JP4971023B2 - Music generation method, music generation device, music generation system - Google Patents

Description

  The present invention relates to an improvement in music generation technology for generating unique music for various data and information such as telephone numbers, names, and schedules.

  As a background art for generating music, there is an “expression generation method, expression generation device, expression generation system” described in Patent Document 1 below, for example, music is automatically generated based on a telephone number. According to this background art, a music phrase table TA as shown in FIG. 15A is prepared. In this phrase table TA, one music phrase “1-0” to “10-9” is assigned to numbers 0 to 9 corresponding to each digit of the telephone number (10 digits in the illustrated example). It has been. For example, when the third digit of the telephone number is “5”, the corresponding music phrase is “3-5”. Music phrases are generally composed according to the principle of functional harmony. As a result, no matter how the music phrases are arranged, the listener can listen to the music as a unitary piece without any sense of incongruity.

A specific example of music generation is shown in FIG. This is a case where the telephone number is “03-5447-5166”. That is, using the phrase table TA, material data corresponding to the intersection of digits and numbers of the telephone number “03-5447-5166” is extracted and arranged, and the expression is expressed as [1-0]-[ 2-3]-[3-5]-[4-4]-[5-4]-[6-7]-[7-5]-[8-1]-[9-6]-[10- 6]. As described above, according to the invention of Patent Document 1, music can be generated by combining musical phrases corresponding to telephone numbers.
Republished Patent W2003 / 081572 (Fig. 2)

  In the background art described above, when music is generated by information, a list of chords and melodies (phrase tables) in accordance with the harmony method is prepared to compensate for the degree of completion of music that a general user can sufficiently accept. Yes. However, there is an inconvenience that it is necessary to compose a large number of music phrases in accordance with the principle of functional harmony. The above phone number is a combination of 10-digit numbers, but if it is an English name / name, a phrase corresponding to an alphabet / symbol combination is required. If it is a Japanese name / name, It is a combination of kanji, hiragana and katakana, and a huge number of phrases must be prepared.

  Also, music has an image that the sound itself reminds the user intuitively or empirically and is often used idiomatically. For example, a slow phrase by a contrabass is rarely used to depict a bird flying around, even if it is used to depict an elephant walking. On the other hand, a quick and detailed phrase by Piccolo may remind you of an elephant march, even though it may remind you of butterflies and small birds.

  The present invention focuses on the above points, and its purpose is to use a phrase of music expression prepared corresponding to the trigger information regardless of how the trigger information is to be arranged for music generation. It is to generate music with a chord progression that is musically comparable. Another object is to grasp specific information contents of a series of trigger information by listening to generated music by assigning musical features to individual trigger information in advance.

  In order to achieve the above object, a music generation method according to the present invention uses a phrase database in which phrase data of music expressions are accumulated to generate music that generates unique music corresponding to a series of trigger information to be generated. A chord progression pattern in which a progression rule is set based on a chord function is prepared, and the phrase data of the musical expression is assigned a chord function that the chord to which the phrase belongs is assigned in a specific key. And the chord is associated with the trigger information directly or indirectly, and when the chord is arranged according to the trigger information from the step of acquiring the trigger information, the phrases of the music expression corresponding to the acquired trigger information A phrase of a music expression having a chord function that matches the progression pattern is stored in the phrase data. Selecting from over scan, characterized in that it comprises the step of generating a music by combining the phrases of the selected musical expression.

  A music generation method of another invention is a music generation method for generating unique music corresponding to a series of trigger information that is a target of music generation, using a phrase database in which phrase data of music expressions are accumulated. A chord progression pattern in which a progression rule is set based on the chord function is prepared, and the phrase data of the musical expression is assigned a chord function indicated by a chord to which the phrase belongs in a specific key, and the trigger information A step of acquiring the trigger information, a step of selecting a phrase of music expression corresponding to the acquired trigger information from the phrase database, and a trigger of the phrase of the selected music expression When arranged according to information, the progression of chords between phrases of these musical expressions is A step of determining whether or not the chord progression pattern matches, and if there is a chord progression that is judged not to match the chord progression pattern, the chord progression pattern is matched at least between phrases of the corresponding music expression Selecting and inserting a music phrase having a chord function from the phrase database, and synthesizing the selected and inserted music phrases to generate music that matches the chord progression pattern. It is characterized by.

  One of the main forms is that musical features other than the chord function are added to the phrase data of the musical expression, and by assigning the musical features to individual trigger information, the trigger information is assigned to the trigger information. A phrase of a musical expression having a corresponding musical characteristic is selected from the phrase database. In another form, the musical features other than the chord function include the type (or texture) of the chord to which the phrase belongs, the chord name, the pitch name and octave identifier of the constituent sounds of the phrase, the pitch in the phrase , The number of sounds in the phrase, and / or rhythm.

  The music generation device of the present invention is a music generation device that generates unique music corresponding to a series of trigger information that is a target of music generation, and a chord progression pattern and music in which a progression rule is set based on a chord function. Phrase data of expressions is accumulated, and the chord function indicated by the chord to which the phrase belongs in a specific key is assigned to the phrase data of the music expression, and directly or indirectly correspondence with the trigger information Is added to the chord progression pattern when arranged according to the trigger information from the phrase database to which the trigger information is acquired, the phrase of the musical expression corresponding to the acquired trigger information, A phrase for selecting a phrase of music expression with a chord function from the phrase database -Option means, characterized by comprising a music generation means for generating a music by combining a phrase selected musical expression.

