JP3489290B2 - Automatic composer - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic music composition apparatus for automatically creating new music information based on inputted music information.

[0002]

2. Description of the Related Art In recent years, with the spread of personal computers, anyone can freely play music by using computer music to play musical instruments, compose, arrange, and synthesize tones. I'm starting to enjoy it. Particularly, in the field of music composition using a computer, music composition can be easily composed by simply inputting and setting various music conditions in accordance with instructions of the computer, even if there is no specialized knowledge of music. In addition, recently, the melody of the original song is classified into harmony and nonharmonic sounds, and the features of the melody are analyzed by further classifying the nonharmonic sounds, and a new melody is synthesized according to the analysis result and chord progression. However, there is a proposal to automatically compose music.

[0003]

However, an automatic music composition apparatus that analyzes the characteristics of the melody of the original song and automatically composes the music based on the analysis result is used for chord progression after changing the input melody. It was just to edit according to the above, and it did not change the musical composition or rhythm of the original melody.

The present invention has been made in view of the above points, and provides an automatic music composition apparatus that can generate a variety of music different from the melody while making the best use of the characteristics of the input melody. It is provided.

[0005]

Means for Solving the Problems An automatic music composition apparatus according to a first aspect of the invention is a performance information supply means for supplying performance information composed of a plurality of note data and a predetermined condition among the performance information. Means for extracting note data to be reproduced, pattern storage means for storing a plurality of performance information patterns, and sound of the note data extracted by the extracting means.
Feature data representing a comfort feature is generated, and the generated feature
A pattern selection means for selecting a performance information pattern from the pattern storage means according to the data and a music generation means for generating a new music piece according to the performance information pattern selected by the pattern selection means are provided.
The note data is composed of pitch and sounding timing. The extracting means extracts only the note data that matches a predetermined condition from the performance information composed of such note data. For example, since the rhythm pattern of the performance information can be recognized based on the sounding timing of the note data that constitutes the performance information, a strong beat sound is generated for each half bar (half of one bar) based on the rhythm pattern. The corresponding note data is extracted, or the note data showing a predetermined pitch change is extracted from the state of the pitch change of the note data. The selecting means generates characteristic data representing musical characteristics of the note data extracted by the extracting means , and the characteristic data is generated.
A corresponding performance information pattern is appropriately selected from the pattern storage means according to the characteristic data . For example, the selecting means selects a performance information pattern in accordance with the pitch difference between adjacent note data pieces extracted or the number of note data pieces extracted. The music generation means generates a new music based on the selected performance information pattern. Therefore, according to the first aspect of the present invention, it is possible to newly generate a song that is rich in variety and that is different from the melody while making the most of the characteristics of the input melody.

[0006] automatic composition apparatus according to a second aspect of the present invention is to divide the performance information supply means for supplying performance information including a plurality of note data, the performance information in a predetermined period, the compartment
The pitch or pronunciation of the note data included between
Change type indicating the change of the group and the parameter corresponding to the change type
Parameter setting means for setting the data for each section, and at least one of the pitch and sounding timing constituting the note data for each of the predetermined sections , corresponding to the section.
Performance information changing means for changing the supplied performance information by changing in accordance with the specified change type and the parameter, and music generation for generating a new music piece according to the performance information changed by the performance information changing means. And means. The parameter setting means divides the performance information into a predetermined section, for example, an appropriate section such as every one bar or every two bars, and each section is included in the section.
Indicates a change in pitch or pronunciation timing of note data
A change type and a parameter corresponding to the change type are set. The change type and the parameters corresponding thereto include, for example, one that reverses a rhythm within a section, one that changes a predetermined rhythm within the section to another rhythm, one that reverses a pitch within the section, and a section Among them, there is one in which the pitch is reversed with reference to a predetermined pitch. The performance information changing means is the type of change set for each section and
Changing the performance information itself by changing the like pitch and tone generation timing constituting the musical note data at every predetermined interval in accordance with the parameters. The music generation means generates a new music based on the changed performance information. Therefore, according to the second aspect of the present invention, it is possible to newly generate a musical piece rich in variety different from the melody while making the most of the characteristics of the input melody.

