JPH03200289A - Automatic player - Google Patents

Abstract

PURPOSE:To enable automatic performance wherein individually shows up by carrying on performance by a player himself in player's tempo over timing and normalizing note data according to performance data on the performance so that the individuality of the player is maintained. CONSTITUTION:The player carries on the performance operation, the performance data based upon the performance operation is outputted, and time timing is inputted to the player together with the performance operation by the player, thereby outputting time timing data based upon the input operation. Time data indicating the input timing of the performance data and time timing data outputted from a performance operation means and a time timing input means in order is generated and written in storage means 7 and 8 together with the performance data and time timing data. Further, the time data stored in the storage means 7 and 8 are normalized and written so that sections determined by time timing data in the storage means 7 and 8 which adjoin to each other with time become equal in time. Consequently, the performance operation wherein the individuality of the player shows up becomes possible.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic performance device, and more specifically, the present invention relates to an automatic performance device, and more specifically, it allows a player to input performance information for automatic performance while actually performing the performance operations, and performs the performance. The present invention relates to an automatic performance device that reproduces information and performs automatic performance.

(Prior Art) When inputting performance note data in an automatic performance device for automatically playing a musical instrument, there are two main ways of inputting the data.One is to input musical score information directly to a computer, sequencer, etc. One is to input encoded note data using a keyboard, etc., and the other is to have the performer play the instrument directly, develop it into digital note information through an interface such as MIDI, and store that information in a sequencer, etc. The former generates accurate note f5 information from the musical score, so it is faithful to the notes, but the performer's individuality regarding performance timing etc. is completely ignored.On the other hand, the latter Since the note information is accumulated as the notes are played, the individuality of the performer is directly reproduced.In the latter case, for example, one or more performers can play different musical score parts separately and automatically When creating note information for performance, compositing them (generally implemented as a function such as a sequencer), and performing automatic performance using unsamples, it is necessary to synchronize the rhythm between each performance. For this purpose, conventionally, the performer performs in accordance with the time signature of a metronome, etc. In other words, the performer listens to the time signature signal of the metronome etc. through headphones etc. and performs in accordance with the time signature. It is something.

[Problem to be solved by the invention]

However, as mentioned above, when a performer performs in time with a metronome, etc., the metronome's time signature is mechanical and is always marked at regular intervals, so there is a difference between the tempo of the performer and the metronome. There is often a difference between the two. Also,
This creates a problem in that the performer's performance becomes awkward because the performer tries to force the performance to match the metronome.

The problem of the present invention is to allow the performer to perform automatically while keeping the beat at his or her own tempo, and to normalize the note data based on the performance data so as to preserve the individuality of the performer. The purpose is to do so.

[Means to solve the problem]

The present invention is based on an automatic performance device that causes the musical tone generating means to perform automatic performance by sequentially outputting performance data to the musical tone generating means. For example, it is a dedicated sequencer that only records and reproduces automatic performance data. Alternatively, it may be an electronic musical instrument with an automatic performance function that includes a built-in musical tone generating means itself.

First, there is a performance operation means for causing a player to perform a performance operation and outputting performance data based on the performance operation.

The means is, for example, a keyboard. The means may be a keyboard such as a keyboard connected to the outside via MIDI or the like.

Next, it has a beat timing input means that allows the performer to input beat timing along with a performance operation, and outputs beat timing data based on the input operation. The means is, for example, a foot switch that allows the performer to input the timing of each beat using a quarter note beat. Note that, of course, the time signature may be a half note, an eighth note, a sixteenth note, or the like.

Subsequently, it has storage means for storing performance data and beat timing data. The means is, for example, a RAM or the like.

In addition, time data indicating the input timing of the performance data and beat timing data sequentially outputted from the performance operation means and the beat timing input means based on the performance operation of the performer is generated, and is stored together with the performance data and beat timing data. It has a writing means for writing into the means. The means includes, for example, a timer circuit that measures the timing from the input point of the previous performance data or beat timing data to the current input point of the same data, and generates time data as its output.

