JP2008209535A - System and method for generating music information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable a user to easily generate music information matching one's preference and feelings, without labor and time. <P>SOLUTION: In a system capable of automatically generating music information composed of eight bar lines according to a melody comprising two bar lines inputted by a user, a tone sequence pattern A of the inputted first bar line and a tone sequence pattern B of the second bar line are extracted, and the extracted tone sequence patterns A and B are transformed, based on a change rule so as to generate tone sequence patterns for other bar lines. Then, after generating the tone sequence patterns to the eighth bar line, a chord progression is set, and sound pitch of each bar line is adjusted from a chord assigned to each bar line, or the like. Finally, backing is imparted to generate the music information comprising eight bar lines. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a music information generation system that automatically generates music information, and more specifically, for example, from a melody composed of two measures input by a user, for example, eight measures, sixteen measures, and the like. The present invention relates to a music information generation system that can generate music information of relatively long measures.

  Conventionally, as a music information generation apparatus, an apparatus (Patent Document 1) that can automatically accompany a melody directly input by a user, or generate music information from text included in an e-mail or the like. Apparatus (Patent Document 2), or an apparatus (Patent Document 3) that can generate music information from an image.

  To describe these devices in detail, first, in a device that is capable of accompaniment to a melody directly input by a user as shown in Patent Document 1, first, the input of all melody is received from the user, and the received melody is received. Judgment of key and chord progression. Then, sub melody data is generated from the key and chord progression obtained by the determination, and this sub melody data is synthesized with the main melody so that music information can be generated.

  An apparatus for generating music information from text (Patent Document 2) is designed to generate music information according to the number of characters and contents from a mail to be transmitted, counting the number of characters in the mail, For each number of characters, a music template composed of a note sequence of 1 sound, 2 sounds, 3 sounds, 4 sounds, and 5 sounds prepared in advance is applied. At the same time, the words that match the given keyword are extracted from the text included in the e-mail. For example, “happy, happy, easy, fun, good, light, good, light, south, white, If it contains words such as “above”, it is set in major, and if it contains words such as “sad, arrogant, disgusting, dark, evil, distressed, thought, north, black, below” Is set in minor, and the pitch of the linked music template is modified to generate music according to the length and meaning of the sentence.

As shown in Patent Document 3, an apparatus that generates music information from an image quantizes an image acquired by a camera into 256 gradations, and uses audio information corresponding to the luminance of each pixel as a template in advance. A pitch corresponding to the brightness of each pixel is applied with reference to this template. Specifically, when the luminance of the analyzed pixel is “0 to 10”, the pitch of “A” that is one octave higher than the reference octave is applied, and the luminance of a certain pixel is “247”. ”To 255”, the sound of “H” that is one octave below the reference octave is applied, and these pitches are arranged in the vertical axis direction or horizontal axis direction of the music data. Is generated.
JP 2002-023745 A JP 2004-226671 A JP 2001-350473 A

  However, when music is generated from text as described above, or when music is generated from an image, it is not always possible to generate music that really suits one's preference and sensitivity.

  On the other hand, as shown in Patent Document 1, it is sufficient to input a melody that suits one's preference and sensitivity and perform accompaniment to this, but in this way all melody can be used by the user. It takes a lot of time and effort to get input. In addition, it is difficult for people who do not have specialized knowledge about music to enter a melody that suits their tastes, and even if they take a long time, they may not be able to produce satisfactory music. .

  Therefore, the present invention has been made paying attention to the above-mentioned problems, and provides a music information generation system that can easily generate music information that suits one's preference and sensibility without requiring the user to take time and effort. For the purpose.

  That is, in order to solve the above-mentioned problem, the present invention extracts a value string pattern that constitutes the input melody in a music information generation system that generates a melody of a plurality of measures from a melody input by a user, A melody composed of a plurality of measures is generated by concatenating tone value strings obtained by changing the extracted tone value string patterns.