The automatic music composition apparatus according to the third invention supplies performance information supply means for supplying performance information composed of a plurality of note data, and lyrics data composed of syllable data corresponding to the lyrics to be composed. Data supplying means, dividing means for dividing the performance information and the lyrics data into predetermined sections in association with each other, and data for adding or deleting the note data in accordance with the number of the syllable data for each divided section. The change means and the music generation means for generating a new music according to the performance information changed by the data change means are provided. The lyrics to be composed are created by the user and are a set of characters.
In particular, Japanese hiragana and katakana are syllable characters in which one character forms one syllable. Therefore, the data supplying means supplies one piece of character when the lyrics to be composed are expressed in hiragana or katakana as syllable data, and supplies the lyrics data which is a collection of the syllable data. The performance information supply means supplies performance information composed of a plurality of note data. The dividing means divides the lyrics data, which is a collection of syllable data, and the performance information, which is a collection of note data, into predetermined sections in association with each other. However, in the divided sections, the number of syllable data and the number of note data may not match. Therefore, the data changing means changes the performance information by adding or deleting the number of note data so that the number of note data is equal to the number of syllable data for each divided section. The music generation means generates a new music based on the changed performance information. Therefore, according to the third aspect of the present invention, it is possible to newly generate a tune that is rich in variation and that matches the input lyrics while making the most of the characteristics of the input melody.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 2 is a hardware block diagram showing the configuration of an electronic musical instrument incorporating the automatic music composition apparatus according to the present invention. The electronic musical instrument is controlled by a microcomputer including a microprocessor unit (CPU) 1, a program memory 2 and a working memory 3. The CPU 1 controls the operation of the entire electronic musical instrument. For this CPU 1, a program memory 2, a working memory 3, a pattern memory 4, a performance data memory 5, a key press detection circuit 6, a performance data input device 7, a switch detection circuit 8 and a display circuit via a data and address bus 1E. 9 and the tone generator circuit 10 are connected to each other.

The program memory 2 stores various programs of the CPU 1, various data, various symbol characters, etc., and is constituted by a read only memory (ROM). The working memory 3 is for temporarily storing performance information and various data generated when the CPU 1 executes a program, and is a random access memory (RAM).
The respective predetermined address areas are assigned and used as registers and flags. The pattern bank memory 4 stores a plurality of melody patterns corresponding to pitch difference data and rhythm pattern data. To read the melody pattern from the pattern bank memory 4, an address corresponding to the pitch difference data or the rhythm pattern data may be given. The performance data memory 5 stores performance data regarding the composed melody.

The keyboard 1A is provided with a plurality of keys for selecting the pitch of a musical tone to be generated, has a key switch corresponding to each key, and detects the key depression speed as necessary. It has a touch detection means such as a device and a pressing force detection device. The key depression detection circuit 6 is configured to include a circuit composed of a plurality of key switches provided corresponding to each key of the keyboard 1A that specifies the pitch of a musical tone to be generated, and a new key is depressed. Output a key-on event,
When the key is newly released, a key-off event is output. In addition, a process of generating touch data is performed by determining the key pressing operation speed or the pressing force when the key is pressed, and the generated touch data is output as velocity data. As described above, the data such as the key-on event, the key-off event, and the velocity are expressed by the MIDI standard, and also include the data indicating the key code and the assigned channel. The performance data input device 7 is for inputting performance data relating to an original music piece, and is composed of a MIDI interface when the performance data is data complying with the MIDI standard.

Numeric keypad & keyboard & various switches 1B
Is an analysis switch for analyzing and extracting the musical characteristics of existing songs, a composition switch for automatically composing music based on the analysis and extraction results, a numeric keypad for entering numerical data and a keyboard for entering character data, It is configured to include various controls such as automatic performance start / stop switches. It should be noted that, in addition to these, various operators for selecting, setting, and controlling pitch, tone color, effect, etc. are included, but since the details thereof are publicly known, description thereof will be omitted. Switch detection circuit 8
Detects the operation state of each operation element of the numeric keypad & keyboard & various switches 1B and outputs switch information according to the operation state to the CPU 1 via the data and address bus 1E. The display circuit 9 displays various information such as the control state of the CPU 1 and the contents of setting data on the display 1C. The display 1C is a liquid crystal display panel (LC
D), CRT, etc., and its display operation is controlled by the display circuit 9. GUI (Graphical User Inte) by this numeric keypad & keyboard & various switches 1B and display
rface) is configured.

The tone generator circuit 10 is capable of simultaneously generating musical tone signals on a plurality of channels, inputs data and performance information (data conforming to the MIDI standard) given via the address bus 1E, and receives this data. A tone signal is generated based on. Any tone signal generation method may be used in the tone generator circuit 10. For example, a memory reading method for sequentially reading the musical tone waveform sample value data stored in the waveform memory according to the address data that changes corresponding to the pitch of the musical tone to be generated, or a predetermined frequency with the address data as phase angle parameter data. A well-known method such as an FM method for performing a modulation operation to obtain musical tone waveform sample value data or an AM method for performing a predetermined amplitude modulation operation using the address data as phase angle parameter data to obtain a tone waveform sample value data. You may employ suitably. The tone signal generated from the tone generator circuit 10 is a sound system 1 including an amplifier and a speaker (not shown).
Pronounced via D.

Next, an example of the operation of the automatic music composition apparatus according to the present invention will be described. FIG. 3 is a diagram showing a main flow of the electronic musical instrument of FIG. The main flow is sequentially executed by the following steps. Step 31: First, initial setting processing is performed, and initial values are set in the registers and flags in the working memory 3 shown in FIG. Step 32: Numeric keypad & keyboard & various switches 1
It is determined whether or not the composition switch on B has been turned on. If there is an on operation (YES), the process proceeds to step 33, and if there is no on operation (NO), the process proceeds to step 34.

