JPS6040030B2 - automatic composer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic composing machine suitable for composing assigned pieces for listening training or performance practice in the field of music education.

In music education, listening training or performance practice (including rhythm practice) is widely practiced.

Generally, to conduct listening training, a teacher plays a short piece of music of about 2 to 4 measures, and has the students respond verbally or write it down on a musical staff. Furthermore, in order to practice playing, a similar short piece of music is printed in advance and distributed to the students, and the students are asked to perform based on it. In the past, it has been customary for the teacher to compose the theme music used for such listening training or performance practice, or to use a portion of an existing song.

However, if the composition of the assigned piece is left to the teacher, there is a problem that it is difficult to obtain a musically excellent assigned piece because the composition ability of the teacher varies among individuals. When using this method, there was a problem in that the room for choice was narrowed and the required songs became uniform.

This invention was made in order to solve the above problem, and its purpose is to create an automatic music composer that can automatically create a series of pitch data corresponding to the above-mentioned task piece. Our goal is to provide the following.

In order to achieve the above object, the present invention stores a series of pitch data in a predetermined pitch data memory, and supplies these pitch data with a predetermined low-level signal. each time, the pitch data that is sequentially read out and then sequentially extracted is compared with a predetermined musical condition, and the pitch data that matches the condition is sequentially stored in a predetermined selected data memory, and Each time a series of data reading operations is completed, one of the plurality of pitch data stored in the selection data memory is randomly selected, and each time the series of readout operations is repeated, whereby the selection is The pitch data read from the data memory is sequentially selected as one of the constituent sounds of the song.

Here, particularly in this embodiment, the pitch data extracted from the selected data memory and adopted as the constituent notes of the song is stored in a predetermined temporary storage memory, and when compared with the musical conditions, , the previously selected pitch data stored in the temporary storage memory is used as the comparative judgment material.

In this way, in the selection data memory, a plurality of pitch data that are most suitable for the later constituent tones are obtained among the two competing pitch data that constitute a predetermined song, and then these pitch data are obtained. By randomly reading out one of the plurality of pitch data of
The individual pitch data match the musical conditions and form a new arrangement relationship that is extremely unpredictable, giving the entire piece an extremely novel style.

An embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the overall configuration of an automatic music composer according to the present invention.

The automatic music composition machine shown in this embodiment has two operating modes: a composition mode and a performance mode.

In the composition mode, as will be described later, a predetermined composition operation is performed based on each data read out from the pitch data memory and the pattern memory, and the composed program data is stored in the score data memory. Ru. On the contrary,
In the performance mode, automatic performance is performed based on the music stored in the musical score data memory, and at the same time the CRT
The song is displayed as a musical score on the screen, and the printer prints out the musical score corresponding to the song, for example, on paper using a staff score. Therefore, when conducting listening training in a music education class, after performing the operations in the composition mode described above, the students are asked to listen to a piece composed using the automatic performance function described above, and the results are displayed on the CRT.
All you have to do is check the staff notation displayed or printed on the screen or printer paper. Also, if you want to practice playing,
Similarly, after operating the composition mode, all you have to do is write out the composed piece of music outside the pudding, distribute it to the students, and have them perform it.

Next, these two operation modes will be explained in order, starting with the composition mode.

To set the operation to the composition mode, turn on the composition start switch SWI. When SWI is turned on, in response to the rising edge of "1", the differential circuit 1 outputs a minute width "1" pulse (hereinafter referred to as ΔCOMST), and at the same time, the RS flip-flop 2 outputs a "1" pulse.
”, and its Q output (hereinafter referred to as COMP) is set to “1”. In this way, when ΔCOMST and COMP are output, the RS
Flip-flops 3 and 4 are set by ΔCOMST supplied through OR circuits 5 and 6, respectively, and the Q output of flip-flop 3 (hereinafter referred to as DST) is "1" and the Q output of flip-flop 4 is "0". Become.

As a result, the address counter 7 starts counting the clock J supplied via the AND circuit 8, and the counter 9 is released from reset and starts counting the clock J.

Thereafter, the count value of the address counter 7 continues to increment in synchronization with the low clock, and when it reaches the preset count completion value, the carry-out output C output from the counter 9.
O causes the latch circuit 10' to latch the numerical data output from the random signal generator 11.