  A music generation device according to another invention is a music generation device that generates unique music corresponding to a series of trigger information that is a target of music generation, and a chord progression pattern in which a progression rule is set based on a chord function, Phrase data of music expression is accumulated, and the chord function indicated by the chord to which the phrase belongs in a specific key is assigned to the phrase data of the music expression, and directly or indirectly corresponding to the trigger information A phrase database to which a relationship is added, trigger information acquisition means for acquiring the trigger information, phrase selection means for selecting a phrase of music expression corresponding to the acquired trigger information from the phrase database, and a phrase of the selected music expression When arranged according to the trigger information, the chord progression between these musical expressions Progress determination means for determining whether or not the chord progression pattern matches, and when there is a chord progression that is determined not to match the chord progression pattern, at least the chord progression pattern between the phrases of the corresponding music expression A phrase insertion means for selecting and inserting a music phrase having a chord function for matching with the phrase database, music for generating music that matches the chord progression pattern by synthesizing the selected and inserted music phrases Generation means.

  One of the main forms includes an assignment table in which musical features other than the chord function are added to the musical expression phrase data, and the musical features are associated with individual trigger information. At the same time, the phrase selecting means selects a phrase of musical expression having a musical characteristic corresponding to the trigger information from the phrase database. In another form, the musical features other than the chord function include the type (or texture) of the chord to which the phrase belongs, the chord name, the pitch name and octave identifier of the constituent sounds of the phrase, the pitch in the phrase Including at least one of the width of the sound, the number of sounds in the phrase, and the rhythm.

  According to still another aspect, the phrase selecting means selects a subsequent phrase based on a predetermined selection criterion set based on the musical feature when there are a plurality of candidates for the subsequent phrase. And In another embodiment, the selection criterion is such that, when a phrase having a different initial sound is included in a plurality of candidates for subsequent phrases, a pitch difference from the last sound of the previous music phrase is minimized. A phrase including such a sound as the first sound is designated as a subsequent phrase. Alternatively, when the initial sounds of a plurality of succeeding phrase candidates are all the same, the selection criterion is followed based on the dynamic or static characteristics of the phrase based on the pitch or number of notes in the phrase. It is characterized by specifying a phrase.

  In another embodiment, transposition information for rewriting a constituent sound of a chord assigned to a specific chord function in a specific key into a constituent sound of a chord indicating the same function in another key is divided into a plurality of functions and keys. Corresponding modulation information table, and phrase rewriting means for rewriting the musical expression phrase data in accordance with an arbitrary key together with the chord function assigned to the phrase based on the modulation information table It is characterized by that.

  A music generation system according to the present invention includes the music generation device according to any one of claims 5 to 12 as a server or a terminal, and provides generated music data to a user through a network. The above and other objects, features and advantages of the present invention will become apparent from the following detailed description and the accompanying drawings.

  The present invention prepares a chord progression pattern in which a progression rule is set based on the chord function, and when a series of trigger information to be generated for music is acquired, between phrases of musical expressions corresponding to individual trigger information The phrase of the music expression is extracted from the phrase database to synthesize music so that the chord progression of the chord matches the chord progression pattern, or a correction phrase is inserted between phrases that do not match the chord progression pattern. It was decided to insert and synthesize music. For this reason, it is possible to generate chord progression music that is not inferior in terms of music by using the prepared musical expression phrases. In addition, by associating the musical features of each musical expression phrase with the individual trigger information in advance, it is possible to recognize the specific contents of the trigger information by listening to the generated music. It is done.

  Hereinafter, the best mode for carrying out the present invention will be described in detail based on examples.

  <Overview> First, an overview of the first embodiment of the present invention will be described with reference to FIGS. 1 and 2. In general, proper chord progression is indispensable for generating music that feels comfortable. Modern music has already established harmony methods, and many music are based on it. For external information (hereinafter referred to as “trigger information”) that is the object of music generation, for example, the numbers “1”, “2”, “3”, and “4”, the types of chords (the relationship between the material = the root and the constituent sounds) ), The chord progression may be inappropriate depending on how the trigger information is arranged. In the present invention, as a method for avoiding such inconvenience, a chord progression rule is used based on the function of the chord.

  Each chord has functions such as tonic, subdominant, subdominant minor, and dominant. In addition, a chord belonging to a tonic is not unnatural even if it proceeds to a chord belonging to any function, and a dominant characteristic is that it sounds unnatural when proceeding to a chord other than a tonic. On the other hand, for example, the chord “Gmajor” has a dominant function in the key C (C major) and has a tonic function in the key G (G major). In the present embodiment, a chord progression control method used for automatic music generation will be described based on the established music theory as described above.

  FIG. 1 shows various conditions for controlling the chord progression of this embodiment. First, as described above, since the chord function changes depending on the key (key), the key is determined. As shown in FIG. 1A, in this embodiment, the key is set to C (C major). Next, as shown in FIG. 1B, there are four types of chord functions to be used: tonic, subdominant, subdominant minor, and dominant. Next, as shown in FIG. 1 (C), the types of chords (textures) used corresponding to the trigger information “1” “2” “3” “4” are respectively □ major (□ M), □ Minor (□ m), □ Minor 7th (□ m7), □ Seventh (□ 7) are assigned. Note that □ in the above notation is an algebraic representation of the root note name. In general, “major” is bright, “minor” is dark, “minor seventh” is static, and “seventh” is dynamic.