  Here, the “tone value” indicates the length of the sound, and corresponds to a quarter note or an eighth note on the musical score. In terms of acoustics, this indicates the time from when a sound is output until the next sound is output. In addition, “pitch”, which will be described later, indicates the pitch of the sound, and corresponds to “doremi ...” of each range on the musical score. It is equivalent.

  In such a system, preferably, a melody accepting unit that accepts input of a melody, and a tone value string pattern setting unit that sets a pattern of a tone value string that constitutes the melody from the melody accepted by the melody accepting unit. A change rule storage means for storing a change rule for making a change to the sound value string pattern, and the sound value string pattern set by the sound value string pattern setting means, with reference to the change rule, Pattern change means for generating a note value string pattern in which the value string is changed and melody connection means for connecting the sound value string patterns changed by the pattern change means are provided.

  In this way, you can create only partial melody fragments that match your preferences and sensibility, and enter them to generate music information that includes other measures that take advantage of the atmosphere of the melody. become able to. Thereby, the music information suitable for one's preference and sensitivity can be generated without taking time and effort. In addition, when another melody is connected to the input melody, another melody is generated in consideration of the tone value pattern of the melody created by the user, so that music information having a natural flow can be generated. It becomes like this.

  Further, in such an invention, the chord progression setting means for setting the chord progression for the input melody is provided, and the pitches of each note value in each measure are set as pitches in each constituent chord assigned by the chord progression. It is provided with a pitch adjusting means for adjusting the pitch.

  In this way, it becomes possible to finish the music with a more natural flow by the natural tempo due to the change in the note value string pattern and the natural pitch corresponding to the chord progression.

  In the present invention, the melody tone value patterns input by the user are changed and connected to each other, so that the user only has to input a characteristic portion (melody fragment) to suit his taste and sensitivity. Music information can be generated. In addition, since the other tone value strings to be connected are obtained by changing the input tone value string pattern, the flow can be made natural.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows a configuration example of the music information generation system 1, and FIG. 2 shows functional blocks of the music information generation system 1.

  The music information generation system 1 according to this embodiment includes a client computer 2 and a server computer 3 connected via the Internet. Among these, the client computer 2 is a mobile phone, PDA, PHS, or the like. In addition to a portable terminal such as a personal computer, the computer has at least a function for receiving a melody input by a user and a function for outputting music information composed of a plurality of measures generated by the server computer 3. Yes. On the other hand, the server computer 3 has a function of generating music information composed of a plurality of measures based on the partial melody transmitted from the client computer 2 and transmitting the music information to the client computer 2. In this embodiment, the music information generation system is configured by the client computer 2 and the server computer 3, but it is not necessarily configured by the client computer 2 and the server computer 3. For example, one This music information generation system may be realized by a computer. Alternatively, the function of the server computer 3 may be distributed to a plurality of server computers, and the music information generation system may be configured by the plurality of server computers and the client computer 2. In this embodiment, a melody input consisting of a plurality of measures (for example, two measures) is received from the user, and music information consisting of eight measures is generated by the server computer 3 based on this input, and the client computer 2 uses the information. Output is enabled. Hereinafter, a specific configuration of the music information generation system 1 according to the present embodiment will be described in detail.

  FIG. 2 shows functional blocks constituting the music information generation system 1. The client computer 2 and the server computer 3 constituting the music information generating system 1 include a CPU, a storage device, a communication device, an input device such as a numeric keypad, a keyboard, and a mouse, a display device, and the like, and are stored in this storage device. The following functions are executed based on the program.

