) of the rest register RE S Ti corresponding to the first channel is 0 (there is no rest in units of measures), and if it is 0 (Y), , the process moves to step 154.

In step 154, bit KEYON(i) corresponding to the first channel of key-on register KEYON is set to "l".
If it is not ``1'' (N), proceed to step 156. In step 156, the data [P
NTl) is the key code data KC.

It is determined which of the rest code data RC and end code data MED corresponds to. If the result of this determination is that the key code data is KC, the process moves to step 158.

In step 158, the value of the next key-on timing data KON of the key code data KC and the tempo counter CL are determined.
The count value of K and Z are compared to determine whether the timing matches. As a result of this determination, if the timing does not match (N), the process returns to the routine of FIG. 10, but if the timing matches (Y), the process moves to step 160.

In step 160, the key code data KC and the fourth byte volume rail data VOLD are supplied to the music TG 28 to perform key-on processing. In other words, TG28 for music
In the first musical tone forming channel, a musical tone signal whose pitch and volume are respectively controlled according to the key code data KC and the volume rail data VOL is formed and supplied to the sound system 34.

As a result, the speaker of the sound system 34 produces a musical tone corresponding to the musical tone signal.

After this, in step 162, the signal corresponding to the first channel of KEYON(1)'ji<"1"
After that, the routine returns to the routine shown in FIG.

If an interrupt occurs after generating musical tones as described above, the determination result in step 1540 becomes affirmative (Y).
Proceed to step 164.

In step 164, key-on timing data KON
The value of the next key-off timing data KOF is compared with the count value of the tempo counter CLK to determine whether the timings match. As a result of this judgment, the timing does not match (
If the timing matches (Y), the process returns to step 166.

In step 166, key-off processing for the music TG 28 is performed. That is, control is performed to attenuate the tone signal being formed in the fifth tone forming channel.

Next, in step 168, the bit KEYON(i)'a''″0′ corresponding to the i-th channel of KEYON is
Make it. Further, the address value of PNTi is increased by 4 to enable reading of the 4-byte data EVM''f of the next input.

Thereafter, in step 170, key-off timing data KOFy2, which was previously subjected to key-off processing, is stored in the second byte of RESTi. This is a necessary process to realize a long rest period in units of measures, as will be described later.

After performing the VC and reading out the four-item data EVM'g as described above, musical tones can be generated one after another by executing key-on/off processing.

4 d-byte data EVMT (If rest data R8'4 is inserted in the middle of sequential arrangement, 4-byte data E
As VMY is read out one after another, the data at the address specified by PNTl (PNT7) eventually becomes the rest code data R.
It is determined in step f156 that it is C, and step 172
Move to. In this step 172, the next bar number data BN of the rest code data RC is stored in the first byte of RESTi.
'4 stores, and then returns to the routine of FIG.

In this next interrupt process, the publication result of step 1520 is negative (N), and the process moves to step 174.

In this step 174, the first shift of the key-off timing data KOF of the second seat of RESTi is compared with the value of the tempo counter CLK to determine whether the timings match.

In this case, the key-off timing data KOF in the 4-byte data EVM of the input immediately before the rest data R8 is stored in the 2nd and 9th ites of RESTi in the same manner as described above regarding step 170. Therefore, timing coincidence cannot be obtained in the measure to which this data KOF belongs.

That is, the publication result of step 1740 is negative (N), and the process returns to the routine of FIG. Then, after repeating this process several times and reaching the next measure, the publication result of step 1740 becomes affirmative (Y) at a timing corresponding to approximately one measure from the key-off timing, and the process moves on to step 176. , ty, - is subtracted from the bar number data BN of RESTI. And the first
Return to the routine shown in Figure 0. In this case, data BN
If the number of measures indicated by is 1, RFJSTi can be calculated by one subtraction.
However, if the number of measures indicated by the data BN is 2 or more, R
The lI of ESTi becomes O. In this way, RES
By executing step 1764 until the 1st to 9th bits of Ti become 0, a long rest period in units of measures can be obtained.

When the 1st byte of RESTi becomes 0 by subtraction, the result of step 1520 becomes affirmative (Y) in the next interrupt process, and the next key-on process becomes possible.