  For the four chord functions, four chords are determined from the chords mainly used by the key C (see FIG. 1D) so that the types (textures) do not overlap. In this embodiment, as shown in FIG. 1E, a chord CM is assigned to the tonic function, a chord Dm is assigned to the subdominant function, a chord Fm7 is assigned to the subdominant minor function, and a chord G7 is assigned to the dominant function. In FIG. 1 (D), the chord names shown below the notes are the three chords in the upper row, the four chords (seven chords) in the middle row, and the four chords with change symbols in the lower row. This is the chord name for a typical example. Each of these chords CM, Dm, Fm7, G7 is assigned an appropriate number of phrases of music expression. In the present embodiment, the phrase is a measure unit as will be described later, but it may be larger or smaller than one measure. In other words, a phrase can be regarded as a minimum unit of musical expression. It is difficult to express thoughts and feelings with only individual sounds such as “do” and “le”, but when they are gathered together, they constitute the smallest unit of music expression (minimum element or cell), Furthermore, by combining them, it can be considered that it becomes music as an expression of thought and emotion.

  Next, for the function to which the chord belongs, for example, a chord progression rule is set as shown in FIG. In the illustrated example, the tonic (T) function can proceed to any of a subdominant (SD) function, a subdominant minor (SDM) function, and a dominant (D) function. , SDM function and D function can be progressed. Further, the SDM function can proceed to either the T function or the D function, and the D function can proceed to only the T function. At this time, if the arrangement of the trigger information is “1” → “4” → “2” → “3”, the type of chord □ M assigned to the trigger information “1” (see FIG. 1C) ) Is a chord CM with reference to FIG. 1E and has a T function. Similarly, the chord types assigned to the trigger information “4”, “2”, and “3” are □ 7, □ m, and □ m7, respectively, and the corresponding chords are G7, SD having the D function. Dm having a function and Fm7 having an SDM function. At this time, if the progression of these chord functions is compared with the chord progression pattern shown in FIG. 2 (A-1), the progression from chord G7 to Dm is allowed as shown in FIG. 2 (A-2). I understand that it is not done.

  As described above, since only the four chords shown in FIG. 1 (E) may cause inconvenience in the progression of chords corresponding to all combinations (arrangements) of trigger information, in this embodiment, the type of chord (texture) In order to avoid duplication, four new chords Am, Em7, C7, FM are selected from the chords used in the key C and added to each function as shown in FIG. Then, the number of chord candidates corresponding to the type □ m assigned to the trigger information “2” increases to two: a chord Am belonging to the T function and a chord Dm belonging to the SD function. Here, as shown in FIG. 2 (A-2), from the trigger information “4” (D function), the chord Dm of the SD function is not allowed to proceed, but FIG. 2 (B-2) As shown, the chord Am belonging to the T function is allowed to proceed. Further, the chord Am can be advanced to the chord Fm7 (SDM function) corresponding to the trigger information “3”. Thus, by preparing two sets of chords having four kinds of textures, it is possible to obtain a comfortable chord progression regardless of the order (combination) of trigger information.

  <Specific Example> Next, with reference to FIG. 3 to FIG. 6, an optimal phrase is selected from among a plurality of music phrases belonging to the same function and the same chord group in order to smoothly advance the music. A specific example of the method will be described. When a melody tries to progress from one phrase to the next, the most important thing is generally the pitch relationship between the last note of the previous phrase and the first note of the next phrase. At that time, if the chord progression is appropriate in accordance with the chord progression pattern described above, the sense of discomfort is eliminated by proceeding from the same sound to the same sound or proceeding sequentially (advancing to the adjacent sound in the scale). Can do. On the other hand, in jumping progress (advancing at a width of 3 degrees or more, or advancing to a sound other than the scale sound), the connection between phrases may become unnatural.

  FIG. 3 (A) shows an example of assignment of phrases belonging to the same function and the same chord group. In the illustrated example, a chord Dm and a chord FM belong to the group GSD (subdominant function), and phrases 1 to 3 of music expression are assigned to the chord Dm. Similarly, phrases 1A to 3A are assigned to the chord FM. On the other hand, chords CM, Am, Em7, and C7 belong to the group GT (tonic function), and phrases 4 to 6 are assigned to the chord CM. Also, an appropriate number of music phrases (not shown) are assigned to other chords. The chords belonging to the groups GSD and GT are the same as the example shown in FIG. 2 (B-1).

  Here, an example of the phrase selection method is as follows, from phrases 1 to 3 belonging to the Dm chord group of the subdominant function in C major (key C) to phrases 4 to 6 belonging to the CM chord group of the tonic function. A specific example will be described. First, the properties of the chord name to which each phrase belongs (see Fig. 3 (B)) and the names of the first and last notes that make up the phrase (see Fig. 3 (C)) are stored in advance in the data for each phrase. Add it. Note that the pitch names are numbered to distinguish octaves. FIG. 4 shows the notation of phrases 1 to 6 and their properties. Looking at phrase 1 here, “Dm_D4_A3” is added as property P1. The Dm is a chord name, D4 is a pitch name indicating the first sound “R” of the phrase, and A3 is a pitch name indicating the last sound “La” of the phrase. Similar properties P2 to P6 are added to the other phrases 2 to 6 as well.

  Here, in the case of proceeding from phrase 1 to phrases 4 to 6, the relationship between the last sound “A3” of phrase 1 and the first sounds of phrases 4 to 6 will be compared. The first sound of phrase 4 is “C4”, and the connection with phrase 1 is “A3” → “C4”, which is a jumping progression that goes up three times. Similarly, when proceeding to phrase 5, “A3” → “G3”, which is a sequential progression that goes down twice, and when proceeding to phrase 6, “A3” → “E3”, a jumping progress that goes down four times It is. Therefore, as shown in FIG. 5 (A), it can be seen that from phrase 1 it is most natural to proceed to phrase 5 and then to phrase 4 and phrase 6 in this order. Even in actual hearing, the connection to phrase 5 is very natural and easy to sing, phrase 4 is somewhat unnatural and phrase 6 is unnatural (some people have difficulty singing).