The random signal generator 11 is constituted by, for example, a shift register configured to operate as a maximum length counter.

The random signal generator 11 sequentially outputs random numerical data at constant minute intervals. Therefore, as mentioned above, when the counters 7 and 9 start counting and reach a certain count value, the carry out output CO
When the latch circuit 10 is driven by this, random numerical data with no regularity is latched in the latch circuit 1. The count output of the address counter 7 is used as an address signal for the pitch data memory 12, while the latched numerical data of the latch circuit 1 is
It is used as an address signal for the pattern memory 13 as SB.

As shown in FIG. 2, a series of pitch data is sequentially stored in each address of the pitch data memory 12 according to a predetermined code.

Therefore, as described above, when the count value of the address counter 7 increments in synchronization with the clock, the pitch data memory 12
From then on, a series of pitch data stored at each address is continuously output in synchronization with the clock. The read pitch data is then supplied to the discrimination circuit 14 and the gate circuit 15 in parallel. Discrimination circuit 1
4 determines whether the pitch data read out from the pitch data memory 12 matches predetermined musical conditions, and various types used for this determination operation are used. The musical conditions are stored in the condition data memory 16.

Further, the grade selection switch 17 is used to selectively combine various musical conditions stored in the condition data memory 6. Among these conditions, the conditions selected by the grade selection switch 17 are appropriately combined to perform the discrimination process. In this embodiment, the musical conditions stored in the condition data memory 16 include the following. '1
} The first note is C.

[2] After C, there is only C.

(3) After Hua, there is only Mi.

{4} Increased fourth progression is prohibited.

{5' increase 5th progression is prohibited.

{6} It does not fly more than one octave.

In addition to the above, items selected by the grade selection switch 17 include the following.

[7' Do not use the black key.

'8} Don't use Mi or U.

(9} Do not use C# and F#.

In the discrimination circuit 14, the pitch data read from the pitch data memory 12 is discriminated against predetermined musical conditions, and if it is determined that the condition is met, a matching signal is generated. A "1" pulse is output as OK.

Therefore, when the determining circuit 14 determines that the condition is met, the matching signal OK is supplied to the enable terminal EN of the gate circuit 15 and the clock terminal CK of the counter 18.

The counter 18 is supplied with ΔCO through the OR circuit 19.
It is reset in advance by MST at the start of the composition mode. Also, at this point, the signal DST
is set to “1”, so DST “1”
”, the selected data memory 20 is set to the write mode, and at the same time the selector 21 selects the A input. Therefore, after starting to output a series of data from the pitch data memory 12, the condition is met. is stored at the top address in the selected data memory 20, except for the first determined pitch data.

Furthermore, when a series of pitch data is continued to be provided from the pitch data memory 12, each address in the selected data memory 20 contains a pitch that matches the predetermined musical condition as the first note of a predetermined song. Data will be written sequentially.

Next, when the count output of the address counter 7 reaches a predetermined count completion value, "1" is output from the carry out terminal CO.
A pulse is output, and in response to this “1” pulse, DST
is reset to "0", and at the same time, this carry-out output CO "1" pulse is also supplied to the random selection signal generation circuit 23.

When DST is reset to "0" in this manner, the selected data memory 2 is set to the read mode, and at the same time, the B input of the selector 21 is selected.

The random selection signal generation circuit 23 has an address counter 7.
Every time the carry out output ``1'' pulse arrives,
A data discrimination circuit compares the numerical data DI outputted from the counter 18 and the numerical data ○2 outputted from the random signal generator 11, and discriminates that there is a relationship of DIZD2, and a discrimination output of this data discrimination circuit. In response, a latch circuit latches the numerical data 02 output from the random signal generator 11, and a pulse generating circuit outputs a signal WTC "1" pulse every time the numerical data D2 is latched in this latch circuit. The output of the latch circuit is supplied to the B input of the selector 21 as an address signal of the selected data memory 20.

As a result, when the address counter 7 completes counting, if the value of the numerical data D2 output from the random signal generator 11 satisfies the relationship DIZD2, the numerical data D2 output from the random signal generator is The signal is supplied to the selection data memory 20 as an address signal, and as a result, only one piece of pitch data stored at the address is extracted from the selection data memory 20.

