JP3239897B1 - Songwriting device and program - Google Patents

Abstract

Abstract: [PROBLEMS] To generate lyric and melody by successfully matching each other without restriction. According to the present invention, common music generation data is used to generate a melody and lyrics. In one embodiment, the original music generation data is supplied from the music generation database A1 to the lyrics generation unit A4 and the melody generation unit (1) A5 via the music generation data selection unit A3,
Lyrics and a melody are generated based on the music generation data. The number of notes of the melody generated by the melody generation unit (1) A5 is calculated by the melody correction unit A6 and the melody generation unit A4.
Modified to match the number of syllables in the lyrics generated by
There is consistency between the melody and the lyrics. In another embodiment, a lyric correction unit is provided instead of the melody correction unit A6. In another embodiment, instead of the melody generation unit (1) A5 and the melody correction unit A6, a melody generation unit (2) having a function of matching the generated melody with the number of syllables of the lyrics is provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a lyric composer and a lyric composer capable of generating melody and lyrics having consistency with each other.

[0002]

2. Description of the Related Art Conventionally, an automatic composition device for generating a melody or a lyric device for generating lyrics have existed as a single unit. For example, as an automatic music composition device,
Japanese Patent Application Laid-Open No. Hei 11-65561 and a lyric writing apparatus are disclosed in Japanese Patent Application Laid-Open No. Hei 10-97529. The former automatic music composition device disclosed in Japanese Patent Application Laid-Open No. 11-65561 is designed to generate a melody matching the lyrics prepared in advance. Further, the latter songwriting device disclosed in Japanese Patent Application Laid-Open No. Hei 10-97529 is designed to generate lyrics suitable for a melody prepared in advance.

[0003] However, an apparatus for generating both a melody and lyrics has not yet been proposed. In particular, there is no device capable of generating consistent melodies and lyrics. Therefore, if the two prior arts described above are combined, a device that composes music first and then composes a song, or a device that composes lyrics first and composes music later, can be considered. I can't get it. In other words, in such a simple combination device, since the lyrics or melody generated later are restricted by the melody or lyrics generated earlier, free lyrics or melody cannot be generated. . In addition, the melody or the lyric generated first is generated without considering the lyric or the melody generated later, so that there is a possibility that the two do not match well.

[0004]

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, one of the melody and the lyrics is not restricted by the other, and the two match well, and the present invention is compatible with each other. An object of the present invention is to provide a songwriting system capable of generating a melody and lyrics.

[0005]

According to the main features of the present invention, a lyrics generating means for generating lyrics data based on music generating data, and a melody generating means generating melody data based on music generating data. A lyrics composer (claim 1) comprising: a lyrics modifying means for increasing or decreasing the number of syllables of the lyrics data generated by the lyrics generating means to match the number of notes of the melody data generated by the melody generating means. Is done.

According to this feature, a lyrics generation step for generating lyrics data based on the music generation data, a melody generation step for generating melody data based on the music generation data, and a lyrics generation step. A lyric composition program (Claim 5) for causing a data processing apparatus to execute a procedure including a lyrics correction step of increasing or decreasing the number of syllables of the lyrics data to match the number of notes of the melody data generated in the melody generation step. Provided.

[0007] The songwriting device according to the present invention further comprises:
It has a vocabulary / sentence database that stores vocabulary or sentences that are candidates for lyrics data, and associates each vocabulary or sentence with a modified example in which the number of syllables is increased or decreased within a range where the meaning or phonology does not change. The correction means can be configured to increase or decrease the number of syllables in the lyrics data by using this modified example.

[0008] In the songwriting device according to the present invention, the lyrics generation means may be configured to generate lyrics data suitable for the genre based on the genre data included in the music generation data. .

In the lyric composer according to the present invention, the tune generating data includes data of the number of notes representing a range of the number of notes in a predetermined section, and the lyric generating means generates the lyrics within a range represented by the data of the number of notes. It is possible to determine the number of syllables and generate lyrics data having the determined number of syllables.

According to the main feature of the present invention, first, music generation data which is a source of melody and lyrics generation is supplied to the lyrics and melody generation means (C4, C5), and the lyric generation means ( C4), the lyrics are generated based on the rhythm feature data and the like included in the music generation data, and the melody generation means (C5) generates the melody based on the music generation data. According to C6), the number of syllables in the lyrics is increased or decreased to match the number of notes in the melody.
Note that the parentheses are corresponding symbols used in embodiments described in detail below for convenience of understanding, and the same applies to the following.

As described above, after the lyrics and the melody are generated based on the music generating data (C4, C5), the consistency between the lyrics and the melody, that is, the number of the melody notes and the number of the lyrics syllables are matched (C6). So that one can generate melodies and lyrics without being constrained by the other,
Moreover, since both are generated based on the common music generation data, a melody and lyrics that match well are generated. In addition, since the lyrics are corrected (C6) so that the number of melody notes and the number of syllable syllables are matched, it is possible to compose lyrics using the melody initially generated based on the music generation data.