  Similarly, as shown in FIG. 5 (B), it can be determined that the progression from phrase 2 to phrase 4 that progresses sequentially is natural. Also, as shown in FIG. 5C, when proceeding from phrase 3 to phrase 4, it is unnatural when jumping up 5 degrees, and when proceeding to phrase 5, “F3” → “G3” It is a natural progression with no sense of incongruity due to the sequential progression of degrees. However, when proceeding to phrase 6, the pitch is “F3” → “E3”, which is closer to a second minor (a semitone relationship), and is more natural than when proceeding to phrase 5. This can also be determined from the note name of the property. In FIG. 6, phrases 4 to 6 that are optimally progressed from phrases 1 to 3 are collectively shown. If there are multiple candidates that start with the same sound as subsequent phrases, for example, by adding the width of the pitch or the number of sounds in the phrase to the property, the dynamic phrase can be changed from a dynamic phrase to a dynamic phrase. New selection criteria can be added, from phrases to quiet phrases.

  <System Configuration> Next, the system configuration of the present embodiment will be described with reference to FIG. FIG. 7 shows the entire system of this embodiment. In the figure, the music generation server 100 is connected to a large number of clients 300 via the Internet 200. The provider is omitted. The client 300 includes various types such as a PC (personal computer), a mobile phone, and a portable information terminal. Among these, the music generation server 100 has a known general system configuration, and includes a phrase database 102, a trigger information allocation table 103, a chord progression pattern 104, a modulation information table 105, a trigger information acquisition program 108, a phrase. In addition to the extraction program 110, the phrase synthesis program 112, the chord progression determination program 114, the phrase insertion program 116, and the phrase rewriting program 118, a processing device 106 for executing each of the above programs is provided. Although not shown, a display device, a printer, a storage device, a communication device, and the like are also provided.

  The phrase database 102 is a collection of music expression phrase data. For example, the phrase of the music expression as shown in FIG. 4 includes the chord name to which the phrase belongs, the pitch names of the first and last sounds of the phrase. Many properties are collected together with properties including These phrase data are stored in an arbitrary form, for example, classified by function or chord name to which the chords constituting them belong. Next, the trigger information allocation table 103 shows the type (texture) of chords corresponding to each trigger information, and an example is shown in FIG. In the present embodiment, the chord type is assigned in advance to the trigger information. However, when the phrase is directly associated with the trigger information, the trigger information assignment table 103 can be omitted. Moreover, such a trigger information allocation table 103 may be prepared in advance on the music generation server 100 side, or may be specified and generated by the user.

  Next, the chord progression pattern 104 indicates the progression of chords used for music generation. As shown in FIG. 2 (A-1) and (B-1), the chord progression pattern 104 proceeds from each chord function. Possible chord functions are shown, and an appropriate number of chords are assigned to each chord function so that the types do not overlap. The chord progression pattern 104 may include the names of phrases belonging to each chord as shown in FIG. The modulation information table 105 will be described later. The processing device 106 includes a CPU, a RAM, a ROM, and the like, and executes various programs.

  Among the programs, the trigger information acquisition program 108 is a program for acquiring trigger information that is a target of music generation. For example, when a request is made from the client 300 through the Internet 200, a screen for inputting conditions for music generation is transmitted to the client 300 and displayed, or the condition is displayed on a display device (not shown) of the music generation server 100. The input screen is displayed and the input trigger information is acquired.

  The phrase extraction program 110 is a program that searches and extracts the phrase corresponding to the trigger information from the phrase database 102. The phrase extraction program 110 selects a subsequent phrase according to a predetermined selection criterion when there are a plurality of subsequent phrase candidates. The selection criterion is set based on a property (musical feature) added to the phrase of music expression. Specifically, for example, when a plurality of candidates for subsequent phrases include a phrase having a different first sound, the sound having the smallest pitch difference from the last sound of the previous music phrase is selected as the first. For example, a phrase included as a sound is designated as a subsequent phrase. Also, if the initial sounds of multiple candidates for the subsequent phrase are all the same, the dynamic phrase is changed from a dynamic phrase to a dynamic phrase based on properties such as the pitch of the pitch and the number of sounds in the phrase. Subsequent phrases may be selected based on the characteristics of the phrase so as to move from to a static phrase. Such selection criteria may be incorporated in the phrase extraction program 110 in advance, or may be arbitrarily set by the client 300.

  The phrase synthesis program 112 is a program that synthesizes the extracted phrases to generate music. In addition, if necessary, a program for providing the generated music to the client 300 in the form of a score or playing the generated music may be provided. The chord progression determination program 114, the phrase insertion program 116, and the phrase rewriting program 118 will be described later.

  <Operation>... Next, the operation of this embodiment will be described. In the music generation server 100, the trigger information acquisition program 108 is executed by the processing device 106, and a screen for inputting the trigger information is displayed on the music generation server 100 side or the client 300 side. When a desired music key or trigger information is input according to the displayed input screen, the phrase extraction program 110 is executed by the processing device 106 in the music generation server 100. The phrase extraction program 110 refers to the trigger information allocation table 103 and the chord progression pattern 104 and corresponds to individual trigger information “1” “4” “2” “3” included in a series of numbers, for example, “1423”. A music phrase that matches the chord progression pattern 104 is extracted from the phrase database 102. At this time, when there are a plurality of phrase candidates, the phrase extraction program 110 refers to the properties of the music phrase and selects a phrase that matches the selection criteria described above. As described above, as described above, for example, the selection may be made such that the difference in pitch between the last sound of the previous phrase and the first sound of the subsequent phrase is minimized. The selection may be made according to other selection criteria.