Step 33: Since it is determined in step 32 that the composition switch has been turned on (YES), any one of the automatic music processings 1 to 6 is executed in response to the on operation. To do. That is, in this embodiment, the automatic music composition device has a total of six types of automatic music composition modes, and one of the modes is used to perform automatic composition. This mode may be set in advance with another switch, or six composition switches are provided, and the automatic music composition processing 1 to
Any one of 6 may be executed. Details of these automatic music pieces processing 1 to 6 will be described later.

Step 34: It is judged whether the automatic performance start / switch on the numeric keypad & keyboard & various switches 1B is turned on, and there is an on operation (YES).
In case of NO, the process proceeds to step 35. In case of no ON operation (NO), the process proceeds to step 36. Step 35: The automatic performance start / stop processing corresponding to the ON event of the automatic performance start / stop switch is performed. That is, when the on event is the automatic performance start event, the performance data is read from the performance data memory 5 and the automatic performance processing is performed based on the performance data. If the on event is a stop event, the currently sounding sound is muted and the automatic performance process is stopped. Step 36: Other processing, for example, sounding processing according to the operation of the keyboard 1A is performed.

FIG. 1 is a diagram showing the details of the automatic music composition processing 1 in step 33 of FIG. Hereinafter, the process from the acquisition of the original melody to the composition of a new melody by the automatic music processing 1 will be described step by step using the musical score of FIG. Step 11: Take in the melody of the existing song. For example,
The melody of an existing song is loaded from the GUI (numerical keypad & keyboard & various switches 1B and display 1C), keyboard 1A, etc. in a form including bar line data. Existing song is MID
In the case of I data, the user may later input bar line data to the imported MIDI data using the GUI. Further, the printed musical score or the like may be read by a scanner. Here, it is assumed that the melody as shown in FIG. Step 12: Input the key data indicating the key of the existing song captured in the step 11. In the case of FIG. 4, C major is input as the key data.

Step 13: Extract the strong beat sound of each section from the fetched melody and use it as the extracted performance data. That is, the imported melody is 2 /
2-beat system such as 2,2 / 4,2 / 8 or 3 / 2,3 / 4,
In the case of 3-beat system such as 3/8, 4/2 for each bar
In the case of 4-beat system such as 4/4 (C) and 4/8, a strong beat sound is extracted for each half measure (half of one measure). Figure 4 (a)
In the case of the melody, since the time signature is 4/4 (C), note data of the first beat and the third beat is extracted as a strong beat. FIG. 4B shows the extracted extracted performance data.

Step 14: Generate pitch difference data between note data of adjacent strong beat sounds in each section. Here, the pitch difference data indicates how many pitches of both note data are separated by a semitone. For example, in the case of the extracted performance data of FIG. 4B, the pitch difference data is "3 semitones-2 semitones-4 semitones-2 semitones-2 semitones-3 semitones-4 semitones". Note that this pitch difference data simply indicates only the pitch difference, but "3 semitones up-2 semitones down-4 semitones up-2 semitones up-2 semitones down-3 semitones up-4 semitones up" Thus, it may indicate whether the pitch of the next note data is high or low.

Step 15: For each section, a melody pattern is selected from the pattern bank memory 4 in accordance with the pitch difference data generated in Step 14. That is, the pitch difference data is used as the upper address, and a combination of the random lower address is used as the read address of the pattern bank memory 4, and the melody pattern is read therefrom. FIG. 4C shows an example of the melody pattern read from the pattern bank memory 4 based on the pitch difference data of FIG. 4B. In addition,
Since the pitch difference data cannot be generated for the last half measure, the note data of FIG. 4 (a) is adopted as it is. That is, the note data “E” and 4 of the last half measure
The minute rests are directly adopted as the melody pattern of the last half measure in FIG. 4 (c). Step 16: Since the melody pattern read out in step 15 is standardized so that the first note becomes the tonic of the key, the first pitch and the extracted performance of the selected melody pattern for each section. The pitch of the melody pattern is shifted so that the pitch of the data becomes equal.
In FIG. 4 (d), the first eighth note data “C” of the first semi-bar is set so as to be the pitch of the half note data “E” of the extracted performance data of FIG. 4 (b). Is shifted to.
Thereafter, the pitch of each note data of each half measure is similarly shifted.

Step 17: Convert the pitch of the note data into the scale note of the key input in Step 12. That is,
By lowering the pitch of the note data with “#” or “♭” from the note data of FIG.
It is converted into a C major scale tone as shown in (e). In addition,
The pitch of the note data with “#” or “♭” may be increased by a semitone to convert it into a C major scale tone as shown in FIG. Step 18: Restore bar lines. For example, FIG. 4 (c)
And, the bar line for each half bar added for data processing as shown in FIG.
Restore the bar line corresponding to the time signature. By the series of processes described above, the melody as shown in FIG.
A completely new melody like (e) is composed.