On the other hand, W output from the random selection signal generation circuit 23
The TC “1” pulse is applied to the load terminal LD of the latch circuit 24.
It is also supplied to the write input terminal WT of the musical score data memory 26 via the OR circuit 25 and to the address counter 28 via the OR circuit 27.

CK input terminal. Here, the address counter 28 receives a signal ΔCO supplied via an OR circuit 29.
It is reset by MST at the start of composition mode. Therefore, as described above, when one of the pitch data is retrieved from the selection data memory 20 in response to the numerical data D2, this read pitch data is latched in the latch circuit 24, and the musical score data memory It is written to the first address within 26. Further, the signal WTC is delayed by one clock period via the D-type flip-flop 30, and then outputted via the clear input CLR of the selected data memory 20 and the OR circuit 5 as the read start signal NEXT.
It is also supplied to the set input S of the S flip-flop 3. Therefore, when one piece of pitch data is read from the selection data memory 20, immediately after that, the selection data memory 20 is cleared, and at the same time, the signal DST becomes "0" and the address counter 7 increments. At the same time, the selector 21 is switched to the A input side again.

Next, a second series of pitch data is retrieved from the pitch data memory 12, and in the same manner as described above, the condition discriminating circuit 14 determines that the pitch is not the same as the one that is read out.
A match with a predetermined musical condition is determined.

In the second and subsequent determination operations, the coincidence determination operation is performed based on the previously selected pitch data latched in the latch circuit 24 and in accordance with predetermined musical conditions. That is, '21 to '6) are executed under the musical conditions described above.

If you get stuck with {1}' in the above musical conditions, use △COM
A condition determination operation is performed based on the strong beat signal output from the ST and pattern memory 13.

In this way, when the second series of pitch data has been output from the pitch data memory 12, the selected data memory 20' is set to the output mode in the same manner as described above, and the selector 21 is set to the output mode. B input is selected, and one of the pitch data is read out from the selection data memory 20 based on the numerical data ○2 output from the random signal generator 11 in the same manner as described above, and the second pitch data in the score data memory 26 is read out. is stored at the address. At the same time, the pitch data is also latched in the latch circuit 24. By repeating the above operations, the score data memory 26
At each address, a sound is stored in the selection data memory 20 based on meeting musical conditions and read out at random based on numerical data D2 output from the random signal generator 11. High data is stored sequentially.

On the other hand, each time the random selection signal generation circuit 23 outputs the signal WTC, the count value of the address counter 31 that specifies the LSB of the pattern memory 13 is incremented by -.

Here, in the pattern memory 13, as shown in FIG. 3, a plurality of note arrangement pattern storage areas (six in this example) are provided, each of which specifies the numerical data latched by the latch circuit 10 as the MSB. ing.

Each of these storage areas is composed of a plurality of addresses, and each of these addresses is configured so that the count value of the address counter 31 is specified as BB.

As shown in FIG. 3, in each address of each note arrangement pattern storage area, a plurality of note arrangement data constituting a predetermined note arrangement pattern are sequentially stored from the first address. In particular, a predetermined end code is stored at the final address of each note arrangement pattern data. Further, the address counter 31 is configured to be released from reset by ΔCOMST supplied via the OR circuit 32 and increment each time the signal WTC arrives. Therefore, as described above, each time one of the pitch data is read out from the selection data memory 20, the LS of the pattern memory 3
B is incremented by -, and the note data stored at each address of the pattern area specified by the MSB is sequentially read out.

Next, the musical note data extracted from each pattern area of the pattern memory 13 is written in parallel to the musical score data memory 26 together with the pitch data read from the selected data memory 20.

On the other hand, as shown in FIG. 3, an end code is stored next to the final note data in the note arrangement pattern area specified by each MSB. Then, as described above, when the note data of each note arrangement pattern area is extracted from the pattern memory 13 and the end code is finally read out, the pattern FINISH detection circuit 33 outputs the pattern FI
A NISH detection signal "1" pulse is output. and,
In response to this detection signal, the RS flip-flop 4 is set again via the OR circuit 6, and the reset of the counter 9 is released. Therefore, the counter 9 starts counting and issues a carry-out output CO when a predetermined count value is reached.
As a result, the latch circuit 10 has a carry out output CO.
In response to this, new numerical data is latched as the MSB, and a new note arrangement pattern area in the pattern memory 3 is specified corresponding to the outside of this numerical data.