Furthermore, time data rewriting is performed to normalize and rewrite each time data stored in the storage means so that the time of each section determined by each temporally adjacent beat timing data stored in the storage means is equal to each other. have the means. The means includes a first real-time calculation section that calculates the real time of each beat interval of, for example, a quarter note, determined by each beat timing data, for example. Further, the second real time calculation section calculates the real time of, for example, a 96th note from the real time of each time signature section of, for example, a quarter note determined by the calculation section. Furthermore, it calculates how many times the time value of each time data in each time signature section is the real time of, for example, a 96th note found in the second real time calculation section, and uses the calculation result to calculate the time data. It has a rewriting section for rewriting.

After the above-mentioned rewriting operation, the musical performance reproducing means sequentially reads out each performance data from the storage means and outputs it to the musical sound generation means at a timing based on each time data stored in the storage means.

For example, when the performance data consisting of each address of the storage means is read out, the means outputs it to the musical sound generation means, and when the time data is subsequently read out, the timing is set by a timer circuit, for example, for the time data. , and after that timing, the next performance data is read out and outputted to the musical tone generating means, and this operation is repeated to cause the musical tone generating means to perform automatic performance.

[For production]

The performer can operate the performance using the performance original means while setting his or her own tempo using a beat timing input means such as a foot switch. Therefore, the performer does not have to force his/her performance to match a fixed time signature such as a metronome, and the performance does not become awkward.

Then, prior to performance reproduction, the time data rewriting means normalizes and rewrites each time data so that the time of each section determined by each temporally adjacent beat timing data becomes equal to each other. That is, for example, each section is normalized to have 24 units of time in units of 96 notes.

Moreover, in each beat interval, normalization is performed while preserving the relative time data ratio of each time data within each interval. This allows the performer to perform while inputting the tempo for each beat, so that the relative time relationship between the time data within each beat is saved as the performer played, and also for each beat. The timing between them is precisely equalized. Therefore, while preserving the individuality of the performers, it is possible to eliminate rhythm deviations between performers or between each performance movement.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail.

This embodiment is realized as an electronic musical instrument having an automatic performance function. Note that the performance operation section such as a keyboard and the musical tone generation section that performs the operation of producing musical tones may be connected to the outside via a MIDI interface and realized as a sequencer or the like having only an automatic performance function.

FIG. 1 is an overall configuration diagram of an embodiment of the present invention.

First, the keyboard section 1 is made up of a plurality of keys (not particularly shown), and the player inputs performance data through the keyboard section 1. An input sense circuit 2 is connected to the keyboard section 1, and based on the output of the keyboard section 1, detects key presses and key releases, outputs key codes corresponding to musical notes, herocity data corresponding to operation speed, etc. do.

Separately from this, there is a performance information input section 3 that includes a configuration for inputting the player's own tempo at the same time as the performance operation. The performance information input unit 3 includes a performance mode changeover switch 3a for selecting a performance mode consisting of three modes: performance, recording, and playback, a foot switch 3b for inputting the tempo during performance, and a foot switch 3b for inputting the tempo during performance, and a switch for ending recording. Alternatively, a switch 3c for interrupting playback, a tempo selection switch 3d for determining whether the tempo at the time of recording is input by the performer or a constant tempo, and a tempo determination switch 3e for determining the tempo value at the time of playback. , 3f.

The output of the performance information input section 3 is input to the performance information detection circuit 4. The performance information detection circuit 4 detects the state of each of the above-mentioned switches 3a to 3f, and depending on the type of the switch, the performance information storage section 7 and the performance data storage section 8, which will be described later, are Accumulates detection data.