In the automatic music processing 1, a case has been described in which strong beat sounds in each section are extracted to create extracted performance data. However, the present invention is not limited to this, and note data is acquired from a melody imported as follows. May be extracted to create the extracted performance data. For example, a predetermined section (1
For each measure, find the case where the pitch of the note data is proceeding at the same pitch based on the state of the change of the pitch of the note data, and extract the appropriate one from the corresponding note data. Alternatively, it is possible to search for a case where the pitch is progressing in order according to the scale note of the input note, and to extract an appropriate one from the note data corresponding to it, It is also possible to search for a case where one or more notes are jumping and advancing, and to extract an appropriate note data from the note data corresponding thereto. Alternatively, a set of note data that matches a predetermined rhythm pattern may be extracted for each section, and an appropriate one may be extracted from the set.

FIG. 5 is a diagram showing the details of the automatic music composition processing 2 in step 33 of FIG. This automatic music processing 2 is a modification of the automatic music processing 1. The difference is that, in addition to the melody acquisition, lyrics data is acquired and a sounding timing pattern, that is, a rhythm pattern is generated on the basis of the number of syllable data constituting the lyrics data. The point is that the melody pattern is selected and the melody pattern is selected based on the rhythm pattern and the pitch difference data. Hereinafter, the process from the original melody to the composition of a new melody by the automatic music composition processing 2 will be described step by step using the musical score of FIG.
The same processing as the automatic music composition processing 1 will be simplified and described.

Step 51: Take in the melody of the existing song. Here, it is assumed that the melody shown in the musical score of FIG. This melody is shown in Figure 4.
Same as (a). Step 52: Input key data. Also in the case of FIG. 6A, C major is input. Step 53: Capture the lyrics data delimited by the bar line data. As shown in FIG. 6 (b), this lyric data is syllable data in which the first measure has six syllables “A, I, U, E, O, or”.
The second measure is 5 syllable data "Ki, Ku, Ke,"
Ko, Sa ”, and the third measure is 8 syllable data“ Shi,
Sue, se, so, ta, chi, tsu, te ”, 4th measure is 4
It consists of one syllable data “and na, ni, n”. Step 54: For each bar, a sounding timing pattern corresponding to the number of syllables in the bar, that is, the number of syllable data, is read out randomly from the pattern bank memory 4. For example, the number of syllable data of the first measure is "6", the number of syllable data of the second measure is "5", the number of syllable data of the third measure is "8", and the number of syllable data of the fourth measure is "4". A sounding timing pattern corresponding to the number of syllable data as shown in FIG. 6C is selected.

Step 55: Extract the strong beat sound of each section of the taken melody and use it as the extracted performance data. Since the melody of FIG. 6A is the same as that of FIG. 4A, the extracted performance data is as shown in FIG.
Same as (b). Step 56: Generate pitch difference data between note data of adjacent strong beats in each section. Also in the case of the extracted performance data of FIG. 6D, the same pitch difference data “3 semitones-2 semitones-4 semitones-2 semitones-2 semitones-3 semitones-as in the case of FIG. 4B.
4 semitones ”are generated.

Step 57: For each section, a melody pattern is selected from the pattern bank memory 4 in accordance with the tone generation timing pattern selected in step 54 and the pitch difference data generated in step 56. That is, the pitch difference data is used as the upper address, and a combination of the random lower address is used as the read address of the pattern bank memory 4, and the melody pattern is read therefrom. When the rhythm pattern of the read melody pattern matches the sounding timing pattern, it is set as the melody pattern, and when the rhythm pattern does not match, the lower address is changed until a further matching melody pattern is read. FIG. 6 (e) shows FIG.
The case where the melody pattern that matches the sounding timing pattern of (c) is read from the pattern bank memory 4 is shown. For convenience of explanation, the case where the melody pattern of FIG. 6 (e) and the melody pattern of FIG. 4 (c) are the same is shown, but the melody pattern is different even if the pitch difference data and the rhythm pattern are actually the same. There are more things. The pattern bank memory 4 stores a melody pattern in accordance with the pitch difference data and the rhythm pattern, and a combination of the pitch difference data and the rhythm pattern is used as an upper address, and a random lower address is combined and read out. May be.

Step 58: Since the melody pattern read out in the step 15 is standardized so that the first note becomes the tonic of the key, the first pitch of the selected melody pattern for each section. And the pitch of the melody pattern is shifted so that the pitch of the extracted performance data becomes equal. As a result, the musical score of FIG.
It becomes like (f).

Step 59: Convert the pitch of the note data into an input scale. That is, the pitch of the note data with "#" or "♭" is lowered by a semitone from the note data of FIG. 6 (f). On the contrary, the pitch of the note data with “#” or “♭” may be increased by a semitone. Step 5A: Restore the bar line. As a result, FIG.
An entirely new melody as shown in FIG. 6G is composed from the melody as shown in FIG.