In this lyric composer, in the vocabulary / sentence database, in addition to the vocabulary / sentence having a predetermined number of syllables, a modified example in which the number of syllables is increased or decreased within a range in which the vocabulary / sentence and the meaning or phonology do not change, The lyrics are stored as lyrics candidates, and the lyric correction means (C6) increases or decreases the number of syllables of the lyrics generated by the lyric generation means (C4) by using a modified example of the vocabulary / sentence database when correcting the lyrics. It is possible to modify the lyrics to match the melody without changing the atmosphere of the lyrics.

Further, in the songwriting device according to the present invention, the song generating data which is the basis of the songwriting includes the genre data of the song, and the word generating means (C4) uses the genre data to categorize the song into a genre. Suitable lyrics can be generated.

Further, in the songwriting device according to the present invention, the music generation data (rhythm characteristic data) includes data on the number of notes, and the lyrics generation means (C4) determines the number of notes based on the data. It is possible to generate lyrics for the number of syllables that have been set.

[0015]

[0016]

[0017]

[0018]

Preferred embodiments of the present invention will be described below in detail with reference to the drawings. The following embodiment is merely an example, and various modifications can be made without departing from the spirit of the present invention.

[Hardware Configuration] FIG. 1 is a block diagram showing a hardware configuration of a songwriting and composition processing system according to an embodiment of the present invention. In this example, the system includes a central processing unit (CPU) 1, a read-only memory (ROM) 2,
In addition to a random access memory (RAM) 3, an external storage device 4, first and second detection circuits 5, 6, a display circuit 7, etc., a sound source circuit 8 and an effect circuit 9 are provided. Are connected to each other via a bus 10.

The CPU 1, which controls the entire system, performs various controls using a clock from the timer 11 in accordance with a predetermined software program, and particularly performs mainly a songwriting process described later. A predetermined control program for controlling the system is stored in the ROM 2. These control programs include various programs and various tables for the songwriting process according to the present invention, along with basic information processing. Various data (database for generating music, vocabulary / sentence database, etc.) can be included. The RAM 3 is used as a work area for storing data and parameters necessary for various processes, and for temporarily storing various data and the like being processed.

The external storage device 4 includes a hard disk drive (HDD), a compact disk read only memory (CD-ROM), a floppy disk (FD), a magneto-optical (MO) disk, and a digital multipurpose disk (DVD). ), A device using a portable storage medium such as a semiconductor memory, and can store various control programs and various data. Accordingly, programs and various data (such as a song generation database and a vocabulary / sentence database) necessary for songwriting and the like can be read from the external storage device 4 into the RAM 3 in addition to using the ROM 2. Accordingly, the processing result can be recorded in the external storage device 4.

The first detection circuit 5 includes a performance operator device 12 having a performance operator such as a keyboard for performing a performance operation.
Is connected. To the second detection circuit 6, a panel operator device 13 having various setting operators such as switches is connected, and these setting operators are called panel operators.
Used to make various settings for processing such as songwriting. The display circuit 7 is connected to a display 14 and various indicators for displaying and indicating various information necessary for the songwriting process.

An effect circuit 9 composed of a DSP or the like is connected to the sound source circuit 8, and a sound system 15 including a speaker is connected to the effect circuit 9. These devices 8,
The sound generating means 9 and 15 have a tone generating function having a tone generating function and a sound generating function. With this musical sound generating means, RA
By reading various performance data and lyrics data stored in the M3 or the external storage device 4, etc., it is possible to reproduce the corresponding musical sound or voice.

The bus 10 is also connected via a MIDI interface (I / F) 16 to another MIDI information processing device 17 such as an electronic musical instrument, a sound source, and a keyboard. Data can be exchanged. A communication interface (I / F) 18 is further connected to the bus 10, and the communication I / F 18
It can be communicably connected to a server computer 20 or the like via a communication network 19 such as a local area network (LAN), the Internet, or a telephone line. Therefore, the control program and various data can be stored in the external storage device 4 from the server computer 20 or the like.

The system shown in FIG. 1 can be implemented by a data processing device in the form of an electronic musical instrument having dedicated performance operators and panel operators. However, the system shown in FIG. 1 uses a personal computer (PC). A songwriting system may be configured using a data processing device having an equivalent data processing function.

[Music generation data and vocabulary / sentence data]
In the songwriting / composition processing system according to one embodiment of the present invention, the songwriting / composing process of the first to third embodiments described later in [1] to [3] is executed, and the music generation data and the vocabulary / sentence data are obtained. To compose a desired song. FIG. 2 shows a configuration example of music generation data used in the songwriting system according to an embodiment of the present invention, and FIG. 3 shows a configuration example of vocabulary / sentence data used in the system.

In the music generation database constructed in the ROM 2 or the external storage device 4 of this system, music generation data is recorded as a template. FIG.
As shown on the left side, each music generation data includes a melody generation data and an accompaniment generation data and is assigned a genre, and the melody generation data and the accompaniment generation data have contents corresponding to the genre. Has become. For example, in the genre of “love ballad”, the number of notes is small, and the accompaniment style is the ballad style.
In the genre of "minor enka", the chord progression is a minor chord progression.

As shown in the center (upper part) of FIG. 2, the melody generation data is composed of music composition data, rhythm characteristic data, pitch characteristic data, chord progression data, and the like. First, as shown on the right side of FIG. 2, the music composition data includes block configuration data, phrase configuration data, phrase configuration data, and the like, and the number of blocks, phrases, and phrases is defined. The block configuration data may be, for example, “Music is“ A block ”and“ B block ”
The phrase configuration data represents, for example, "the song is composed of four phrases," the phrase a "-" the phrase b "-" the phrase a '"-" the phrase c ". And the phrase configuration data represents a phrase configuration such as "each phrase is composed of two phrases".