As the music data for the first channel is read out one after another as described above, the data (PNTI) at the address specified by PNTi eventually becomes the end code data MED.
If so, it is determined in step 156 that the process moves to step 178. In this step 178, the bits END (i)'! corresponding to the first channel of END! Set it to “1”,
Thereafter, the process returns to the routine shown in FIG. 1θ. In the subsequent interrupt processing, the publication result in step 1500 is positive (
Y), the routine returns to the routine of FIG. 10, and no substantial processing regarding the first channel is performed thereafter.

According to the processing shown in FIGS. 10 and 1J described above, the 16 channels Y? of the music TG 28 are determined based on the music data MDK stored in the performance data memories. 1f11H, so a variety of automatic performances are possible.

Rhythm sound processing (FIG. 12) Next, details #I of the rhythm sound processing (step 76 in FIG. 8) will be explained with reference to FIG.

First, in step 180, the rhythm run flag RHYR
It is determined whether UN is l'', and if it is 'l'' (Y), the process moves to step 182. In this step 182, the least significant bit CLK of the count data of the tempo counter CLK is
Determine whether l, SB are '0'. This determination is made in order to reduce the number of sound timings per measure in rhythm accompaniment to half the number of times when the music is played automatically. For example, in the case of 4 beats, CLK takes a count value of θ~950 within one measure, but regarding rhythm sound, it is 0.2.4...
・94's 48 timing 7, the sound timing is within the -1 measure.

In the determination at step 182, CL KLSB is 10.
If not (N), then it is not the sound generation timing, and the routine returns to the routine of FIG. 8. In addition, in the determination at step 182, if CL KLSB is 0'' (Y), it is the sound generation timing, and the process moves to step 184.

In step 184, rhythm pattern data (PTI~
PTN) is selected, and 2.5 bits of sound generation control data SD corresponding to the count value of CLK are read out and stored in the sound generation control data register RHYDAT. In this case, successive addressing can be performed according to the following equation.

PNTR+l+N+%x (RHYNO-t) +C
LK In this formula, l'-PNTRJ is the starting address of the rhythm data RD stored in the rhythm data read pointer PNTR, [J is the rhythm pattern a, rR) (
YNOJ is the rhythm pattern data J and [CL] stored in the rhythm pattern number register RHYNO.
KJ is the count value of the tempo counter CLK. Referring to Figure 6'26, P of the above formula
NTR is the address of the rhythm pattern number data PN, and adding 1'4 to this means that the data PN is skipped. Also, Ny! on (PNTR+1)!
Adding / means skipping N-items of attribute data PS, and (PNTR-)1+N) represents the starting address of the first rhythm pattern data p'r1.

As described above, the performance data memo IJ 24 stores 96 bytes of each rhythm pattern data for each of four time signatures, three time signatures, and two time signatures.

Therefore, the desired rhythm is determined by %X(RHYNO-1),
You can choose e-turn. For example, R in the above formula
If HYNO is 1, %x(RHYNo-1) becomes O and can be selected as the first rhythm pattern data PTI'. Also, if RHYNO in the above formula is N, %X
(N-1), the (N-1) set of data is skipped, and the Nth rhythm data turn PTN i can be selected.

By changing the rhythm pattern data selected in this way to CLK'40, 2, 4, .

In step 186, the pronunciation control data SD stored in RHYDAT in step 184.
30 to control the generation of impact sound. That is, since the eight percussion sound forming channels of the rhythm TG 30 have already been assigned percussion sounds to be formed in step 11H of FIG.
Specify whether to produce or not produce sound for each channel of 8, and draw the accent (volume) AC+J' at the corresponding pronunciation timing.
By specifying this, a percussion sound signal for one sound generation timing is generated and supplied to the sound system 34. As a result,
After the sound system speaker produces one or more percussion sounds corresponding to the percussion sound signal,
Return to the routine of FIG.

The process of reading out the sound generation control data SD and controlling the rhythm TG3oY system as described above is carried out in the same manner at each sound generation timing thereafter. When the reading of the rhythm pattern for the l small portion is completed, the process returns to the beginning of the same rhythm pattern and continues reading. This results in the selected rhythm,? In the middle of such rhythm accompaniment, in which the rhythm accompaniment is automatically performed according to the turn, the rhythm de-turn number in RHYNO is executed at step 114 in FIG.
When 9 is changed, the instrument group number and rhythm pattern are also changed, creating a rhythmic accompaniment performance with a wide variety of changes.