  First, the melody accepting means 21 accepts a partial melody input created by the user, and accepts a melody input from the user with a numeric keypad, mouse, keyboard, or the like. Here, the melody will be described when a single melody is input. When accepting an input of this melody, first, as shown in FIG. 3, a staff score is displayed on the display device, and a note value designated by the user can be arranged on the staff score with a predetermined pitch. Here, the default is set to C major, and the time signature is set to 4/4 time, but these keys and time signatures can be changed by selection from the pull tab menu. In the case of accepting partial melodic input, in this embodiment, a staff notation consisting of two measures is displayed, and a desired note value is arranged at a desired pitch in these two measures. Note that the partial melody input by the user does not necessarily have to be two measures, and may be one measure or three or more measures. Moreover, it is not limited to each measure group, The note string of suitable length may be sufficient. In this embodiment, a case where melody input is received within two measures will be described as an example. When accepting an input of a melody consisting of these two measures, first, a user selects a note value from the pull tab menu, and then places this note value at a predetermined pitch of the staff. When assigning the note value, the selected note value is moved up and down with the arrow keys and placed at a desired pitch position, and this note value and pitch are input in sequence, thereby providing two measures. After receiving the input of the melody, the input of symbols such as tie, slur, staccato is accepted. For the input such as tie or slur, selection of the music symbol is accepted by an input key, and selection of a start point or end point of a slur is accepted by an arrow key and a determination button. Note that the setting of music symbols such as ties may be performed simultaneously with the input settings for each note value and pitch. Then, by accepting the input of musical symbols such as note value, pitch, tie and slur in this way, as shown in FIG. 4, the first measure is “do, do, do, shi, la”, A melody in which the second measure is composed of “La, De, Shi, Seo” is input. The input melody is temporarily stored in the storage unit of the client computer 2 and viewed by the user, and then the melody is adapted to the user's preference by being appropriately modified.

  The music type setting means 22 accepts selection of the genre of music to be finally generated. For example, the music type setting means 22 accepts selection of types such as pops, classical music, jazz, rock, and folk songs. When accepting this type of selection, as shown in FIG. 3, a “genre” pull tab menu is displayed on the display screen of the client computer 2, and selection of the type of music provided in this pull tab is accepted. When the music type is set by the music type setting means 22, the server computer 3 performs processing such as providing backing according to the music type. Then, the melody and music type information composed of two measures input by the client computer 2 is transmitted to the server computer 3 in this way.

  When the server computer 3 receives these pieces of information via the internal receiving unit, the tone value string pattern extracting means 31 extracts a tone value string pattern from the melody composed of these two measures. Here, the “sound value string pattern” indicates a pattern of a sequence of sound values divided by each measure, and when there is a tie or the like, it indicates a timing pattern of the output sound. . For example, considering the melody shown in FIG. 4, the note value pattern of the first measure is “scored eighth note, sum of sixteenth note and eighth note, sum of eighth note and quarter note” on the score. , Eighth note, eighth note ”is set as the note value sequence pattern A, and the note value sequence pattern of the second measure is“ sum of half note and eighth note, eighth note, 8 “Note note, eighth note” is set as the note value string pattern B. Since the second and third sounds in the first measure are tied with a “do” sound, the timing of the output sound is the sum of the eighth note and sixteenth note, respectively. And the sum of an eighth note and a quarter note. These tone value sequence patterns A and B are electronically set as time intervals of sounds that are actually output. Here, for the sound value string patterns A and B, only the order pattern of the sound value strings is extracted, and the pattern not considering the pitch is extracted.

  The pattern changing unit 32 generates a sound value string pattern to be connected after the sound value string patterns A and B based on the sound value string pattern A and the sound value string pattern B thus extracted. . In general, music often flows with the same note sequence pattern repeated for each measure and related changes. For this reason, in this embodiment, the sound value string pattern A and the sound value string pattern B are used as basic sound value string patterns, and the sound value string pattern A, the sound value string pattern B itself, or the basic sound value string pattern is defined as The changed tone value string patterns are connected. The tone value string pattern change rule is stored in advance in the change rule storage means 33, and the basic tone value string pattern is changed based on the change rule. This change rule is shown in FIG.

  In the third measure, the basic tone value string pattern B is arranged as it is.