A central control unit (CPU, hereinafter the same) 5 is connected to a program storage section 6, a performance information storage section 7, and a performance data storage section 8 via a bus.
Each process is performed based on the programs stored in the . Specifically, the performance data input by the performer is extracted from the input sense circuit 2 and stored in the performance data storage section 8, and at the same time, the tempo information output from the foot switch 3b of the performance information input section 3 is stored as performance information. The data is stored in the performance data storage section 8 via the detection circuit 4. Further, each piece of information output from the switches 3a, 30 to 3f of the performance information input section 3 is stored in the performance information storage section 7 via the performance information detection circuit 4 as needed. The contents of the performance information storage section 7 are configured as shown in FIG. 2, for example. When the switch is pressed, "l" is displayed in the table in the figure, and when the switch is not pressed, "l" is displayed.
0" is stored. The CPU 5 constantly monitors the performance information stored in the performance information storage section 7, reads a necessary program from the program storage section 6, and executes the program depending on the state of the performance information.

Further, in this embodiment, a timer unit 9 that counts the tempo of the performer, and an LE unit that displays the tempo etc. during playback of the performance are provided.
A D display section 10 and a tempo value conversion table 11 that stores the tempo at the time of performance reproduction are connected to the CPU 5 via a bus.

Furthermore, in this embodiment, the performance data storage section 8
A musical tone generating circuit 12 that generates a digital musical tone signal based on the performance data stored in the CPU 5 is connected to the CPU 5 via a bus, and a D/A converter 13 and an amplifier 14 that convert the digital musical tone signal into an analog musical tone signal are connected to the CPU 5. and a speaker 15 are connected in sequence to constitute a musical tone generating section.

The operation of the electronic musical instrument shown in FIG. 1 having the above configuration will be explained while explaining the detailed configuration of each part.

FIG. 3 shows a series of execution steps from input of performance data by the player to playback (automatic performance).

First, the performance mode changeover switch 3a of the performance information input section 3 is selected by the performer to indicate the recording mode.
In addition, the tempo selection switch 3d is set to a mode in which the performer himself inputs the tempo (FIG. 3 Ml).

Next, when the player presses the foot switch 3b, the performance start mode is entered (M2 in FIG. 3).

The performer inputs musical tones from the keyboard section 1 while inputting the tempo himself from the foot switch 3b. The above-mentioned tempo information and musical tone information are stored as performance data in the performance data storage section 8 as needed through the input sense circuit 2 and the performance information detection circuit 4, respectively.

At this time, the performance data stored in the performance data storage section is as shown in FIG. 4, for example. The first address is 5t when the foot switch is pressed for the first time and the performance starts.
The art event data is written and the timer part 9 starts operating. Thereafter, each time musical tone information or tempo information is input from the keyboard section 1 or the foot switch 3b, corresponding event data (key-on event, key-off event, tempo event, etc.) is written, and the event is measured by the timer section 9. and the time until the next event is written (M3 in Figure 3).

The above operations are repeated until the switch 3C indicating the end of the performance is pressed, at which point an end event indicating the end of the performance is written in the performance data storage section 8 (see FIG. 4). When the performer presses the switch 3C indicating the end of the performance, the information is stored in the performance information storage section 7,
The central controller 5 reads the information stored in the performance information storage section 7 and finishes storing the performance data in the performance data storage section 8 (M4 in FIG. 3).

Subsequently, the CPU executes a process of rewriting only the time data of the note data stored in the performance data storage section 8. In this process, the time axis is normalized based on the tempo signal that the performer inputs from the foot switch 3b during performance, and only the time data of the performance data stored in the performance data storage section 8 is rewritten. At the same time, the LED indicating that time data is being rewritten is lit on the LED display unit 10 (M5 in FIG. 3).

After that, when the performance tempo during playback is inputted by the switches 3e and 3f of the performance information input section 3, the central control device 5 displays the tempo on the LED display section 10 in accordance with the signal,
The value is stored in the tempo value conversion table 11 (M6 in FIG. 3).