The rhythm characteristic data is composed of basic note length data and note number data, information indicating the presence or absence of syncopation, and the like. It is used as instruction data. The basic note length data represents the note length of a basic note in a block, a phrase, or a phrase, for example, “eighth note” or “1”.
The maximum number of notes per block, phrase, or phrase is defined accordingly. The number-of-notes data indicates the number of notes (frequency of notes) in each block, each passage or each phrase by the number of notes relative to the maximum number of notes. The range of the number of notes is defined by “place”, “small eyes” and the like.

The pitch feature data is composed of a range, dynamics and the like (not shown), and the chord progression data is a string of chord names (including time data if necessary).

The accompaniment generation data is shown at the center (lower part) in FIG.
As shown in (1), it is composed of chord progression data and style designation data, and the chord progression data is used in common with the melody generation data. The style designation data is data for designating any of a large number of separately stored accompaniment style data. The accompaniment generation data may include other data.

On the other hand, the vocabulary / sentence database constructed in the ROM 2 or the external storage device 4 of this system is:
As illustrated in FIG. 3, a vocabulary or a short sentence (right side) to be lyrics is divided by a minimum of M syllables to a maximum of N syllables (left side), and for example, a vocabulary for each syllable corresponding to one phrase or It is a database of sentences. The division of the number of syllables is indicated based on the melody generation data [tune composition data and rhythm characteristic data (maximum pitch data and note number data)] in generating the lyrics, and the vocabulary and vocabulary belonging to the specified division are specified. An appropriate vocabulary / sentence is selected from the sentences by a predetermined method.

As shown on the right side of FIG. 3, vocabulary / sentence data representing vocabulary / sentences corresponding to the syllable number is stored in each syllable number section. "Genre" data indicating whether the syllables can be used, "Modified example" data for slightly increasing or decreasing the number of syllables, and the like are stored.

The genre data indicates the approximate contents of the lyrics in a song genre, and is used to select lyrics suitable for the genre. For example, in "all genres", there are many words that represent general scenes, and in "love songs", many words that show the affection of men and women appear.

The modified example data is class data of the vocabulary / sentence data in which particles are inserted or deleted or replaced with other words having a similar meaning, and the number of syllables is slightly increased or decreased from the original vocabulary or sentence. Let me. The modified example data is used to increase or decrease the number of syllables in the lyrics in a third embodiment described later.

[1] Lyrics Composition Process of First Embodiment FIG. 4 is a functional block diagram of the lyric composition process of the first embodiment executed by the lyric composition system according to one embodiment of the present invention. Here, the outline of the songwriting function according to each embodiment will be described very simply with reference to FIG. In the songwriting system according to one embodiment of the present invention, common melody generating data is used to generate a melody and lyrics.

In the first embodiment, the melody and the melody generation data which are the basis of the lyric generation are obtained from the tune generation database A1 via the tune generation data selection unit A3 through the lyric generation unit A4 and the melody generation unit (1). ) A5, and the lyrics and the melody are generated based on the music generation data. The number of notes of the melody generated by the melody generation unit (1) A5 is corrected by the melody correction unit A6 so as to match the number of syllables of the lyrics generated by the lyrics generation means A4. Consistency is achieved.

In the second embodiment, a melody generating unit having a function of matching the generated melody with the number of syllables of the lyrics instead of the melody generating unit (1) A5 and the melody correcting unit A6 of the first embodiment. (2) is provided (FIG. 1)
0). Furthermore, in the third embodiment, a lyric correction unit is provided in place of the melody correction unit A6 of the first embodiment, and the number of syllables in the lyrics is corrected to match the number of notes in the melody (FIG. 12).

Now, the first embodiment will be described in more detail with reference to FIG. As described above, the music generation database A1 and the vocabulary / sentence database A2 store the music generation data and the vocabulary / sentence data, respectively. The music generation data selection unit A3 selects music generation data (template) stored in the music generation database A1 according to the genre designated by the user, etc., and, together with the genre data corresponding to the music genre,
Among the melody generation data, music composition data indicating the number of phrases, etc., the number of notes (some data of notes) and 1
Rhythm feature data or the like defining the maximum number of notes of a phrase (determined by basic tone length data) is sent to the lyrics generation unit A4, and music composition data, rhythm feature data, pitch feature data, chord progression data and the like are transmitted to a melody generation unit. (1) Send to A5.

[Generation of lyrics data] First, the lyrics generation unit A4
Then, based on the music composition and rhythm characteristic data of the music generation data, the vocabulary / sentence data of the vocabulary / sentence database A2 is referred to, and vocabulary / sentence candidates to be assigned to each phrase are extracted according to the genre data. . The extracted data is displayed on the display 14 and presented to the user. Therefore, the user can select any vocabulary
When a sentence is selected, lyrics data representing the selected vocabulary / sentence is generated. The lyrics data generated in this case may be randomly selected on the system side from the extracted vocabulary / sentence candidates. Further, the user may be able to edit the vocabulary / sentence of the lyrics data as needed.