  In the fourth measure, the average value V of the melody note values in the basic note value sequence pattern A and the basic note value sequence pattern B is calculated by the following equation 1, and when V ≦ 240, that is, a relatively short note value is obtained. When there are many notes, four quarter notes are arranged, and when V> 240, that is, when there is a relatively long note value, two half notes are arranged.

  In the fifth measure, the basic note value string pattern A is arranged as it is.

  In the sixth measure, the basic note value string pattern B is arranged as it is.

  In the seventh measure, the basic tone value string pattern B is inverted and arranged. That is, a note value string of “eighth note, eighth note, eighth note, eighth note, half note” is arranged for the basic note value string pattern B shown in FIG.

  In the eighth measure, one half note is arranged when V ≦ 240, and one whole note is arranged when V> 240.

  As a result, the input note value string pattern as shown in FIG. 4 is converted into an eight-measure melody having a note value string pattern as shown in FIG. Here, for the 4th bar and the 8th bar, only the rhythm is set based on the average sound value of the sound value string pattern A and the sound value string pattern B.

  Here, an 8-value note value string pattern is generated, but a rule is set in advance in the change rule storage means 33, thereby generating a 16-measure or 32-measure note value string pattern. You may do it. The change pattern is not limited to the change pattern from the third measure to the eighth measure as described above, and various change patterns may be provided. In this way, the tone value string patterns of each measure are generated, and these tone value string patterns are sequentially connected to the sound value string patterns A and B via the connecting means 34. At this time, for the pitches of the connected melody, the pitches of the basic tone value patterns A and B are used as they are as temporary pitch information, or they are all set to the same pitch.

  The chord progression setting means 35 sets the chord progression of the generated music, and sets one or a plurality of chords in each measure. This chord progression pattern is input from the client computer 2 by the user or automatically set for each music genre. When the user inputs using the client computer 2, the user specifies all chords corresponding to a length of 8 bars. On the other hand, when the chord progression is set automatically, the first two measures entered by the user must correspond to the chords of the melody, so the first and third beats of the constituent notes of each input measure The chord of the measure is extracted from the constituent sounds included in the eye, and the chord progression including the extracted chord is set. Here, in the melody of FIG. 4, since the first beat of the first measure is “do” and the third beat is “do”, a chord suitable for the constituent sound of “do” is set, For the second measure, since the first beat is “La” and the third beat is “Do”, a chord suitable for the constituent sounds of “La” and “Do” is set. Here, when setting a chord, not only the chord of the root, the third, the fifth, and the seventh, but also the tension code (9th, 11th, 13th) may be included. In this way, chords are set up to the second measure, chord progressions including these chords are set, and then chords are assigned to each measure. In FIG. 7, “FM7”, “EM7 A7”, “Dm7 G7”, “CM7 Cdim7”, “FM7”, “EM7 A7”, “Dm7 G7”, and “CM7] are assigned to each bar after the third bar.

  Next, the pitch adjusting means 36 adjusts the pitch of the melody of each measure according to the chord assigned to each measure in this way. The pitch adjusting means 36 converts the sound output at the beginning of the melodic sound in each assigned chord into the sound of the chord sound (hereinafter referred to as “target sound”). At this time, it is arranged so that the pitch between the two tones before and after the melody to be processed falls within one octave, thereby avoiding sudden jumps.

  Specifically, when the target sound is arranged, as a general rule, it is performed based on the priority order table as shown in FIG. 8, and when the following rule 1 is applied, the priority order of the target sound is lowered. Will place the subordinate sound.

<Rule 1 for Lowering Priority Order>
1. 1. After the target sound is processed, the pitch between the two notes before and after the melody to be processed is 1 octave or more. 2. After the target sound is processed, the pitch between the two notes before and after the melody to be processed is the same (that is, the same pitch). After the target sound is processed, the melody to be processed is the same as the melody given as a motif

  That is, first, when converting the target sound, the sound to be arranged is set according to the “type (Maj7, dim7, etc.)” of the given chord, for example, “La” set in Dm7 of the third measure In consideration of the priority order of FIG. 8, the sound value (F) of 3 degrees is selected from “Le” which is the route of Dm7, and placed in the same octave (FIG. 9). For the first melodic sound “do” set for the next G7, the sound “shi” is set three times from the first priority G, but is the same as the next melodic sound. The sound “Fah” of 7 degrees which is the second highest priority is set. However, since this “Fa” is the same degree as the pre-melody, this is excluded, and as a result, “Mi” of 13 degrees of the third priority is arranged.