Then, when the performer sets the performance mode changeover switch 3a to 'auto' indicating automatic playback and presses the foot switch 3b, performance data with normalized performance tempo is sequentially read out from the performance data storage section 8, It is played back (M7 in Figure 3).

Next, the time data rewriting process shown in FIG. 3 M5, which is particularly relevant to the present invention, will be described in detail.

FIG. 5 is a detailed operational flowchart of the time data rewriting process M5 in FIG. 3, and this control is realized by the CPU 5 executing the program in the program storage section 6. Incidentally, each of the variable data ADR, Rs, Re, P, Q, T, will be secured as In the following description, each variable data is given a functional name as appropriate.

Furthermore, in the following processing, an example will be explained in which performance data as shown in FIG. 7(a) is stored in the performance data storage section 8 of FIG. 1. In the example of FIG. 7(a), E2
~E6 are data such as note-on events and note-off events based on performance operations on the keyboard section 1, and P4
, P5 is tempo event data based on the tempo input operation with the foot switch 3b. And t1 to t6
is time data indicating the timing between each of the above-mentioned event data.

In FIG. 5, first, the state of the switch 3d is detected at Sl, and it is determined whether or not the tempo is being input by the performer. Unless the tempo is input by the performer, the following process is not performed, and the time data rewriting process M5 in FIG. 3 is not substantially executed.

Conversely, if the tempo is being input by the performer, S
In step 2, initial state setting processing is performed. Here, 321
And in S22, address A of the performance data to be processed next.
Both DR and addition start address Rs are set to 1. Then, in S23, the value of the timer value Q, which will be described later, is 255.
Further, in S24, an LED indicating that time data is being rewritten is lit on the LED display section 10. In addition,
The timer value Q set in S23 may be any value other than 0. The meaning of this will be explained later.

Next, in S3, in the storage contents of the performance data storage section 8, focusing on the tempo event that indicates that the foot switch 3b has been pressed, the process starts from each tempo event to the next tempo event (5 tarts at the start of the performance). The actual time (from the event to the first tempo event) is calculated.

That is, first, cumulative time data X (described later) is reset to O in S31.

Next, in S32, the contents of the address indicated by the address ADH of the performance data storage section 8 are read out and set in the variable Y. Now, the content of address ADR is 1 (S21
), the odd address on the performance data storage section 8 has the seventh
As shown in the figure, time data and event data are stored in even-numbered addresses, so the first time data is stored in variable Y. The time data read into this variable Y is accumulated into cumulative time data X (which has been reset in S31) in S33.

Subsequently, in S34, the address ADR is incremented to an even address, and in S335, the event data stored in the even address of the performance data storage section 8 is read into the variable Z.

While this event data is not an end event or a tempo event, the determination in S36 is NO, the address ADR is changed to an odd address again in S37, S32 is executed again, and the next time information is read into variable Y, and so on. Then, the processes from 332 to S37 are repeated.

After repeating the above processing, the tempo event or end event is read out to the variable Z in S35, and when the determination in 336 becomes YES, the cumulative time data X becomes 5ta.
rt event or the actual time from the previous tempo event to the next tempo event. This real time indicates the cumulative time of one section input by the performer as one beat.

In the example of FIG. 7(a), first, in the section from the 5 tart event to the tempo event P4, S32 to S3 of FIG.
7 is repeated. As a result, the value of cumulative time data X is: X=t++tz+t, :+...
(1) (see FIG. 7(b)).

Next, in S35, the tempo event or end event is read into the variable Z, and when the determination in S336 becomes YES,
Proceed to S4. In S4, if the content of variable Z is not an end event but a tempo event, the determination of 341 is YE.
The result is S, and the process advances to S42. Here, cumulative time data
The number of seconds of a 96th note in the one-beat interval represented by is calculated. In this embodiment, the player is instructed to operate the foot switch 3b with a quarter note as one beat. Since a quarter note is equal to 24 96th notes, in order to calculate the actual time of a 96th note, the value of cumulative time data Just divide it.