FIG. 5 shows an example in which lyrics data is generated when a syllable (character) number is specified for each block with a song structure of two blocks and four phrases. Hereinafter, the procedure for generating lyrics data will be described with reference to this example. In the song generation data selection unit A3, the leftmost “block” to
"A block" as indicated by "data for generation"
And music composition data representing the music structure with eight blocks of “first block” to “eighth phrase”, two blocks of “B block”, four phrases of “first phrase” to “fourth phrase”,
Melody generation data having rhythm characteristic data in which the basic pitch and the frequency of the number of notes are specified for each of the blocks A and B are selected together with the genre data specifying the music genre, and sent to the lyrics generation unit A4.

The lyric generation unit A4 first calculates lyric generation conditions relating to the number of syllables (characters) based on the transmitted music generation data (genre data, music composition data, and characteristic data). In other words, based on the number of phrases obtained from the music composition data and the rhythm feature data, the maximum number of notes of one phrase (depending on the basic pitch), the number of notes (frequency), and the like, the fifth row in FIG. As described above, the maximum number of syllables and the number of syllables (characters) for each phrase are calculated as lyric generation conditions, and the calculated lyric generation conditions are used as syllable (character) number indication data for the lyrics candidates.

For example, when the basic note length is designated as "eighth note" as in the block A in FIG. 5, the maximum number of syllables is set to "maximum 16 syllables" (FIG. 5: fifth stage). ("Lyrical condition") right side]. For the number of syllables (characters),
Number of notes "Small", "Medium" and "Multiple" of data
Respectively, "small" = range of 1 to 7 syllables (characters), "medium" = range of 8 to 12 syllables (characters), and "multiple" = range of 13 to 16 syllables (characters). Is automatically selected. In the example of FIG. 5, since the data of the number of notes in the A block is somewhat “small”, as shown on the left side of the fifth row (“lyric generation condition”), the number of characters “small” [=
1 to 7 syllables]. Then, the number of syllables (characters) randomly determined from the range selected according to the content of the data to some extent is the syllable for the lyrics of each phrase (first to fourth phrases) in the block (A block). This is a condition for generating a (character) number.

When the basic note length is specified as "sixteenth note" as in the B block, the maximum number of syllables is set to "maximum 32 syllables" (FIG. 5: fifth stage (" Lyric generation condition ") right side]. Regarding the number of syllables (characters), the number of syllables corresponds to “small”, “middle” and “multiple” of the data, respectively, and “small” = 1 to 14 syllables (characters), The range of “medium” = 15 to 24 syllables (characters) and the range of “multiple” = 25 to 32 syllables (characters) are automatically selected. In the example of FIG. 5, since the data of the number of notes is somewhat “middle”, the number of characters “middle” [= 15 to 24 syllables (characters) as shown on the right side of the fifth row (“lyric generation condition”) ] Range is selected. Then, the number of syllables (characters) randomly determined from the range selected according to the content of the data or the number of notes is determined by the block (B block).
Is the lyric generation condition of the number of syllables (characters) for the lyrics of each phrase (the fifth to eighth phrases).

Next, a vocabulary / sentence which satisfies the lyric generation condition (syllable number instruction data) calculated in this way and satisfies the music genre specified by the genre data is obtained from the vocabulary / sentence database A2. Searched and extracted. When there are a plurality of extracted vocabularies and sentences, as described above, the vocabulary and sentences are selected by the user from the vocabulary and sentences or randomly selected by the system, and the selected vocabulary and sentences are each phrase. Are sequentially generated as lyrics data. 5 ("Lyrical example")
Shows an example of the lyrics generated in this way, and the parentheses indicate the number of syllables (characters) for each phrase of the generated lyrics.

[Generation and modification of melody data]
The melody generation section (1) A5 generates melody data based on the melody generation data of the music generation data. For example, the number of notes in the melody data is determined at random according to the number of notes, the number of notes represented by the data, the maximum number of notes in one phrase defined from the maximum length data, and the like. Here, the data relating to the number of notes is similar to the number of syllables of the lyrics because the original melody generation data is also used in the lyrics generation unit A4, but does not completely match due to the presence of random elements. . Therefore, the correction of the mismatched melody part is performed by the melody correction unit A6.
It is left to the function of.

On the other hand, the melody modifying unit A6 increases or decreases the number of notes in the melody data generated by the melody generating unit (1) A5 according to the number of syllables of the lyric created by the lyric generating unit A4. A process for correcting the melody data is performed, and this melody correction process is referred to as "matching the melody to the lyrics". This "matching of melody to lyrics" process is not a process of generating a new melody,
In principle, the number of notes is simply increased or decreased based on the melody already generated by the melody generation unit (1) A3.

That is, the following matching rules (a) and (4) are applied to the lyrics data generated by the lyrics generation unit A4.
By applying (b), the number of melody notes is adjusted (increased / decreased) in accordance with the number of notes in the melody data and the number of syllables in the lyrics data: (a) When the number of notes> the number of syllables (characters) Priority is assigned to the type of notes (dots) to be reduced. For example, a priority order such as priority = “other> chord constituent sound> strong beat> phrase end> phrase start” is set. To delete notes (dots). Basically, the deleted note is set as a rest, but the length of the note immediately before the deleted note may be extended as necessary. If there are many notes of the same rank as deletion candidates, a note to be deleted is selected at random from these notes. (B) In the case of the number of notes <the number of syllables (characters): ・ Duplicate the sound length and add the same sound. For example, one quarter note is divided into two eighth notes of the same pitch. -The rest part is filled with the note immediately before or after the rest. And so on.