  In the 4th bar, as shown in FIG. 10, when the pitches of the two notes from the end of the previous bar (3rd bar) transition from a high state to a low state, 1 beat of the 4th bar For the eyes, a pitch that is two degrees below is placed, and if this is not a chord component of the beat, a pitch that is three degrees below is placed. On the other hand, if the pitches of the two notes from the end of the previous bar are transitioning from a low state to a high state, a second higher pitch is placed on the first beat of the fourth bar, which is the chord of the beat. If it is not a constituent sound, a pitch three times higher is placed. For the third beat, a chord constituent sound whose pitch is twice from the note information of the first beat is arranged, and when there are a plurality of candidates, the pitch is arranged according to the rules of target sound arrangement. For the 4th beat, when the average value V of the note value of the measure is 240 or less, when the pitch of the first beat is higher than the pitch of the third beat, 2 beats A pitch that is twice lower than the first note is placed in the eye, and a pitch that is twice higher than the third beat is placed on the fourth beat. If the pitch of the first beat is lower than that of the third beat, the second beat is placed twice higher than the first beat, and the fourth beat The pitches that are two times lower than the third beat are respectively arranged.

  On the other hand, when the average value of the note values of the fourth measure is larger than 240, when the pitches of the two notes from the end of the previous measure (third measure) are transitioning from a high state to a low state, 4 A pitch 2 degrees below is placed on the first beat of the bar, and if it is not a chord constituent sound of the beat, a pitch 3 degrees below is placed. On the other hand, if the pitch of the second note from the end of the previous bar is transitioning from a low state to a high state, a second higher pitch is placed on the first beat of the fourth bar, If the beat is not a chord component, a pitch that is three times higher is placed. Also, for adjusting the pitch of the 3rd beat, arrange a chord structure in which the pitch from the note information of the 1st beat is 2 degrees, and if there are multiple candidates, the pitch is adjusted according to the principle of target sound placement Set.

  For the 8th bar, one whole note is set when the average note value V> 240, and one half note is arranged when the average note value V ≦ 240. At this time, the pitch is the main tone.

  Then, the pitch adjusting means 36 further converts an extraordinary tone (a black key in the case of C major) included in the whole melody into an inner tone for each pitch set in this way. In this case, if the melodic sound that is an extraordinary sound is moving up to a subsequent melodic sound, it is moved up a semitone, and if the melodic sound that is an extraordinary sound is descending to a subsequent melodic sound Move down a semitone.

  As a result, an 8-measure melody composed of a note value string as shown in FIG. 11 is generated from a two-measure note value pattern input by the user.

  The music information generating means 37 generates a melody composed of 8 bars by connecting the bars generated in this way. Then, music information is generated by adding a backing corresponding to the selected genre to the melody consisting of 8 measures generated in this way. For the backing, a backing file having a plurality of rhythm patterns is prepared according to the music genre, and the backing is given while applying the set chord constituent sound thereto.

  The music information generated in this way is transmitted to the client computer 2, and the client computer 2 outputs the received music information through the output means 23 so as to be viewable.

  Next, the flow of processing in the music information generation system 1 configured as described above will be described with reference to the flowchart of FIG.