A detailed flowchart of this operation is shown in FIG.

As shown in the figure, the division operation at 342 in FIG.
At 5423 and 5424 in Figure 6, 1 from cumulative time data
This is achieved by sequentially subtracting 2 by 2, and the result is divided by 11 and 12, and the timer value Q is finally accumulated to 5423 (reset to O in S421).
It can be found as

After the above operations, at 343 the current address AD
The value calculated by IK from R is set as the addition end address Re.

As a result, the addition end address Re is 332 to 3 mentioned above.
This is the address where the last time data in the section detected in the process of No. 37 is stored. This role will be explained later.

Note that in S4, if the content of variable Z is an end event, the determination in step 341 is No, and the process advances to S44. Here, the above-mentioned timer value Q is not calculated, and the value obtained by adding 1 to the current address ADR is the addition end address Re.
It is said that As a result, the addition end address Re is determined as the end address Re in the section detected in the processing from 332 to S37 described above
This is the address following the last address where the event is stored. This role will be explained later.

The process of S4 will be explained below using an example of the section from the 5tart event to the tempo event P4 in FIG. 7(a) or (b). That is, since the value of the cumulative time data X is X = t+ 1 tz + tz as in the above-mentioned equation (1), the timer value Q is: Q= (t++b+b) / 24...
(2) becomes. Also, since the current address ADR indicates address 6 in FIG. 7(a), the addition end address Re is 343, Re=6-1=5...13), and the storage of time data t3 is completed. It becomes the address.

Next, in S5 of FIG. 5, each time data between event data based on performance operations within one beat is rewritten into information indicating how many times longer the time is in units of 96th notes.

First, in 5501, the note cumulative value P is reset to O for the operation described later, and then, in 5502,
Addition start address Rs is set to address ADH. Then, by repeating the processes 8503 to 5508, in the section from the addition start address Rs to immediately before the addition end address Re (3503), odd addresses where time data is stored are sequentially specified (S50
8), each time data is rewritten as follows.

That is, in 5504, the time data indicated by the current address ADR is stored in the variable Y, and in 5505, the time data is divided by the above-mentioned timer value Q to obtain musical note data T. Then, this note data T is newly set at the current address ADR. Now, the timer value Q is
Since it represents the actual time of a 96th note in the current section, this calculates how many times each time data is a 96th note. Also, in the above repetition, at 5507,
The note data T, which is the above-mentioned normalized time data, is sequentially accumulated in the note accumulation value P (which was reset in 8501 described above).

The role of this cumulative note value P will be described later.

Then, the current address ADH is the addition end address Re
, the determination in 5503 becomes YES, and if the content of variable Z is not an end event but a tempo event, the determination in 5509 becomes YES, and 5510~
The process of 5512 is executed. Here, the last time data of the current section is normalized, and the addition start address R
s and address ADR are reset.

First, the last time data is not normalized as in step 3505 described above, but is normalized in step 5510. In other words, since one beat (quarter note) is equal to 24 96th notes, the final time data is calculated from 24, which indicates the entire beat, to each note corresponding to each time data calculated by repeating 5505 steps. It is expressed as a value obtained by subtracting the accumulated value of data T. This accumulated value is 850 as mentioned above.
Since the cumulative note value P has been determined by the process in step 7,
The final time data is found by subtracting the cumulative note value P from 24. By processing in this manner, it is possible to prevent the accumulation of errors due to the division processing in FIG. 6 corresponding to 342 described above.

Next, in 5511, the value obtained by adding 2 to the value of the addition end address Re is set as a new addition start address Rs. now,
Since the addition end address Re is the address where the last time data is stored in the section detected in the process from 332 to S37 described above, by adding +2, the first time data of the section next to the above section is stored. This is the address where it is stored. And this addition start address Rs is
At 5512, address ADR is also set at the same time.