FIG. 6 shows an example of melody data correction by the "matching of melody to lyrics" process. For example, assuming that the first phrase of the melody is generated as shown in FIG. 6A, the number of notes is five, and the lyrics corresponding to this phrase are “shadow shakes” (lower in FIG. 6B).
Less than 7 syllables. Therefore, the matching rule (b)
Need to be applied to add two. In this example,
As shown in (1), the original half note is divided into two to add a quarter note, and as shown in (2), the same note immediately after the rest is added to the original rest position, The melody is modified as shown in FIG. 6B.

The fourth phrase of the melody is shown in FIG. 6C.
, The number of notes is 21
Therefore, the number of syllables of the lyrics “you… で に” (lower in FIG. 6D) corresponding to this phrase needs to be reduced by four to four. Therefore, four notes indicated by * (all of them are notes corresponding to "others") are reduced according to the priority of the matching rule (a). Then, for the third * mark, as shown in (3),
After deleting this note, extend the previous note, and eventually
The melody is modified as shown in FIG. 6D.

Similarly, the melody of the second to fourth and sixth to eighth phrases can be corrected (the description is omitted).

The following is a list of items to be noted with respect to the generation or modification of the lyrics data and the melody data: (1) In the example of FIG. 6, the phrase is changed from the beginning of the first bar to the second bar. However, if there is a pickup (the part into which the previous phrase is embedded), the phrase section is shifted to the left only by this pickup. (2) For syllables, in the explanation so far, syllables (number)
== characters (numbers), but sound-repellent flights (n) and prompting flights (tsu) may be ignored, or "are" may be treated as one syllable. (3) If dissatisfaction occurs when assigning the generated lyrics to the melody, editing may be performed by, for example, generating the lyrics or selecting words again. (4) Although the large number of notes (slight number of notes data) is used as an example of the melody generation data, the number of notes may be specified. In this case, although the number of melody types to be generated is reduced, the number of notes of the lyrics can be specified, so that the assignment to the notes is simplified. (5) In the melody matching process (melody correction process), the priority of the chord sound may be set lower as the deletion target.

[Generation of Accompaniment Data and Output of Performance Data] By the way, the accompaniment generation unit A7
From 1, accompaniment generation data such as style designation data and chord progression data are provided from the music generation data via the music generation data selection unit A3, and the accompaniment data is generated based on the accompaniment generation data. . The accompaniment generation unit A7 is provided as necessary. For example, when singing with a capella, the accompaniment generation unit (A7, B7, C) is used in this embodiment and other embodiments described later.
7) may not be necessary.

Next, the synthesizing unit A8 synthesizes the lyric data generated by the lyrics generation unit A4, the melody data corrected by the melody correction unit A6, and the accompaniment data generated by the accompaniment generation unit A7. Create performance data with lyrics in one file. The output section A9 outputs the performance data.

Examples of the performance data output at the output unit A9 are as follows: (1) Tone generation based on performance data composed of melody data and accompaniment data is performed by a music generation system (8, 9, 15).
Through the display or in the form of a musical score
To be displayed. (2) The lyrics based on the lyrics data are displayed on the display 14 or sung by voice through the tone generation system (8, 9, 15). (3) The data is stored in the storage medium of the external storage device 4 in the format of “performance data + lyrics data” or the format of the audio data described in (2). (4) Attach performance data with lyrics to e-mails and send them to others, or upload them to server computer 20. (5) Play the accompaniment while displaying the lyrics, make the user sing a melody, and record it. And so on.

[Lyson Generation Processing, Melody Generation Processing 1 and Melody Correction Processing Flow] FIGS. 7 to 9 show the lyrics generation unit and the melody of the lyrics composition processing executed in the lyrics composition system according to one embodiment of the present invention. It is a flowchart showing an example of the verb generation process, the melody generation process 1, and the melody correction process respectively performed by the generation unit (1) and the melody correction unit.

First, in the lyrics generation processing of FIG.
In step L1, the lyrics generation unit A4 acquires genre data, music composition data, and rhythm characteristic data from the music generation database A1 via the music generation data selection unit A3. In the next step L2, the number of syllables of the current phrase is determined based on the acquired music composition data and rhythm characteristic data. In the following step L3, the syllable number represented by the genre data and the number of syllables determined in step L2 are determined. To extract vocabulary / sentence candidates by referring to the vocabulary / sentence database.

Next, in step L4, the vocabulary / sentence candidate extracted in step L3 is presented to the user on the display 14 to prompt the user to select a vocabulary / sentence. Then, in step L5, the selection from the user is received. To determine the vocabulary / sentence, and generate lyrics data of the phrase.

In the next step L6, it is checked whether or not the current phrase is the last phrase. If the current phrase is not the last phrase, the process returns to step L2, and the processes in steps L2 to L6 are repeated for the next phrase. Then, if it is determined in step L6 that the phrase is the last phrase, the flow of the word generation processing ends.