  First, for example, when a user makes a mobile ringtone that suits his / her preference, a music creation program is started by operating a button on the mobile phone, and a musical score and various kinds as shown in FIG. 3 are displayed on the display screen. Show pull tab menu. Then, selection of a note value to be arranged on a staff notation composed of two measures is accepted, and the note value is arranged at a predetermined pitch using an arrow key or the like. In the same manner, the selection of the note value and the setting of the pitch are performed in the same manner, and the input of a melody consisting of two measures is accepted (step S1). Then, these pieces of information are transmitted to the server computer 3 (step S2).

  When the server computer 3 receives these pieces of information, the server computer 3 extracts a tone value string pattern A and a tone value string pattern B for each measure from the received two-measure melody (step T1), and a change rule. Referring to Fig. 4, the tone value string pattern of each measure after the third measure is generated as follows (step T2).

  First, for the third measure, the note value string pattern B is arranged as it is, and the average value V of the melody note values in the note value string patterns A and B (the quarter note is set to 480 and the eighth note is set to 240). Is equal to or greater than 240, four quarter notes are arranged, and when the average value V is smaller than 240, two half notes are arranged. The tone value string pattern A is arranged as it is for the fifth measure, and the note value string pattern B is arranged as it is for the sixth measure. For the 7th measure, the tone value pattern B is inverted in time series, and for the 8th measure, when the average value V of the note values is 240 or less, one half note is placed. When V is larger than 240, one note is arranged at 240. Then, the tone value strings of each measure set in this way are connected in time series to generate a note value sequence consisting of eight measures.

  Next, a chord suitable for the melody is extracted from the chord progression set by the user or the melody of two measures input by the user, and the chord progression suitable for the progression of the chord for the two measures is set. A chord is assigned to each measure (step T3).

  After a chord is assigned to each measure as described above, the pitch adjustment means 36 adjusts the pitch to a pitch suitable for the chord assigned to that measure (step T4). In this pitch adjustment, in the third measure, the first sound in the melody corresponding to each chord is converted into a constituent sound of the chord, and as shown in FIG. In addition, the “do” sound of the eighth note is converted into the “mi” sound. In the 4th measure, if the average value V of the note value is less than 240, the 3rd higher or lower pitch is arranged according to the pitch of the 3rd measure. Arranges the note at a position where the pitch is different from the first beat twice. For the second and fourth beats, a sound that is twice lower and a sound that is two times higher are arranged depending on the pitch of the first and third beats. On the other hand, if the average value V of the note values is greater than 240, the sound that is two degrees below or two degrees above the first two beats in the order of the last two notes in the third measure will be used. A high note is arranged, and for the third beat, a note is arranged at a pitch two times away from the first beat. For the 8th bar, one whole note is set when the average note value V> 240, and one half note is arranged when the average note value V ≦ 240. At this time, the pitch is the main tone.

  And after arranging the pitch of each measure in this way, the subtone included in the whole melody is converted into the internal tone.

  Finally, a backing suitable for the music genre input by the user is generated (step T5), and these are combined to generate music information with backing (step T6), which is then sent to the client computer 2. Transmit (step T7).

  When the client computer 2 receives the music information (step S3), the client computer 2 stores the music information in the storage unit and outputs the sound via the output means 23 such as a speaker (step S4).

  As described above, according to the above embodiment, a tone value pattern is extracted using a melody composed of two measures input by a user, and the extracted note value sequence pattern is modified. By connecting the melody to the received melody, a melody consisting of a plurality of measures is generated, so that it is possible to generate music information utilizing user preferences and sensibilities. In addition, by inputting partial melodies instead of inputting all melodies as in the past, music information consisting of 8 bars can be generated. The music information suitable for can be generated. In addition, other melodies are generated in consideration of the input melody value string pattern and connected to each other, so that music information having a musical flow can be generated.

  Furthermore, in the above embodiment, the chord progression is set for the input melody, and the pitch of each note value of the melody composed of a plurality of connected measures is set to the pitch in each chord assigned by the chord progression. Since the pitch is adjusted to be high, music information having a natural pitch that matches the chord progression can be generated.

  In addition, this invention is not limited to the said embodiment, It can implement in a various aspect.