After this operation, the process returns to step 33 described above, the real time of one beat's worth of time until the next tempo event is calculated, and the same process as described above is repeated.

The process of S5 above will be explained as follows using an example of the section from the 5tart event to the tempo event P4 in FIG. 7(a) or (b). That is, S42
In step 5505, each time data L and 2 are normalized as follows using the timer value Q determined as in the above-mentioned equation (2).

Also, when the normalization of Ll and t2 is completed, 550
In the process of step 7, the cumulative note value P is calculated as P'= from equation (4).
t 1 / Q + t 2 / Q ・
...(5). According to equation (5), the last time data t3 of the above section is 5510, T=24-P=24-D+/Q+tz/Q)...(6)
It is normalized as follows.

After this operation, the addition start address Rs of the next section is 55
11, from the above equation (3), Rs=Re+2=
It is determined as 5+2=7 4 + (7), and becomes the storage address of the first time data L4 in the section determined by the tempo events P4 to P5. This addition start address Rs
is simultaneously set as address ADR in 5512.

As described above, 5 tarts in FIG. 7(a) or (b)
Time data L1, L2 and t, 3 between each event data for one beat from event to tempo event P4
is normalized.

Next, the tempo event P4 in FIG. 7(a) or (b)
Exactly the same processing as in the above case is executed for the section from 5502 to 5503, but since no events other than the tempo event have occurred in this section. 8503 at the time
The determination immediately becomes YES, and the processes 8504 to 5508 are not executed. Then, the determination in 5509 is YES, and the process in 5510 is executed. In this case, the cumulative note value P remains at the value 0 that was reset at 5501, so the value obtained by normalizing the time data t4 at 5510 is 2.
It becomes 4. That is, since the time data L4 is the time of one beat from tempo event P4 to P5, that is, the time of a quarter note, it becomes a normalized value representing 24 96th notes.

For the above operation in S5, end in S5
When normalizing the final section up to the event, the procedure is as follows. That is, it is set in the above-mentioned 835 and S36
The content of the variable Z determined in is the end event, and the content of the addition end address Re is as follows by 344 described above.
This is the address following the final address where the end event is stored. Therefore, by repeating steps 3503 to 350B, normalization processing based on step 3505 is performed on all time data in the final section up to the end event. In this case, the timer value Q in the previous interval is used (the timer value Q has not been calculated in S44). Further, the normalization processing by 3510 is not executed. These processes take into consideration that the last section may last less than one night.

In this case, when the address ADR becomes equal to the addition end address Re and the normalization of the last time data of the final section is completed, the determination in 5503 becomes YES, the determination in 5509 becomes No, and in S6, the first The LED indicating that the time data is being rewritten on the LED display section 10 shown in the figure is turned off, and the time data rewriting process M5 in FIG. 3 is completed.

The above processing will be explained as follows using an example of the section from tempo event 1-P5 to the end event in FIG. 7(a) or (b). That is, first, since the process of S44 is executed instead of S42, the value of the previous interval Q=L/24 (8) is used as the timer value Q.

Using this timer value Q, each time data t in the final section
, and L6 are each normalized at 5505, and so on.

By the above operation, for example, 5t shown in FIG. 7(b)
From art event to tempo event P4, from tempo event P4 to tempo event P5, and tempo event P5
The interval of one beat from to the end event is normalized to be 24 time units with 96 notes as a unit. Moreover, in each beat interval, normalization is performed while preserving the relative time ratio of each time data within each interval.

This allows the performer to perform while inputting the tempo for each beat, so that the relative time relationship between the time data within each beat is saved as the performer played, and also for each beat. The timing between them is accurately normalized to 24 units of 96 notes. Therefore, while preserving the individuality of the performers, it is possible to eliminate rhythm deviations between performers or between each performance movement.