Next, in the same manner as described above, each time the pitch data read from the selection data memory 20 is written into the musical score data memory 16, one pitch data is written in the specified note arrangement pattern area in the pattern memory 13. The address is incremented one by one, and the note data constituting each note arrangement pattern is extracted one by one and written into the musical score data memory 26 together with the pitch data extracted from the selected data memory 20. be.

On the other hand, the detection signal "1" pulse output from the pattern FINISH detection circuit 33 is supplied to the measure counter 34 as a counting input.

The bar counter 34 is reset by ΔCOMST, and is incremented by one each time the detection signal output from the pattern FINISH detection circuit 33 arrives. The measure counter 34 is configured to have pitch data of about 2 to 4 measures written in the musical score data memory 26 and to issue a carry-out output CO. Therefore, as described above, while reading pitch data from the pitch data memory, determining it, writing the data to the score data memory, and reading out each note data from the pattern memory 13, the bar counter 34
When the count value in reaches the number corresponding to the predetermined number of measures,
The FINISH data generation circuit 35 is driven by the carryout output CO of the 4-node counter 34, and predetermined FINISH data (for example, ALL "1") is output.

At the same time, the carry-out output of the measure counter 34 is also supplied to the write terminal WT of the musical score data memory 26 via the OR circuit 25, so that the F-melon ISH data generated from the F-melon ISH data generation circuit 35 is , is stored at the final address in the score data memory. Fourth
The figure is a memory map showing the status of each pitch data and note data stored in the musical score data memory.

As shown in the figure, pitch data and note data are sequentially stored in pairs at each address of the musical score data memory 26, and preset FINISH data is written at the final address. be. In this way, when the composition start switch SWI is turned on, each stored pitch data is successively read out one by one from the pitch data memory 12, and musical conditions are determined via the determination circuit 14. Ru.

Then, only when it is determined that the pitch data matches a predetermined musical condition, the read pitch data is transferred and written to the selected data memory 20, and one of these data is randomly selected. and written into the musical score data memory 26. At the same time, each note data constituting a predetermined note arrangement pattern is sequentially written from the data memory 3 and from the predetermined note arrangement pattern memory 13 to the score data memory 26 simultaneously with the pitch data. I will go.

When pitch data and note data of approximately 2 to 44 verses are written into the score data memory 26, reading of the pitch data and note data is automatically finished, and then FINISH data is written. It is generated and written into the musical score data memory 26. Next, the operation in the performance mode will be explained. In order to operate the performance mode, first turn on the performance start switch SW2. When SW2 is turned on, in response to the rise of “1”, the differentiation circuit 36 outputs a minute width “1”.
A pulse (hereinafter referred to as ΔPLYST) is output, and at the same time, in response to the rising edge of the "1" pulse, the Q output (hereinafter referred to as PLAY) of the RS flip-flop 37 is reset to "1". Further, when the Q output of the RS flip-flop 37 is set to "1", this "1" is delayed by one clock via the D flip-flop 38 and then supplied to the input side of the AND circuit 39.
As a result, the logical condition of the AND circuit 39 is satisfied after waiting for "1" of the program end signal FINISH, which will be described later. In this way, ΔPLYST and PLAY
is output, the ANDRES counter 28 outputs the OR circuit 2.
At the same time, the note length counter 40 is reset by ΔPLYST supplied through the OR circuit 41.

As mentioned above, each address of the musical score data memory 26 stores pitch data and note data as a pair, as shown in FIG. 4, and the final address stores FINISH data. remembered.

Therefore, when the address counter 28 is reset as described above, the pitch data and note data stored at the leading address are retrieved from the score data memory 26 in parallel.

The pitch data is then supplied to the musical tone forming circuit 42,
A musical tone forming process is performed, and the musical tone signal thus formed is amplified via an amplifier 43 and then outputted from a speaker 44. Further, the pitch data read from the musical score data memory 26 is also supplied to the CRT control circuit 45 and the printer control circuit 46 in parallel.