  For example, in the above embodiment, the user inputs a melody of two measures and generates music information consisting of eight measures based on this note value sequence pattern. Alternatively, music information may be generated by extracting a sound value string pattern at an arbitrary break.

  In the above embodiment, each tone value string pattern is generated based on the change rule after the third measure. However, the present invention is not limited to this change rule, and is based on various change rules. A sound value string pattern can be generated.

  Furthermore, in the above embodiment, the chord progression is set after concatenating the note value string patterns, and then the pitch is adjusted. However, when the note value string patterns are concatenated, it is set in advance. It is also possible to adjust the pitch with reference to the chord progression, and then connect the note value string patterns.

  In addition, when the user inputs a melody in the above embodiment, the notes are arranged on the staff. However, the present invention is not limited to this. For example, the voice input from a humming or a musical instrument or the above-mentioned patent document. It may be possible to input a partial melody based on music generated from the text or image that is generated.

The block diagram of the music information generation system in one embodiment of this invention Functional block diagram in the same form Example of client computer input screen in the same form The figure which shows the melody of the input two measures in the form The figure which shows the change rule of each measure in the form The figure which shows the melody of the eight measures generated in the same form Diagram showing chord progression in the same form The figure which shows the rule regarding arrangement | positioning of the target sound in the form The figure which shows arrangement | positioning of the target sound of the 3rd bar in the form The figure which shows arrangement | positioning of the sound of the 4th bar in the form The figure which shows the music information of 8 measures produced | generated in the same form The figure which shows the flowchart of the music information generation system in the form

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Music information generation system 2 ... Client computer 3 ... Server computer 21 ... Melody reception means 22 ... Music type setting means 23 ... Output means 31 ... Sound value string pattern extraction Means 32 ... Pattern changing means 33 ... Change rule storage means 34 ... Connecting means 35 ... Chord progression setting means 36 ... Pitch adjustment means 37 ... Music information generating means

Claims (6)

  In a music information generation system for generating a multi-measure melody from a melody inputted by a user, a pattern of a note value string constituting the melody is extracted from the input melody, and the extracted note value string pattern is extracted A music information generation system characterized in that a melody consisting of a plurality of measures is generated by connecting a note value string with a change to the received melody. In a music information generating system that includes a melody accepting unit that accepts a melody input from a user, and that automatically generates a melody of a plurality of measures from the melody accepted by the melody accepting unit,
A note value string pattern extracting means for extracting a pattern of a note value string constituting the melody from the melody received by the melody receiving means;
A change rule storage means for storing a change rule for making a change to the note value string pattern;
Pattern change means for generating a sound value string pattern obtained by changing the sound value string with reference to the change rule for the sound value string pattern extracted by the sound value string pattern extracting means;
A music information generating system comprising: a melody linking unit for linking the tone value string patterns changed by the pattern changing unit.   The music information generation system according to claim 1 or 2, further comprising a chord progression setting means for setting a chord progression for the received melody, wherein a pitch of a note value in each measure is set in each constituent chord assigned by the chord progression. A music information generation system provided with a pitch adjusting means for adjusting to a pitch.   In the music information generation method for generating a melody composed of a plurality of measures using a melody input by a user, a pattern of a note value sequence constituting the melody is extracted from the input melody, and the extracted note value A music information generation method characterized in that a melody composed of a plurality of measures is generated by connecting a note value string obtained by changing a pattern of a string to the received melody. In a music information generation method that automatically generates a multi-measure melody from a melody input by a user,
Extracting a note value string pattern constituting the melody from the input melody;
Storing a change rule for making a change to the note value string pattern, generating a note value string pattern in which the tone value string is changed with reference to the change rule with respect to the sound value string pattern; ,
And a step of connecting the changed tone value string patterns. The music information generation method according to claim 4 or 5,
Setting the chord progression for the entered melody;
Adjusting a pitch of each note value in each measure to a pitch in each constituent chord assigned by the chord progression.

Images