Note that when RHYRUN is set to "+" in step 112 of FIG. 9, the determination result of step 1800 of FIG. 12 becomes negative (N), and the process returns to the routine of FIG. 8.

Therefore, the rhythm accompaniment stops after this point.

This invention is not limited to the above embodiments, but can be practiced with appropriate modifications.For example, the following (tl
~ (6) can be done.

(1) Only some of the 16 channels (for example, half of the 8 channels) may be used for automatic playback of songs. For this purpose, limit the channels to be used in the header and do not use the end register END. The focus v corresponding to the channel may be set to 1" in advance. Channels not used for automatic performance can be used for manual performance sound generation. (2) Transferring the focus value from the external storage medium 32 to the internal performance data memory The external storage medium 32 may be directly accessed without transferring the performance data.

(3) Performance data for a plurality of performances may be stored in the external storage medium 32, and automatic performance and rhythm accompaniment may be performed by selecting songs.

(4) It is also easy to include repetition codes in music data and control data.

(5) A rhythm pattern memo IJ' may be provided separately from the performance data memory, and the rhythm pattern may be selected from these two memories.

(6) The performance data memo + 724 stores music data other than rhythm patterns, control data, etc., while the RAM or IIi layer removable ROMK provided separately from the memory 24 stores multiple songs (for example, waltz songs). ) may be stored with a large number of commonly usable rhythm patterns (for example, waltz rhythm patterns), and the rhythm pattern χ may be selected from this RAM or ROM. In this case, if the type of music to be played changes from a waltz style to a Latin style song, for example, should a rhythm pattern storage medium other than the external storage medium 32 be used to rewrite the contents of the RAM to a Latin style song? , or an RO that stores Latin rhythm patterns instead of a ROM that stores waltz rhythm patterns.
All you have to do is install M4.

〔Effect of the invention〕

As described above, according to the present invention, the rhythm pattern is selectively read out from the second storage section based on the rhythm pattern selection information sequentially received from the first storage section, and seven rhythm accompaniments are performed. The troublesome operation of selecting a rhythm for each song is no longer necessary, and by storing arbitrary rhythm pattern selection information in the first storage unit, it is possible to create a rhythm accompaniment that is suitable for the performance song and is rich in variety. is obtained.

[Brief explanation of drawings]

1 is a block diagram showing the circuit configuration of an electronic musical instrument according to an embodiment of the present invention; FIG. 2 is a formant diagram of performance data in an external storage medium; FIG. 3 is a header format diagram; The figure is a format diagram of music data, and FIG. 5 is a format diagram of control data. Figure 6 is a format diagram of rhythm data, Figure 7 is:
Flowchart of the main routine, Figure 8 is a flowchart of the interrupt routine, Figure 9 is a flowchart of regulatory data processing, Figure 10 is a flowchart of music data processing, and Figure 11 is a flowchart of processing by channel. Flowchart FIG. 12 is a flowchart of rhythm sound generation processing. 10...Path, 14...Control switch circuit, 16.
...Central processing unit, 18...Program memory, addition...
・・Working memory, etc.・Reading device, lyrics・performance data memory, song・tempo oscillator, public music tone generator, addition・rhythm tone generator, 32・external Storage medium, 34...Sound system. Applicant: Nippon Gakki Rizo Co., Ltd. Agent Patent Attorney: Toshiaki Izawa Figure 2 (Hamaso Tedai 7-JR-7) Figure 4 @ (Gakugeki Theta) Figure 5 (Total Distribution Data) Fig. 6 (Suite Mudita) Fig. 8 (Konsha→N)

Claims (1)

[Claims] (a) A first storage unit that stores music information corresponding to the note progression of a desired song together with rhythm pattern selection information; (b) A second storage unit that stores a plurality of rhythm patterns. (c) a first reading means for reading out the music piece information and the rhythm pattern selection information from the first storage unit; (d) a first reading means for reading out the music piece information and the rhythm pattern selection information from the first storage unit; (f) a musical tone generating means for generating a musical tone signal; (e) a second reading means for selecting and reading out one of the rhythm patterns from the second storage section based on the read rhythm pattern selection information; ) Rhythm sound generating means for generating a rhythm sound signal according to a read rhythm pattern.