  When a phrase corresponding to the trigger information is selected, in the music generation server 100, the phrase synthesis program 112 is executed by the processing device 106, and the selected phrase is synthesized as a series of music. Note that the phrases may be combined as a file that can be played back on the music generation server 100 or the client 300 side, for example, MIDI data or WAV file, or a plurality of WAV or MIDI files may be consecutively arranged in a predetermined order. Then, the program may be played back.

Thus, according to the first embodiment, there are the following effects.
(1) A chord progression pattern 104 in which a progression rule is set based on the chord function is prepared, and when a series of trigger information is acquired, it corresponds to each trigger information and matches the chord progression pattern 104 It is assumed that a simple musical expression phrase is extracted from the phrase database 102 to synthesize music. For this reason, it is possible to generate chord progression music that is not musically inferior using the prepared musical expression phrases.
(2) A property including the musical characteristics of the phrase is added to the phrase data of the music expression, and the trigger information allocation table 103 in which the musical characteristics are associated with each trigger information is prepared. By listening to the music, the contents of a series of trigger information can be grasped. Further, even when there are many types of trigger information, it is possible to generate a chord progression music without a sense of incongruity with a limited number of music phrases without preparing trigger information and music phrases one-on-one.
(3) By preparing two sets of chords having four kinds of textures, it is possible to obtain a comfortable chord progression regardless of the order (combination) of trigger information.
(4) When there are a plurality of subsequent phrase candidates, the phrase extraction program 110 selects a subsequent phrase according to the selection criteria set based on the properties of the music phrase. Even in some cases, phrase selection can be performed smoothly.

  Next, Embodiment 2 of the present invention will be described with reference to FIGS. In addition, the same code | symbol shall be used for the component which is the same as that of Example 1 mentioned above, or respond | corresponds (it is the same also about a following example). In the first embodiment described above, a musical feature is assigned to each trigger information in advance, thereby selecting a phrase that matches the chord progression pattern from among musical phrases belonging to the chord having the musical feature. This embodiment shows an example in which music that matches the chord progression pattern is generated by another method. In the music generation system of this embodiment, the music generation server 100 shown in FIG. 7 includes a chord progression determination program 114 and a phrase insertion program 116. In this embodiment, the music phrases Ph1,... Accumulated in the phrase database 102 are, as shown in FIG. 8B, a specific key as a property of the phrase together with corresponding trigger information and a score. For example, “_t”, “_sd”, “_sdm”, and “_d” are added. That is, in the present embodiment, phrases are assigned one-to-one with respect to individual trigger information.

  Further, as shown in FIG. 8A, the chord progression pattern 104 is set only with the chord function, and no chord is assigned to each function. When the music phrase corresponding to the trigger information is selected from the phrase database 102 by the phrase extraction program 110, the chord progression determination program 114 refers to the property of the music phrase and matches the chord progression pattern 104. It is determined whether or not it has a function. When the chord progression determination program 114 determines that there is a chord progression that does not match the chord progression pattern 104, the phrase insertion program 116 inserts a correction music phrase between the corresponding music phrases. Thus, the chord progression matched with the chord progression pattern 104 is corrected. The music phrase for insertion is also prepared in the phrase database 102.

  FIGS. 8C and 8D show examples of inserting correction phrases. In the example shown in FIG. 8C, when the phrases selected by the trigger information are Ph1, Ph4, Ph2, and Ph3, the chord progression from the phrase Ph4 having the dominant function to the phrase Ph2 having the subdominant function is: The chord progression pattern 104 is not permitted. In this case, the phrase insertion program 116 extracts the insertion phrase SPh1 having a tonic function from the phrase database 102 and inserts it between the phrases Ph4 and Ph2. Alternatively, as shown in FIG. 8 (D), inserted phrases SPh1, SPh2, belonging to a desired function, not only between phrases that are not permitted to proceed, but also between all phrases selected corresponding to the trigger information. A music phrase according to the chord progression pattern 104 may be synthesized by inserting SPh3,.

  The insertion phrases SPh1 to SPh3 may be prepared separately from the phrase corresponding to the trigger information, or a phrase having a function suitable for insertion is extracted from the phrases corresponding to the trigger information. May be used instead. As described above, according to the present embodiment, when the phrase selected corresponding to the trigger information does not match the chord progression pattern 104, the chord progression pattern 104 is matched at least between the corresponding phrases. Since the correction phrase having the chord function is inserted, it is possible to generate chord progression music according to a predetermined rule. In addition, a parameter indicating the chord function for a specific key is added to the properties of the phrase data of the music expression, so the desired chord progression rules can be applied to randomly collected music phrases. It becomes.

  Next, Embodiment 3 of the present invention will be described with reference to FIG. In the first embodiment described above, one musical feature (chord type = texture) is assigned to each trigger information that is a target of music generation. However, in this embodiment, two types of musical information are assigned. This is an example in which features are assigned to trigger information. FIG. 9A is a diagram showing a trigger information allocation table 103A showing the correspondence between trigger information and musical features, FIG. 9B is a diagram showing an example of the arrangement of trigger information, and FIG. 9C is trigger information. It is an example of the music produced | generated corresponding to.