In addition, in the performance input process of FIG. 3 M3, the performer
If the switch 3c of the performance information input section 3 in FIG. 1 is pressed without playing anything, only the end event is stored in the performance data storage section 8. Therefore, 332 in FIG.
~The process in S35 is executed, but it has no meaning, and the process in S35 is executed immediately.
After 6 becomes YES, 341 becomes NO. 55 more
03 is NO only once, and the processes 5504 to 5508 are executed, but these processes are also meaningless. and,
The process proceeds from 5503 to 5509 to S6, and virtually nothing is done. In this case, a value other than 0 is input in advance to the timer value Q in S23 so that division by the value 0 does not occur in 5505.

In this embodiment, the time signature by the footstool 3b in Fig. 1 is set as a quarter note, but the time signature is not limited to this, and it is of course possible to set the time signature as a half note, an eighth note, a sixteenth note, etc. , and they can be selected arbitrarily. Further, in this embodiment, normalization was performed in units of 96th notes, but normalization may be performed in units other than 96th notes.

〔Effect of the invention〕

According to the present invention, the performer can operate the performance using the performance operation means while setting his or her own tempo using the beat timing input means such as a foot switch. Therefore, the performer does not have to force his performance to match a fixed time signature such as a metronome, and it is possible to carry out performance operations that make use of the performer's individuality.

Then, when the performance is played back, the relative time relationship between the time data within each beat is saved as performed by the performer.
Moreover, since the timing between each beat is accurately equalized and automatic performance is performed, it is possible to eliminate rhythm deviations between performers or between performance movements while preserving the individuality of the performers.

This allows, for example, when one or more performers play different musical score parts separately to create note information for automatic performance, and then synthesize them to perform automatic performance using unsamples. , it becomes possible to accurately synchronize the rhythm between each performance while preserving the individuality of the performer.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention; FIG. 2 is a diagram showing the storage contents of the performance information storage unit 7; FIG. 3 is an operation flowchart showing the execution procedure of the embodiment; Fig. 4 is a diagram showing an example of the storage contents of the performance data storage unit, Part 5 is an operation flowchart of time data rewriting processing, Fig. 6 is an operation flowchart of division processing, and Fig. 7 is:
FIG. 3 is an explanatory diagram of the operation of this embodiment. Engineering...SS1 panel, 2...Manual sense circuit, b 4 ・ ・ 5 ・ ・ 6 ・ ・ 7 ・ ・ 8 ・ ・ 9 ・ ・ 10 ・ ・ 11 ・ ・ 12 ・ ・ 13 ・ ・ 14 ・ ・15 ・ ・ Performance information input section, ... foot switch, performance information detection circuit, central control unit (CPU), program storage section, performance information storage section, 1.0 copy of performance data record, part of the timer, LED display section, Tempo value conversion table, musical tone generation circuit, D/A converter, amplifier, speaker.

Claims (1)

[Scope of Claims] 1) An automatic performance device that sequentially outputs performance data to a musical sound generating means to cause the musical sound generating means to automatically perform a performance, wherein the performer is made to perform a performance operation, and the performance is performed based on the performance operation. performance operation means for outputting data; beat timing input means for causing a performer to input beat timing along with the performance operation; and outputting beat timing data based on the input operation; and storing the performance data and the beat timing data. a storage means; generating time data indicating input timings of the performance data and the beat timing data sequentially outputted from the performance operation means and the beat timing input means based on a performance operation of a performer; writing means for writing into the storage means together with the beat timing data; and writing means for writing into the storage means so that the time of each section determined by each temporally adjacent beat timing data stored in the storage means is equal to each other. time data rewriting means for normalizing and rewriting each of the stored time data; and after the rewriting operation, rewriting each of the performance data from the storage means at a timing based on the time data stored in the storage means. an automatic performance device comprising: performance reproduction means for sequentially reading out and outputting the musical tone generation means to the musical tone generation means. 2) The automatic performance apparatus according to claim 1, wherein the beat timing input means is a foot switch that allows the player to input the beat timing with his/her foot.