In the melody generation processing 1 of FIG. 8, in the first step M1, the melody generation unit (1) A5
Acquires rhythm feature data, pitch feature data, chord progression data, and song configuration data from the song generation database A1 via the song generation data selection unit A3. In the next step M2, the number of notes of the current phrase is determined based on the acquired rhythm feature data, and the following step M3
Then, a rhythm pattern is determined based on the rhythm characteristic data and the number of notes. Further, in step M4, a pitch is assigned to each hit point of the rhythm pattern based on the pitch characteristic data and chord progression data, and the melody data of the phrase is converted. Generate.

In the next step M5, it is determined whether or not the current phrase is the last phrase. If not, the process returns to step M2, and the processes in steps M2 to M5 are repeated for the next phrase. Then, step M5
If it is determined that the melody is the last phrase, the melody generation processing 1 is terminated.

Returning to step M2, when melody generation is repeated for a new phrase, a melody is newly generated for a phrase which is the same as the preceding phrase in the phrase configuration, as in steps M2 to M5. It is better to copy the melody of the previous passage without doing it. In addition, for a similar phrase in the phrase configuration, it is preferable to copy a part of the melody of the preceding phrase.

In the melody correction process of FIG. 9, when the melody correction unit A6 acquires the melody data generated by the melody generation unit (1) A5 and the lyrics data generated by the lyric generation unit A4, the first step N1 Then, it is checked whether or not the number of notes in the melody data in the current phrase is larger than the number of syllables in the lyrics data. If the number of melody notes in the current phrase is greater than the number of (song) syllables, in step N2,
After deleting the notes (dots) in accordance with the priority of the fitting rule (a) described above, the process proceeds to step N3, and if not, the process immediately proceeds to step N3.

At step N3, it is determined whether or not the number of notes in the melody data in the current phrase is smaller than the number of syllables in the lyrics data. If the number of melody notes in the current phrase is smaller than the number of (song) lyric syllables, in step N4, note division and rests are converted into notes in accordance with the fitting rule (b) described above. Immediately proceeds to step N5.

At step N5, it is determined whether or not the current phrase is the last phrase.
Returning to step N1, step N for the next phrase
The processing of 1 to N5 is repeated. If it is determined in step N5 that the phrase is the last phrase, the melody correction process ends.

[2] Lyrics Composition Process of Second Embodiment FIG. 10 is a functional block diagram of a lyric composition process of the second embodiment which is executed by the lyric composition system according to one embodiment of the present invention. In the songwriting process of the second embodiment,
The melody data is generated according to the number of syllables in the lyrics data. In other words, the melody generation unit (2) B5 is provided instead of the melody generation unit (1) A5 of the first embodiment, the melody correction unit A6 of the first embodiment is eliminated, and another database and processing unit B1 ~ B4, B7
As shown, B9 to B9 have substantially the same configuration as the database and processing units A1 to A4 and A7 to A9 of the first embodiment, respectively.

When determining the number of notes in the melody data, the melody generating unit (2) B5 in the songwriting process according to the second embodiment sets the number of syllables equal to the number of syllables of the lyrics data generated by the lyric generating unit B4. . That is, the melody generation unit (2)
In B5, melody generation processing 2 described below is executed,
The number of notes is determined from the number of syllables in the lyrics data without using data on the number of notes in the melody generation data. Therefore, the number of syllables of the (song) lyrics and the number of notes of the melody can be matched with a simple configuration without providing the melody correction unit A6 as in the first embodiment.

[Melody Generation Process 2] FIG. 11 shows an example of the melody generation process 2 executed by the melody generation unit (2) in the songwriting process performed in the songwriting system according to one embodiment of the present invention. FIG. In the first step P1 of this processing flow, the melody generation unit (2) B5 sends the rhythm feature data, pitch feature data, chord progression data, and song configuration data from the song generation database B1 via the song generation data selection unit B3. The data is obtained, and the lyrics data generated by the lyrics generation unit B4 is obtained.

In the next step P2, the number of notes of the current phrase is determined in accordance with the number of syllables of the (song) phrase represented by the lyrics data. In the next step P3, the rhythm pattern is determined based on the rhythm characteristic data and the number of lyric notes. Then, in step P4, based on the pitch feature data and the chord progression, a pitch is given to each hit point of the rhythm pattern to generate melody data of the phrase.

In the next step P5, it is checked whether or not the current phrase is the last phrase. If not, the process returns to step P2, and the processes in steps P2 to P5 are repeated for the next phrase. Then, step P5
If it is determined that the melody is the last phrase, the melody generation process 2 ends.

When returning to step P2 and repeating the melody generation of the new phrase, in the melody generation process 2, as for the passage which is the same as the previous passage in the passage configuration, the processing in steps M2 to M5 is performed. It is better to copy the melody of the previous passage without generating a new melody. In addition, for a similar phrase in the phrase configuration, it is preferable to copy a part of the melody of the preceding phrase. However, since it is necessary to match the number of notes with the number of syllables of the lyrics, it is preferable to increase or decrease the number of hit points.