  In this embodiment, visual information is used as trigger information. As shown in FIG. 9A, three types of trigger information TA1 to TA3 are divided by color for an ellipse, and for each color, a bright chord (□ M), a dark chord (□ m), and an intermediate chord A chord type (texture) such as (□ sus4) is assigned. Similarly, three types of chords are assigned to the square trigger information TB1 to TB3, the triangle trigger information TC1 to TC3, and the star trigger information TD1 to TD3. Furthermore, the rhythm of the musical feature is also assigned to the shape of the trigger information. In the example shown in the figure, the rhythm changes from “coarse” to “fine” as it goes from an ellipse to a star. As in the first embodiment, a necessary number of chords are assigned to chord functions used for music generation so that the types do not overlap. Or you may make it allocate one music phrase with respect to one trigger information (for example, trigger TA1) like the said Example 2. FIG. In any case, as shown in the first or second embodiment, a chord progression pattern (not shown) based on the chord function is prepared, so that the chord progression matches the chord progression pattern. Music phrases are synthesized.

  As shown in FIG. 9B, if the trigger information is arranged in the order of TD2, TA1, TC3, TB2, and TA3, the chord type and rhythm are combined as shown in FIG. 9C. Music is generated. The note notation in FIG. 9C is a rhythm score. When synthesizing a music phrase corresponding to the trigger information, a phrase that matches the chord progression pattern may be selected from a plurality of phrase candidates as in the first embodiment. If necessary, a correction phrase that satisfies the chord progression pattern may be inserted in a portion where unacceptable chord progression occurs as in the second embodiment. Thus, according to the present embodiment, it is possible to generate music along a certain chord progression even for visual trigger information. Further, since the trigger information is associated with the rhythm in addition to the chord type, the generated music can be prevented from becoming monotonous.

  Next, Embodiment 4 of the present invention will be described with reference to FIGS. In the first embodiment described above, the phrase data of music expression is prepared for a specific key, but this embodiment can cope with transposition by rewriting properties such as the chord function of the phrase data. It is an example. FIG. 10A is a diagram showing the relationship between keys and chords, FIG. 10B is a diagram showing an example of a modulation information table, and FIG. 10C is an explanatory diagram showing an example of phrase rewriting using modulation. As described above, to which chord function a specific chord belongs depends on the selected key. It is very simply based on the law that when a key is raised by a semitone, all of the constituent sounds of the chord having the same function are also raised by a semitone. FIG. 10 (A) schematically shows this relationship, in which one turn composed of twelve kinds of pitch names is one octave, and adjacent ones represent a relationship of different semitones.

  Using this rule, for example, a modulation information table 105 as shown in FIG. 10B is prepared in the music generation server 100. In the illustrated example, the modulation information table 105 includes modulation information for rewriting a chord component sound assigned to a specific chord function in a specific key into a chord component sound indicating the same function in another key. A list of four functions and 12 keys is displayed. As an example, the chord CM is assigned to the tonic function in the key C. On the other hand, if the key is raised halfway from the key: C and becomes the key: C # (= D ♭), there is C # M (= D ♭ M) as a chord indicating the tonic function in the key. This chord is composed of sounds obtained by shifting all the constituent sounds of the chord CM of the same function in the key C by a semitone. Note that the chord shown in FIG. 10B is a typical example.

  In addition, the music generation server 100 of this embodiment is provided with a phrase rewriting program 118 for rewriting a phrase of music expression based on the modulation information table 105. As an example, as shown in FIG. 10C, a musical expression phrase PhX to which a property Pc indicating “key: tonic function in C” is added is prepared, and another key (not shown) is prepared by the client 300 or the like. In the example, a case will be described in which a key five times higher: G) is designated. At this time, the phrase rewriting program 118 refers to the modulation information table 105 and rewrites the property Pc to the property Pg indicating “key: G tonic function” and also rewrites the musical score data. As described above, according to the present embodiment, it is possible to efficiently use one material by automatically rewriting the property related to the chord function added to the phrase according to the specified (or selected) key. it can. Further, when the sound material is, for example, MIDI data, the phrase database 102 can be more easily enriched by automatically transposing the melody.

In addition, this invention is not limited to the Example mentioned above, A various change can be added in the range which does not deviate from the summary of this invention. For example, the following are also included.
(1) In the above embodiment, one measure is one phrase, but this is because the measure is a rhythm cycle and is easy to grasp as a group of musical expressions, and is not limited to this. Long and short settings may be made as appropriate.
(2) In the first and second embodiments, a relatively simple example with four chord functions has been shown. However, the present invention can also be applied to the progression of chords using chord functions of less or more.

(3) In the above embodiment, music of a predetermined length is generated using a fixed number of trigger information. However, in the present invention, it is not necessary to predetermine the length of music, and information that continues continuously Can continue to generate music. For example, as shown in FIG. 11A, it is also possible to generate music that lasts almost forever based on the DNA base sequence.
(4) The assignment of chords to the chord functions shown in the first embodiment is also an example, and as shown in FIG. 11 (B), by increasing the number of chords as appropriate, richer and more varied music is generated. You may make it do.
(5) In the above embodiment, a general chord progression rule is applied, but it is also possible to set a user's own chord progression rule as in the chord progression pattern shown in FIG. is there.

  (6) The trigger information acquisition method shown in the above embodiment is also an example. As shown in FIG. 12 (A), the trigger information displayed on the game machine 310 can be displayed in any order by the user using a touch panel or the like. The trigger information may be acquired by selecting with or by automatically recognizing the shape drawn by the user. Alternatively, as shown in FIG. 12B, an arbitrary object (object, human face, map, etc.) is acquired as visual trigger information using the camera 322 of the mobile phone 320 or the like. Also good. Furthermore, as shown in FIG. 12C, a chord progression pattern set 340 composed of a plurality of chord progression patterns set for each acceleration is prepared, and the motion of the object is photographed using the video camera 330. The chord progression pattern to be referred to may be switched in accordance with the acceleration.