FIG. 7 is a diagram showing the details of the automatic music composition process 3 in step 33 of FIG. This automatic music processing 3 is different from the automatic music processing 1 by dividing the captured melody into predetermined sections and setting how to change the melody data for each section by setting the change type and parameters. The point is that the loaded melody data is changed and new melody data is composed. Hereinafter, the process from the original melody to the composition of a new melody by the automatic music composition process 3 will be described step by step using the musical score of FIG. The same processing as the automatic music composition processing 1 will be simplified and described.

Step 71: Take in the melody of the existing song. In this automatic music composition processing 3, a melody as shown in the musical score of FIG. Step 72: Input key data. In the case of FIG. 8A, C major is input. Step 73: Using the GUI (numeric keypad & keyboard & various switches 1B and display 1C) and the like, divide the captured melody into predetermined sections, and change types that indicate what kind of change is to be made for each section, and the change Set the parameter corresponding to the type.

As the change type, for example, a first rhythm change type in which a rhythm is reversely advanced in a section (rhythm reverse progression), and a second rhythm change type in which a predetermined rhythm in the section is changed to another rhythm (Rhythm replacement), a first pitch change type (pitch reverse progression) for reversing the pitch within the section, and a second pitch change type (pitch reversal) for reversing the pitch within the section based on a predetermined pitch There is.
Other than this, various rhythm change types and pitch change types are conceivable, but description thereof will be omitted here. In the case of the first rhythm change type, the pitch of the starting point for reverse progression is set, and in the case of the second rhythm change type, the rhythm before the change and the rhythm after the change are set in the first pitch change type. In some cases, the pitch of the starting point for reverse traveling is
In the case of the pitch change type, it is necessary to set the reference pitch, which is the reference of inversion, as a parameter.
In this embodiment, the rhythm change type and the pitch change type can be set simultaneously for one section, but two rhythm change types and pitch change types cannot be set at the same time. When two or more rhythm change types and pitch change types are set for one section, the AND or OR of the melody patterns generated by these changing processes may be adopted as new melody data. It goes without saying that it is good. In this case, if two note data occur at the same sounding timing, one of them is given priority, or if the sounding timing is very short, a minimum note length is set and a note length less than this is set. Should be deleted.

Step 74: For each section, the melody data in that section is changed according to the set change type and parameters. An example of the set change type and parameters is shown in FIG. Here, the captured melody is divided into two bars, the first bar and the second bar are set as section 1, and the third bar and the fourth bar are set as section 2. And
In the section 1, the first rhythm change type (rhythm reverse progression) and the starting point pitch “E” are set as its parameters. The second pitch change type (pitch inversion) in section 2
And the reference pitch "G" is set as the parameter. Therefore, the melody data of FIG. 8C is changed as shown in FIG. 8D according to the change type and the parameter.

That is, in section 1 of FIG. 8C, the rhythm pattern is an eighth rest-8th note (C) -8th note (C) -8th note (D) -8th note (E). -8th note (G) -4th note (E) -8th rest -8th note (D)-
8th note (D) -8th note (E) -8th note (F) -8
Since it is a quarter note (A) -fourth note (F), after the rhythm reverse proceeding with reference to the starting pitch "E", the rhythm reverses from the end of the section 1 toward the beginning, as shown in FIG. As in section 1, quarter note (E) -8th note (E) -8th note (F
#)-8th note (G #)-8th note (B) -8th note (G #)-8th rest-4th note (F #)-8th note (F
#)-8th note (G #)-8th note (A) -8th note (C #)-8th note (A) -8th rest. Figure 8
In section 2 of (b), the pitch pattern is a quarter rest −16.
Minute Note (G) -16th Note (A) -16th Note (G)-
16th note (F) -8th note (E) -4th note (G)-
8th note (C) -2nd note (B), but after pitch inversion centered on the reference pitch "G", as shown in section 2 of FIG. 8D, a quarter rest-16th note ( G) -16th note (F)
-16th note (G) -16th note (A) -8th note (B
♭) -4th note (G) -8th note (D) -2nd note (E
♭)

Step 75: The pitch of each note data constituting the melody data changed in the step 74 is converted into an input scale. That is, by lowering the pitch of the note data with "#" or "♭" from the note data of FIG. 8 (d) by a semitone, a melody as shown in FIG. 8 (d) is obtained. On the contrary, the pitch of the note data with “#” or “♭” may be increased by a semitone. Step 76: Restore bar lines. As a result, FIG.
A completely new melody as shown in FIG. 8F is composed from the melody as shown in FIG. In addition, FIG.
In the reversal processing from FIG. 8C to FIG. 8D, when the pitch is higher than the reference pitch “G”, the pitch is lower than the reference pitch “G” by the pitch, and conversely when it is low. Although the case where the pitch is increased by the pitch has been described, the invention is not limited to this, and the pitch inversion degree may be set. For example, the pitch may be inverted to be several times or a fraction of the pitch with respect to the reference pitch, or the degree may be different depending on whether the pitch is high or low.