The CRT control circuit 45 performs predetermined video processing based on the pitch data output from the musical score data memory 26, and displays notes corresponding to the pitch data on the CRT 47.
For example, it is displayed using a staff score. Further, the printer control circuit 46 processes the pitch data into a predetermined print format, and similarly prints it out on staff sheet using the printer 48.
On the other hand, the note data read from the score data memory 26 is fed to the data input side of the note length comparison circuit 49.

Here, the length of each note is expressed by the number of notes, with one period of a predetermined digital clock as a reference, except for each note. The reference input side of the note length comparison circuit 49 is supplied with the counted value of the note length counter 40 . The note length counter 40' is configured to count the tempo clock TCL output from the tempo clock oscillator 50. Therefore, when it is determined in the note length comparison circuit 49 that the note data currently being output matches the count value of the note length counter 40, a coincidence signal EQ "1" is output.

At this time, the input conditions of the AND circuit 51 are PLAY “1”,
EQ becomes "1", and the address counter 28 is incremented by the output of the AND circuit 51 via the OR circuit 27. In this way, when the count value of the address counter 28 is incremented by one, the pair of pitch data and note data stored at the next address are transferred from the score data memory 26 as shown in FIG. The data is read out and repeatedly supplied to the musical tone forming circuit 42, CRT control circuit 45, and printer control circuit 46 as described above.

Therefore, when you turn on the performance start switch SW2,
A melody corresponding to the composed musical score stored in the musical score data memory 26 will flow from the speaker 44, and the musical score will be displayed on the CRT 47 in the form of a staff notation, and the printer 48 will output the musical score. is also printed out on staff paper.

While repeating the above, the FINISH data stored at the final address is transferred to the score data memory 26 as shown in FIG.
When the FNISH detection signal FmISH is detected, the FINISH detection circuit 52 is driven and the FNISH detection signal FmISH is output.

This signal F . Ru. That is, by this,
The discussion of each note data and pitch data from the musical score data memory 26 is completed. Thus, in the automatic music composer shown in this embodiment, a series of pitch data is stored in the pitch data memory 12, and the pitch data is continuously discussed to determine if it matches musical conditions one by one. , all the data matching the conditions are stored in the selection data memory 20 at once, and then one of the data matching the conditions stored in the selection data memory 20 is selected at random. Since the melody is read out and captured as one of the constituent sounds of the song, the melodies stored in the musical score data memory 26 have an extremely innovative style, making it easier for teachers to If you repeat the composition mode many times, you can get a different melody each time, without relying on a certain individuality as you would if you were composing a song. Unlike when using this method, the room for selection is not limited when obtaining a topic piece, and if this method is used, it is extremely suitable for articulation training, performance practice, etc. carried out in music education classes.

Furthermore, in this embodiment, various musical conditions can be selectively combined and used depending on the grade as the discrimination conditions in the discrimination circuit 14. degree or
Depending on the grade, etc., the musical level of the piece to be composed can be arbitrarily selected from easy to advanced. By conditioning the music to a specific style, it becomes possible to freely create melodies that are in tune with one's own preferences and have an unpredictable pitch arrangement.

Furthermore, in this embodiment, as a means for obtaining note data accompanying each pitch data, note arrangement pattern data is constructed by arranging a plurality of notes in advance in a predetermined order.
This note arrangement pattern data is stored in multiple sets of pattern memories and is configured to be read out at random, so by combining a plurality of these note arrangement patterns, the note arrangement of 2 to 4 measures can be achieved accurately. You can definitely get the pattern.

In other words, since the length of each note arrangement pattern data is set in advance to a certain time length, by reading out multiple sets of these, it is possible to reliably obtain a note arrangement pattern that ends at a bar break. . Furthermore, in this embodiment, each note data constituting each note array pattern data is represented by the number of pieces based on one period of the tempo clock TCL, and each note data is represented by the number of pieces of data constituting the note array pattern data, and each note data is expressed in correspondence with each MSB of the pattern memory. Since the arrangement pattern storage area is set and each address of this note arrangement pattern storage area is specified one address at a time with B in the pattern memory, the memory area for storing the note arrangement pattern data is as follows. of,
In addition to being able to be used extremely effectively, it is also possible to simultaneously store pitch data and note data constituting each note arrangement pattern as parallel data at one address in the score data memory 26. It becomes possible to effectively utilize the memory area within the data memory 26.