  (7) In the third embodiment, musical features are assigned to shapes and colors of graphics, but this is also an example, and various types of trigger information can be applied. For example, the example shown in FIG. 13 is an example in which the rhythm of the incoming melody changes depending on the distance between the mobile phone 360, 370, 380 of the other party who has called the mobile phone 350 and the mobile phone 350. . According to the illustrated example, if the distance is within 100 m, it is called a fine rhythm that is not busy, if the distance is between 1000 m (1 km) to 100 m, it is a normal rhythm, and if the distance is 10 km or more, it is called a relaxed rhythm. Condition. The distance between the mobile phone 350 and the mobile phones 360, 370, and 380 can be obtained using the GPS 390.

  Or as shown in FIG. 14, you may make it utilize this invention as a means to notify the processing precision of a workpiece. For example, if the surface of the workpiece is rough, assign chords □ dim7, □ m7 (-5), □ m + 7 with a low degree of consonance, apply a fine rhythm, and the surface of the workpiece is almost flat For example, chords □ M and □ m having a high degree of harmony are assigned and a relaxed rhythm is applied. By applying the smoothness of the processed surface to the harmony and rhythm of the chord in this way, the processing accuracy can be known by music, so that the work can be performed without keeping an eye on the workpiece. Also, unlike the alarm, the music changes gradually, making it easier to grasp the processing status. The assignment of musical features to individual trigger information as described above may be set in advance on the music generation server 100 side, or may be set independently by the user.

  (9) The present invention is widely used not only for entertainment such as “music building blocks” in various forms, but also for information systems for visually impaired persons, automatic music generation systems for specific images linked with image recognition technology, etc. Can be applied. Also, music may be generated according to the emotion by digitizing the emotion. In addition to the emotions described above, the present invention can be applied to general information that can be quantified, such as sensitivity information, biological information, and environmental information, and music generated using these information as a trigger is generated in real time. May be.

  According to the present invention, when a chord progression pattern in which a progression rule is set based on the chord function is prepared and a series of trigger information is acquired, the chord progression between phrases of the music expression corresponding to each trigger information is The music expression phrase is extracted from the phrase database so as to match the chord progression pattern, and the music is synthesized, or the correction phrase is inserted between the phrases that do not match the chord progression pattern. It was decided to synthesize. For this reason, it is applicable to the use which produces | generates the original music corresponding to trigger information. In particular, it is suitable for the purpose of grasping the specific information content of the trigger information from the generated music by associating the musical characteristics of the phrase of music expression with the individual trigger information in advance.

It is a figure which shows Example 1 of this invention, (A) is a figure which shows the allocation of a key (tone), (B) is a figure which shows the chord function to be used, (C) is a trigger information and the kind of chord (texture) (D) is a diagram showing typical chords in the key C, and (E) is a diagram showing an example of chord assignment to each chord function. It is a figure which shows the said Example 1, (A-1) and (B-1) are figures which show the example of a chord progression pattern, (A-2) and (B-2) are when making it respond | correspond to trigger information It is a figure which shows the propriety of a chord progression. It is a figure which shows the said Example 1, (A) is a figure which shows the example of phrase assignment for every chord function, (B) is a figure which shows a chord name, (C) is an example of the scale of key C (C major) It is a figure which shows the note name at the time of setting. It is a figure which shows the example of progress between the phrases of the said Example 1. FIG. It is a figure which shows the example of progress between the phrases of the said Example 1. FIG. It is a figure which shows the example of progress between the phrases of the said Example 1. FIG. It is a block diagram which shows an example of the system configuration | structure of this invention. It is a figure which shows Example 2 of this invention, (A) is a figure which shows an example of a chord progression pattern, (B) is a figure which shows an example of a music phrase, (C) and (D) are the phrases for correction. It is a figure which shows the example of insertion. It is a figure which shows Example 3 of this invention, (A) is a figure which shows the example of allocation of the trigger information and the musical feature of a chord, (B) is a figure which shows the example of arrangement | sequence of trigger information, (C) is produced | generated. FIG. It is a figure which shows Example 4 of this invention, (A) is a figure which shows the relationship between a key and a chord, (B) is a figure which shows an example of a modulation information table, (C) is an example of phrase rewriting using a modulation. It is explanatory drawing which shows. It is a figure which shows the other Example of this invention. It is a figure which shows the other Example of this invention. It is a figure which shows the other Example of this invention. It is a figure which shows the other Example of this invention. It is a figure which shows an example of background art.