FIG. 9 is a diagram showing the details of the automatic music composition processing 4 in step 33 of FIG. This automatic music processing 4 is a modification of the automatic music processing 3 of FIG. 7. The difference is that, apart from the acquisition of the melody, the lyrics data is acquired and the melody after the change is made based on the number of syllable data constituting the lyrics data. Is the point that has changed. Hereinafter, the process from the original melody to the composition of a new melody by the automatic music composition processing 4 will be described step by step using the musical score of FIG. It should be noted that the same processing as the automatic music composition processing 3 of FIG. 7 will be briefly described.

Step 91: Take in the melody of the existing song. Here, a melody as shown in the musical score of FIG. This melody is the same as in FIG. Step 92: Input key data. In the case of FIG. 10A, C major is input. Step 93: Take in the lyrics data delimited by bar line data as shown in FIG. First of this lyrics data
The measure has four syllable data “I, Ro, Hani”, the second measure has four syllable data “Ho, He, Tochi”, and the third measure has 8
Syllable data "Ri, Nu, Ru, Oh, Wa, Ka, Yo, Ta",
The fourth measure is composed of two syllable data "re, so". Step 94: The melody captured in step 91 is divided into predetermined sections, and a change type indicating what kind of change is to be made for each section and a parameter corresponding to the change type are set.

Step 95: For each section, the melody data in that section is changed according to the set change type and parameters. FIG. 10 shows an example of change types and parameters.
It is shown in (b). Change types and parameters are shown in Fig. 8.
Same as (b). Therefore, the melody data shown in FIG.
It is changed as shown in FIG. 10D, which is the same as FIG.

Step 96: The pitch of each note data forming the melody data changed in the step 95 is converted into an input scale. That is, by lowering the pitch of the note data with “#” or “♭” from the note data of FIG. 10 (d) by a semitone, a melody as shown in FIG. 10 (e) is obtained. FIG. 10 (e) melody and FIG.
The melody of (e) is the same. Step 97: Restore bar lines. As a result, FIG.
The musical score of 0 (e) is as shown in FIG. 10 (f). The melody changed by the processing so far is the same as in the case of FIG.

Step 98: Match the number of note data with the number of syllable data for each measure. That is, note data is added or deleted so that the number of syllable data assigned to each measure is equal to the number of note data. For example,
In the first measure, the syllable data number of the lyrics data of FIG. 10 (h) is “4”, and the note data number of FIG. 10 (f) is “6”, so two note data are deleted. Second
In the measure, since the number of syllable data of the lyrics data of FIG. 10 (h) is “4” and the number of musical note data of FIG. 10 (f) is “6”, two musical note data are deleted. In the third measure, the number of syllable data of the lyrics data of FIG. 10 (h) is “8” and the number of note data of FIG. 10 (f) is “7”, so one note data is added. In the fourth measure, the number of syllable data of the lyrics data of FIG. 10 (h) is “2” and the number of musical note data of FIG. 10 (f) is “1”, so one musical note data is added. As a result, the melody shown in FIG.
A completely new melody like (g) is composed.

FIG. 11 is a diagram showing details of the automatic music composition processing 5 in step 33 of FIG. This automatic music processing 5 is shown in FIG.
This is a modified example of the automatic tune processing 3 of the above, and a different point is that a predetermined parameter is extracted for each predetermined section, the extracted parameter is changed, and the melody data is changed based on the changed parameter. is there. Hereinafter, the process from the original melody to the composition of a new melody by the automatic music composition processing 5 will be described step by step using the musical score of FIG. Note that the same processing as the automatic music composition processing 3 will be simplified and described.

Step 111: Take in the melody of the existing song. Here, a melody as shown in the musical score of FIG. This melody is the same as in FIG. Step 112: Input key data. In the case of FIG. 12A, C major is input. Step 113: The melody captured in step 111 is divided into predetermined sections, and a predetermined parameter is extracted for each section. As the parameter to be extracted, for example, the state of the change of the pitch of the note data in the interval is a parameter, that is, the number of times the pitch of the note data progresses at the same pitch (homogeneous progression), in order according to the scale of the input note. It is the number of times the pitch progresses (sequential progress) and the number of times the sound jumps in the input tone by one or more tones (jump progress). In addition to this, various rhythms and pitches can be considered, but description thereof will be omitted here. here,
The captured melody is divided into two bars, the first bar and the second bar are set as section 1, and the third bar and the fourth bar are set as section 2. The result of parameter extraction for each section is shown in FIG.
As shown in FIG. 2 (c), in the section 1, the jump progress is 4 times, the sequential progress is 4 times, and the same sound progress is 2 times.
1 time, 5 times in sequence, 0 times in same sound. The illustration is omitted when the same sound progresses 0 times.

Step 114: Change a part of the parameters of each extracted section. For example, four jumps in section 1 are three times, four steps in sequence are five times, and section two is two.
2 jumps in 3 times, 5 jumps in 4 times
Change each time. Therefore, the changed parameters are as shown in FIG. Step 115: For each section, change the melody data in that section according to the changed parameters. Therefore, in the section 1, the jumping progress from the eighth note (G) to the quarter note (E) of FIG.
The progress is changed from the eighth note (G) in (e) to the quarter note (A) in sequence, and in section 2, 16 in FIG.
The sequential progression from the 1st note (G) to the 16th note (A) is changed to the jumping progression from the 16th note (G) to the 16th note (D) in FIG.