In addition, since the pitch data and the note data are stored as a pair at each address in the score data memory 26, the configuration of the automatic performance processing circuit also changes to the output of the score data memory. By dividing the line into upper and lower bits, note data and pitch data can be easily separated, and the configuration of the automatic performance circuit can be simplified. Incidentally, in the embodiment described above, conditions m to {circle around (2)} were given as conditions for the discriminating circuit 14, but various other musical conditions may be used, and they are listed as follows.

00 Black keys are prohibited for the last note.

(11) The number of black keys is allowed up to two times.

(12) The last note is not the same as the previous note.

In this way, various musical conditions can be cited, and by appropriately combining these conditions, it becomes possible to freely set or select a musical style. In addition, particularly in this embodiment, when the discriminating circuit determines whether the musical conditions match, the previously selected pitch data stored in the latch circuit 24 is used as part of the discriminating data. Therefore, when determining the difference in pitch between the previous note and the next note, this can be done extremely easily.

As is clear from the description of the embodiments above, the automatic music composer according to the present invention stores a series of pitch data in a predetermined pitch data memory, and stores these pitch data in a predetermined humility. Each time a start-up signal is supplied, the pitch data that is sequentially read out in sequence is compared with a predetermined musical condition, and the pitch data that matches the condition is stored in a predetermined selected data memory. Each time the series of data writing operations is completed, one of the plurality of pitch data stored in the selected data memo is randomly stored, and each time the series of data output operations is By repeating the reading operation, the pitch data extracted from the selected data memory is sequentially selected as one of the constituent sounds of the song, so if you use this, you will not have any musical knowledge. Even those who do not have the following skills can easily obtain a series of musically arranged pitch data, and if they add note arrangement data to the obtained pitch data, for example, as in the embodiment, It is extremely easy to obtain a track of a predetermined length.

Furthermore, as long as the pitch data can be arranged in a series while matching musical conditions, it is relatively easy to add note arrangement pattern data to this to compose a piece of music. As shown in the example, it becomes possible to easily compose a piece of music manually without automatically composing even the note arrangement pattern data. That is, by printing out pitch data arranged in a series to match musical conditions and adding note arrangements to this, it is possible to compose a piece of music. Therefore, in the field of music education, when performing performance practice or listening training, there is no need to entrust the composition of the required piece to the teacher to be used for this purpose, and there is also less room for selection of the required piece. This eliminates problems such as the spread of musical compositions and the fact that the compositions are biased toward a particular style, and the educational effects of this type of music education can be significantly improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the electrical configuration of an automatic music composer according to the present invention, FIG. 2 is a memory map in the pitch data memory shown in FIG. 1, and FIG. 4 is a memory map showing the data structure in the musical score data memory shown in FIG. 1. FIG. 4 is a memory map showing the data structure in the musical score data memory shown in FIG. 11...Random signal generator, 12...
・Pitch data memory, 14...condition discrimination circuit, 15...
...Gate circuit, 16...Condition data memory,
17...Grade selection switch, 20...
...Selection data memory, 24...Latch circuit,
26... Musical score data memory. Figure 2 Figure 3 Ship Figure 4

Claims (1)

[Scope of Claims] 1. A pitch data memory storing a series of pitch data, and a pitch data memory that sequentially and continuously reads the series of pitch data from the pitch data memory each time a predetermined read start signal is supplied. a first reading control means for reading the pitch data sequentially from the pitch data memory, and a music condition determining means for comparing the pitch data sequentially read from the pitch data memory with a predetermined musical condition and selecting pitch data that matches the condition. , a selection data memory for sequentially storing pitch data selected by the music condition determining means; and a plurality of pitch data stored in the selection data memory each time a series of data read operations by the first read control means are completed. a second reading control means for randomly reading one of the pitch data of the selected data memory, and configured to sequentially select the pitch data read from the selected data memory as one of the constituent sounds of the song. An automatic music composer featuring: 2. The music condition determining means has a temporary storage memory for temporarily storing pitch data read from the selected data memory and adopted as constituent notes of a song, and when compared with the music conditions. 2. The automatic music composition machine according to claim 1, wherein the pitch data which is stored in the temporary storage memory and which has been selected immediately before is used as the comparative judgment material.