Explanation of symbols

100: Music generation server 102: Phrase database 103, 103A: Trigger information allocation table 104: Chord progression pattern 105: Transposition information table 106: Processing device 108: Trigger information acquisition program 110: Phrase extraction program 112: Phrase synthesis program 114: Chord Progress determination program 116: Phrase insertion program 118: Phrase rewriting program 200: Internet 300: Client 310: Game machine 320: Mobile phone 322: Camera 330: Video camera 340: Chord progression pattern set 350, 360, 370, 380: Mobile phone 390: GPS
GSD, GT: Groups P1-P6, Pc, Pg: Properties Ph1-Ph4, PhX: Phrases SPh1-SPh3: Inserted phrases TA1-TA3, TB1-TB3, TC1-TC3, TD1-TD3: Trigger information

Claims (13)

A music generation method for generating unique music corresponding to a series of trigger information to be generated by using a phrase database in which phrase data of music expressions are accumulated,
A chord progression pattern with a progression rule set based on the chord function is prepared,
The phrase data of the music expression is assigned a chord function indicated in a specific key by the chord to which the phrase belongs, and is directly or indirectly associated with the trigger information,
Obtaining the trigger information;
A step of selecting, from the phrase database, a phrase of music expression having a chord function that matches the chord progression pattern when arranged according to the trigger information, from phrases of music expression corresponding to the acquired trigger information. ,
Synthesizing phrases of the selected musical expression to generate music;
A method of generating music, comprising: A music generation method for generating unique music corresponding to a series of trigger information to be generated by using a phrase database in which phrase data of music expressions are accumulated,
A chord progression pattern with a progression rule set based on the chord function is prepared,
The phrase data of the music expression is assigned a chord function indicated in a specific key by the chord to which the phrase belongs, and is directly or indirectly associated with the trigger information,
Obtaining the trigger information;
Selecting a phrase of music expression corresponding to the acquired trigger information from the phrase database;
Determining whether or not the chord progression between the phrases of the music expression matches the chord progression pattern when the phrases of the selected music expression are arranged according to the trigger information;
If there is a chord progression that is determined not to match the chord progression pattern, a music phrase having a chord function for matching the chord progression pattern is at least between phrases of the corresponding music expression from the phrase database. Step to select and insert,
Synthesizing the selected and inserted music phrases to generate music that matches the chord progression pattern;
A method of generating music, comprising: Musical features other than the chord function are added to the phrase data of the music expression,
The musical expression phrase having a musical feature corresponding to the trigger information is selected from the phrase database by assigning the musical feature to individual trigger information. 2. The music generation method according to 2.   Musical characteristics other than the chord function include the type (or texture) of the chord to which the phrase belongs, the chord name, the pitch name and octave identifier of the constituent sound of the phrase, the pitch of the pitch in the phrase, the phrase The music generation method according to claim 3, further comprising at least one of the number of sounds and the rhythm. A music generation device that generates unique music corresponding to a series of trigger information to be generated,
Chord progression pattern with progression rules set based on the chord function,
Phrase data of music expression is accumulated, and the chord function indicated by the chord to which the phrase belongs in a specific key is assigned to the phrase data of the music expression, and directly or indirectly corresponding to the trigger information Phrase database with relations added,
Trigger information acquisition means for acquiring the trigger information;
Phrase for selecting from the phrase database a phrase of music expression having a chord function that matches the chord progression pattern when arranged according to the trigger information, from phrases of music expression corresponding to the acquired trigger information Selection means,
Music generation means for generating music by synthesizing phrases of selected music expressions;
A music generation apparatus comprising: A music generation device that generates unique music corresponding to a series of trigger information to be generated,
Chord progression pattern with progression rules set based on the chord function,
Phrase data of music expression is accumulated, and the chord function indicated by the chord to which the phrase belongs in a specific key is assigned to the phrase data of the music expression, and directly or indirectly corresponding to the trigger information Phrase database with relations added,
Trigger information acquisition means for acquiring the trigger information;
Phrase selection means for selecting a phrase of music expression corresponding to the acquired trigger information from the phrase database,
A progress judging means for judging whether or not the chord progression between the phrases of the music expression matches the chord progression pattern when the phrases of the selected music expression are arranged according to the trigger information;
When there is a chord progression determined not to match the chord progression pattern, a music phrase having a chord function for matching the chord progression pattern is selected from the phrase database at least between phrases of the corresponding music expression Phrase insertion means to insert,
Music generating means for synthesizing the selected and inserted music phrases to generate music that matches the chord progression pattern;
A music generation apparatus comprising: Musical features other than the chord function are added to the phrase data of the music expression,
An allocation table in which the musical features are associated with individual trigger information;
With
The music generating device according to claim 5 or 6, wherein the phrase selecting means selects a phrase of a music expression having a musical feature corresponding to the trigger information from the phrase database.   Musical characteristics other than the chord function include the type (or texture) of the chord to which the phrase belongs, the chord name, the pitch name and octave identifier of the constituent sound of the phrase, the pitch of the pitch in the phrase, the phrase The music generation device according to claim 7, wherein the music generation device includes at least one of a number of sounds and a rhythm. The phrase selecting means is:
9. The music generation apparatus according to claim 8, wherein when there are a plurality of subsequent phrase candidates, the subsequent phrase is selected based on a predetermined selection criterion set based on the musical feature. The selection criteria are:
If a phrase with a different initial sound is included in a plurality of candidates for the subsequent phrase, a phrase including the sound with the smallest pitch difference from the last sound of the previous music phrase as the first sound is selected. The music generation device according to claim 9, wherein the music generation device is designated as a subsequent phrase. The selection criteria are:
When the initial sounds of a plurality of subsequent phrase candidates are all the same, the subsequent phrase is designated based on the dynamic or static characteristics of the phrase based on the pitch or number of pitches in the phrase. The music generation device according to claim 9. Transposition that shows the modulation information for rewriting a chord component assigned to a specific chord function in a specific key to a chord component that shows the same function in another key, corresponding to multiple functions and tones Information table,
Phrase rewriting means for rewriting the phrase data of the music expression in accordance with an arbitrary key together with the chord function assigned to the phrase based on the modulation information table;
The music generation device according to claim 5, further comprising:   13. A music generation system comprising the music generation device according to claim 5 as a server or a terminal, and providing the generated music data to a user through a network.

Images