Step 116: The pitch of each note data forming the melody data changed in the above step 115 is converted into an input scale. This process is not performed in the case of FIG. Step 117: Restore the bar line. As a result, a melody as shown in FIG. 12 (f) is composed from the melody as shown in FIG. 12 (a). In the melody data changing process in step 115, the case where a part of the note data is changed has been described. However, since the changed melody and the original melody are almost the same in this case, the changed parameter All the contents of the note data in FIG. 12B may be changed so as to match the contents. In this way, in the section 1, the jump progress is 3 times, the sequential progress is 5 times, the same sound progress is 2 times, and in the section 2, the jump progress is 3 times and the sequential progress is 4 times.
It is possible to obtain a completely new melody as shown in FIG.

FIG. 13 is a diagram showing the details of the automatic music composition process 6 in step 33 of FIG. This automatic music processing 6 is shown in FIG.
This is a modified example of the automatic tune processing 5 of No. 1, except that melody data is captured separately from the melody data, and the melody after the modification is further modified based on the number of syllable data that compose the lyrics data. Is. Hereinafter, the process from the original melody to the composition of a new melody by the automatic music composition processing 6 will be described step by step using the musical score of FIG. Note that the same processing as the automatic music composition processing 5 in FIG. 11 will be described in a simplified manner.

Step 131: Take in the melody of the existing song. Here, the melody as shown in the musical score of FIG. This melody is the same as in FIG. Step 132: Input key data. In the case of FIG. 14A, C major is input. Step 133: Take in the lyrics data delimited by the bar line data as shown in FIG. The first measure of this lyrics data is 7 syllable data "I, Ro, Ha, Ni, Ho, He,
And, the second measure is 7 syllable data "Chi, Ri, Nu, Ru,
"O, wa, ka", the third measure is 6 syllable data "Yo, Ta,
"So, Tsu, Ne", 4th measure is one syllable data "U"
Each is composed of Step 134: The melody captured in step 131 is divided into predetermined sections, and predetermined parameters are extracted for each section. The parameters to be extracted are the same as in the case of the automatic music composition processing 5 and indicate the state of the pitch change of the note data in the section. Therefore, the extracted parameters are as shown in FIG. 14C, which is the same as FIG. 12C.

Step 135: Change a part of the extracted parameters of each section. As in the case of FIG. 12,
4 jumps in section 1 to 3 jumps, 4 jumps in sequence
5 times, 2 jumps in section 2 to 3 times,
Change 5 times in sequence to 4 times respectively. The changed parameters are as shown in FIG. 14 (d), which is the same as FIG. 12 (d). Step 136: For each section, change the melody data in that section according to the changed parameters. An example of the changed parameters is the same as in FIG.
It becomes like (d). Needless to say, all the contents of the note data in FIG. 14B may be changed so as to match the contents of the changed parameters as described above.

Step 137: The pitch of each note data constituting the melody data changed in the above step 115 is converted into an input scale. This process is not performed in the case of FIG. Step 138: Restore the bar line. As a result, the musical score of FIG. 14E becomes as shown in FIG. 14F. Step 139: The number of note data is adjusted to the number of syllable data for each measure. That is, note data is added or deleted so that the number of syllable data assigned to each measure is equal to the number of note data. For example, in the first measure, the number of syllable data of the lyrics data of FIG. 14 (h) is seven, that is, “i, ro, ha, ni, ho, ha, to”.
Since the number of note data in (f) is 6, one note data is added. In the second measure, the number of syllable data of the lyrics data of FIG. 14 (h) is “Chi, Ri, Nu, Ru, Oh, wa,
14 ”, and the number of note data in FIG. 14 (f) is 6, one note data is added. In the third measure, the number of syllable data of the lyrics data of FIG. 14 (h) is six, “yo, ta, lea, so, tsu, ne”.
Since the number of note data in (f) is 7, one note data is deleted. In the fourth measure, the number of syllable data of the lyrics data of FIG. 14 (h) is one “u”,
Since the number of note data in (f) is also one, it is not changed as it is. The note data to be added or deleted is randomly selected. As a result, a completely new melody as shown in FIG. 14 (g) is composed from the melody as shown in FIG. 14 (a). In the automatic music processing 5 or 6, the case where the number of parameters in the predetermined section is the same as the number of parameters after the change has been described. The data may be reduced or increased.