[3] Lyrics Composition Process of Third Embodiment FIG. 12 is a functional block diagram of a lyric composition process of the third embodiment executed by the lyric composition system according to one embodiment of the present invention. In the songwriting process of the third embodiment,
It is characterized in that after generating the lyrics data, the number of syllables is increased or decreased in accordance with the number of notes in the melody. That is, a lyric correction unit C6 is provided in place of the melody correction unit (1) A6 of the first embodiment, and the other database and processing units C1 to C5, C
7 to C9 have substantially the same configuration as the database and the processing units A1 to A5 and A7 to A9 of the first embodiment, respectively, as illustrated.

The lyric composer C6 in the lyric composer of the third embodiment executes a lyric amendment process, which will be described later, and calculates the number of syllable syllables of the lyrics data generated by the lyric generator C4 to the melody generator (1). ) Increase or decrease according to the number of notes in the melody data generated in C5. To increase or decrease the number of notes, the modified meaning data (the rightmost column in FIG. 3) attached to each vocabulary / sentence data in the vocabulary / sentence database C2 is used to determine the substantial meaning of the already generated vocabulary or sentence. The number of syllables can be changed without changing. For example, in the case where the seven-syllable word “leaf shakes” in the lyrics data is corrected to six syllables according to the number of notes 6 in the melody data, the modification of “the leaf shakes” Leaf shaking ”(6
Syllable). This allows the number of syllables of the lyrics to match the number of notes of the melody without correcting the melody.

[Lyric Correction Processing] FIG. 13 is a flow chart showing an example of the lyrics correction processing executed by the lyric correction section in the lyric composition processing executed in the lyric composition system according to one embodiment of the present invention. Upon acquiring the melody data generated by the melody generation unit (1) A5 and the lyrics data generated by the lyrics generation unit A4, the lyrics correction unit C6 determines the lyrics data of the current phrase in the first step Q1 of this processing flow. Check whether the number of syllables is greater than the number of notes in the melody data.

If the number of syllables in the current phrase is larger than the number of melody notes, the process proceeds to step Q2 to reduce the number of syllables before proceeding to step Q3, otherwise to step Q3. In step Q2, referring to the vocabulary / sentence database, according to the modified example data corresponding to the lyrics data, a modified example of the number of syllables corresponding to the number of melody notes is selected as a new (song) word, and the desired number of lyric syllables is determined. To reduce.

At step Q3, it is checked whether or not the number of syllables in the lyrics data in the current phrase is smaller than the number of notes in the melody data. If the number of syllables in the current phrase is smaller than the number of melody notes, the number of syllables in the current phrase is determined in step Q4. , And then proceeds to step Q5, otherwise proceeds directly to step Q5. In step Q5, referring to the vocabulary / sentence database, according to the modified example data corresponding to the lyrics data, a modified example of the number of syllables corresponding to the number of melody notes is selected as a new (song) word, and the desired number of lyric syllables is determined. To increase.

At step Q5, it is determined whether or not the current phrase is the last phrase.
Returning to step Q1, step Q is performed for the next phrase.
The processing of 1 to Q5 is repeated. Then, when it is determined in step Q5 that the phrase is the final phrase, the word correction processing ends.

[Various Embodiments] While the songwriting process according to the present invention has been described with reference to specific embodiments, the embodiments are not limited to the illustrated ones. For example,
Regarding the functions of the system, (1) the lyrics and composition algorithm [(song) lyrics generation / correction unit and melody generation / correction unit processing method] are not limited to those illustrated. (2) In each embodiment, the accompaniment generation portion may not be provided. (3) At least two of the three embodiments may be provided at the same time, and the user may be able to select which of the three embodiments is to be used for songwriting.

Further, with respect to the modification to be stored in the vocabulary / sentence database, in the embodiment, vocabulary / sentences having different numbers of syllables and having substantially the same meaning are stored in association with each other. However, if the meaning of the lyrics is not considered very important, vocabulary / sentences whose meaning changes may be associated. For example, if a word emphasizes the sound (phoneme) of a word, and the meaning is not important if the sound (phoneme) is similar, use a vocabulary or sentence whose sound (phoneme) is more similar than meaning. It may be preferentially associated as a modification.

The method of storing the modified example in the vocabulary / sentence database is not limited to the method of storing the entity of the modified example for each vocabulary / sentence as in the embodiment. The vocabulary and the pointer to the sentence may be stored as a modified example in the same manner as the sentence.

In the melody correction processing, an increase / decrease rule (matching rules (a) and (b)) for increasing / decreasing the number of notes in the melody may be selectable by the user. For example, in the case of (a) an increase in the number of notes, the user can select note division or conversion of a rest, and in the case of (b) a decrease in the number of notes, the user selects the priority of the hit point type. It can be so.

When increasing the syllables of a lyric, the vowels may be lengthened to form a long vowel (the one followed by the notation "-"), and the syllables may be increased by counting the expanded vowel "-" as one syllable. . Conversely, in order to reduce the syllable of the lyric, the vowel "-" which is extended may be deleted to reduce the syllable.

The form of the system (data processing device)
The present invention is not limited to the form of the electronic musical instrument but may be a form of PC + application software, and may be applied to a karaoke device, a game device, a portable communication terminal such as a mobile phone, an automatic performance piano, and the like. When applied to a portable communication terminal, not only the case where the predetermined function is completed only with the terminal, but also a part of the function is provided on the server side, and the predetermined data processing function as a whole system consisting of the terminal and the server May be realized.