In the above-described automatic music composition processing 1, the case of extracting note data that becomes a strong beat sound for each section has been described, but the same note progression, sequential progression or jumping as performed in the automatic music composition processing 5 has been described. You may make it extract the note data applicable to progress. In this case, when there are a plurality of corresponding note data (in the case of jumping progress 4 times, sequential progress 4 times, same sound progress 2 times as shown in FIG. 12C), the earliest one , The last one, the lowest pitch, the highest pitch, and so on. In the above embodiment, Japanese has been described as an example of syllable data, but it goes without saying that the same can be applied to the case of other languages. In this case, single vowels, double vowels,
Syllable data may be determined based on phonetic symbols such as plosives, nasal sounds, sidetones, fricatives, and siphons, and composition may be performed based on the determined syllable data. Further, in the above-mentioned embodiment, the case where one melody is composed has been described, but a plurality of melody may be generated at the same time. For example, different melodies as shown in FIG. 12E and FIG. 12G may be generated at the same time. Furthermore, the note data to be deleted or added may be randomly selected from the melody data, or may be deleted in a form that reduces homophonic progression, sequential progression or jump progression,
You may add in the form which increases homophony progress, a sequential progress, or a jump progress.

[0048]

According to the present invention, while utilizing the characteristics of the inputted melody, it is possible to newly compose a musical piece rich in variety different from the melody.

[Brief description of drawings]

FIG. 1 is a diagram showing details of automatic music composition processing 1 in step 33 of FIG.

FIG. 2 is a hardware block diagram showing a configuration of an electronic musical instrument incorporating the automatic musical composition apparatus according to the present invention.

FIG. 3 is a diagram showing a main flow of the electronic musical instrument of FIG.

FIG. 4 is a diagram showing a process from the original melody to the composition of a new melody by the automatic music composition processing 1 of FIG.

FIG. 5 is a diagram showing details of the automatic music composition processing 2 in step 33 of FIG.

6 is a diagram showing a process from the original melody to a new melody composition by the automatic music composition processing 2 of FIG. 5;

FIG. 7 is a diagram showing details of the automatic music composition processing 3 in step 33 of FIG.

FIG. 8 is a diagram showing a process in which a new melody is composed from an original melody by the automatic music composition processing 3 of FIG. 7;

FIG. 9 is a diagram showing details of the automatic music composition processing 4 in step 33 of FIG.

FIG. 10 is a diagram showing a process from the original melody to a new melody composition by the automatic music composition processing 4 of FIG. 9;

11 is a diagram showing details of the automatic music composition processing 5 in step 33 of FIG.

FIG. 12 is a diagram showing a process from the original melody to a new melody composition by the automatic music composition processing 5 of FIG.

FIG. 13 is a diagram showing details of the automatic music composition processing 6 in step 33 of FIG.

FIG. 14 is a diagram showing a process from the original melody to the composition of a new melody by the automatic music composition processing 6 of FIG.

[Explanation of symbols]

1 ... CPU, 2 ... Program memory, 3 ... Working memory, 4 ... Performance data memory, 5 ... Key detection circuit, 6 ...
Switch detection circuit, 7 ... Display circuit, 8 ... Sound source circuit, 9 ...
Keyboard, 1A ... Numeric keypad & keyboard & various switches, 1
B ... Display, 1C ... Sound system, 1D ... Data and address bus

Continuation of the front page (56) Reference JP-A-63-286883 (JP, A) JP-A-63-250696 (JP, A) JP-A-6-348261 (JP, A) JP-A-6-274171 (JP , A) JP-A-3-119381 (JP, A) JP-A-1-167882 (JP, A) JP-A-1-167881 (JP, A) Jitsukaihei 2-76794 (JP, U) (58) Fields investigated (Int.Cl. ⁷ , DB name) G10H 1/00-7/12 G10G 1/04

Claims (3)

(57) [Claims] 1. Performance information supply means for supplying performance information composed of a plurality of note data, extraction means for extracting note data matching a predetermined condition from the performance information, and a plurality of performance information patterns. And a pattern selecting means for generating characteristic data representing musical characteristics of the note data extracted by the extracting means and selecting a performance information pattern from the pattern storing means according to the generated characteristic data. And a music composition means for generating a new music piece in accordance with the performance information pattern selected by the pattern selection means. 2. Performance information supply means for supplying performance information composed of a plurality of note data, and the performance information is divided into predetermined sections and included in the sections.
Indicates a change in pitch or pronunciation timing of the note data
The change type and the parameter corresponding to the change type.
And parameter setting means for setting each between, for each of the predetermined interval, at least one of the pitch and tone generation timing constituting the musical note data, corresponding to the inter-compartment
Performance information changing means for changing the supplied performance information by changing according to the set change type and the parameter, and music generation for generating a new music piece according to the performance information changed by the performance information changing means. And a music composition device. 3. Performance information supply means for supplying performance information composed of a plurality of note data, data supply means for supplying lyrics data composed of syllable data corresponding to lyrics to be composed, and the performance information. And dividing means for associating the lyrics data with each other and dividing the lyrics data into predetermined sections, data changing means for adding or deleting the note data according to the number of the syllable data for each divided section, and the data changing means. An automatic music composition device comprising a music composition generation unit for generating a new music composition according to changed performance information.