When the form of the electronic musical instrument is adopted, the form is not limited to the keyboard instrument, but may be a string instrument type, a wind instrument type, a percussion instrument type, or the like. Further, the present invention is not limited to the case where the sound source device, the automatic performance device, and the like are built in one electronic musical instrument main body.
Each of them may be a separate device, and each device may be connected using communication means such as MIDI or various networks.

As for communication with an external device, the communication interface and the communication network are not limited to wired ones, but may be wireless or both. Also, M
IDI is not limited to a dedicated MIDI interface, but may be RS-232C, USB (Universal Serial Bus), IEEE 1394 (I Triple E1).
394) and other general-purpose interfaces
An interface may be configured. In this case, MID
Data other than the I message may be transmitted and received at the same time.

[0086]

As described above, according to the present invention, first, tune generation data, which is the basis of melody and lyrics generation, is supplied to the lyrics and melody generation means (C4, C5). After the lyrics and the melody are generated based on the data, the consistency between the two (the number of notes and the number of syllables) is obtained by the lyrics correcting means (C6).
One can generate a melody and lyrics without being constrained by the other, and since both are generated based on common music generation data, it is possible to generate a melody and lyrics that match well.

[0087]

[Brief description of the drawings]

FIG. 1 is a block diagram showing a hardware configuration of a songwriting and composition processing system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration example of music generation data used in a songwriting composition processing system according to an embodiment of the present invention;

FIG. 3 is a diagram showing a configuration example of vocabulary / sentence data used in the songwriting composition processing system according to one embodiment of the present invention;

FIG. 4 is a functional block diagram of a songwriting process according to the first embodiment, which is executed in the songwriting system according to the embodiment of the present invention;

FIG. 5 is a diagram showing an example of (song) lyrics generation for explaining a procedure for generating lyrics data according to an embodiment of the present invention.

FIG. 6 is a diagram showing a specific correction example for explaining the melody data correction processing (“matching of melody to lyrics” processing) according to one embodiment of the present invention;

FIG. 7 is a flowchart showing (song) word generation processing according to an embodiment of the present invention.

FIG. 8 is a flowchart showing a melody generating process 1 according to an embodiment of the present invention.

FIG. 9 is a flowchart showing a melody correction process according to an embodiment of the present invention.

FIG. 10 is a functional block diagram of a songwriting process according to a second embodiment executed in the songwriting system according to the embodiment of the present invention;

FIG. 11 is a flowchart showing a melody generating process 2 according to an embodiment of the present invention.

FIG. 12 is a functional block diagram of a songwriting process according to a third embodiment executed in the songwriting system according to the embodiment of the present invention;

FIG. 13 shows a song according to an embodiment of the present invention (song).
It is a flowchart showing a word correction process.

[Explanation of Signs] A1, B1, C1 music generation database, A2, B2, C2 vocabulary / sentence database, A3, B3, C3 music generation data selection unit, A4, B4, C4 (song) lyric generation unit, A5 , C5 melody generation unit (1), B5 melody generation unit (2), A6 melody correction unit, C6 (song) lyric correction unit, A7, B7, C7 accompaniment generation unit, A8, B8, C8 synthesis unit, A9, B9 , C9 output section.

────────────────────────────────────────────────── ─── Continuation of front page (56) References JP-A-11-237881 (JP, A) JP-A-11-224089 (JP, A) JP-A-9-62263 (JP, A) JP-A-10- 97529 (JP, A) JP-A-11-161267 (JP, A) JP-A-9-50278 (JP, A) JP-A-11-65561 (JP, A) JP-A-4-248658 (JP, A) JP-A-4-256160 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10H 1/00 101-102 G10G 1/00-3/04

Claims (5)

(57) [Claims] 1. A lyric generation means for generating lyrics data based on music generation data, a melody generation means generating melody data based on music generation data, and a syllable of the lyric data generated by the lyrics generation means. A lyric composer comprising: a lyric correcting means for increasing or decreasing the number so as to match the number of notes of the melody data generated by the melody generating means. 2. A vocabulary / sentence storing a vocabulary or a sentence which is a candidate for lyrics data, and a vocabulary / sentence in which each vocabulary / sentence is associated with a modified example in which the number of syllables is increased or decreased within a range in which the meaning or phonology does not change. 2. The songwriting device according to claim 1, further comprising a sentence database, wherein the lyrics correction unit increases or decreases the number of syllables in the lyrics data using the modification. 3. 3. The songwriting device according to claim 1, wherein the lyrics generation means generates lyrics data suitable for the genre based on genre data included in the music generation data. 4. The music generation data includes note number data indicating a range of note numbers in a predetermined section, and the lyric generation means determines the number of syllables of the lyrics within the range indicated by the note number data. The songwriting device according to any one of claims 1 to 3, wherein the lyrics data having the determined number of syllables is generated. 5. A lyric generation step for generating lyrics data based on music generation data, a melody generation step for generating melody data based on music generation data, and a syllable of the lyrics data generated in the lyrics generation step. A songwriting program for causing a data processing device to execute a procedure including a word correcting step of increasing or decreasing the number to match the number of notes of the melody data generated in the melody generating step.