JP2009020323A - Automatic music composition device, automatic music composition method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an automatic music composition device and an automatic music composition method, enabling a user hardly having knowledge for music to enjoy arrangement of a base piece (original piece) while changing the degree of arrangement. <P>SOLUTION: An original piece storage means 105 stores a score of the original piece. A numerical string generation means 109 generates a numerical string. A transition probability storage means 106 stores two or more kinds of transition probability data of sounds in the original piece. A playing sound determination means 110 determines a sound to be played next from each sound of the score of the original piece stored in the original piece storage means 105 based on transition probability data selected by input from an input device 101 of the two or more kinds of transition probability data stored in the transition probability storage means and the numerical string generated in the numerical string generation means 109, and generates an arranged score 107-1. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an apparatus and method for automatically creating music, and more particularly to an automatic music apparatus and an automatic music method that allow a user to select the degree of arrangement of a base music (original music).

  As an automatic composer, it memorizes the sound sequence data that expresses each sound as a melody material with pronunciation timing, pitch, volume, and length, and as chord progression (chord sequence) and scale as music background information There has been proposed an automatic music composition device that generates a melody by inputting (scale type and key) and changing each pitch of the material based on input music background information (see, for example, Patent Document 1).

JP-A-9-120283

  Since the automatic musical composition apparatus described in Patent Document 1 changes each pitch of a certain melody material, it can create a musical piece in which the original musical piece is arranged. However, each pitch is changed according to music background information such as chord progression (chord sequence) and scale (scale type and key), so if you do not have some degree of music knowledge, It was difficult to adjust the degree.

  An object of the present invention is to provide an automatic music composition device and an automatic music composition method that enable even a user who has little knowledge about music to enjoy the arrangement of a base music (original music) while changing the degree of arrangement. It is to provide.

  The first automatic music composition apparatus of the present invention comprises an original music storage means for storing a musical score expressing each sound of a sound string constituting the original music by a pronunciation timing, a pitch, a loudness, and a sound length; Transition probability storage means for storing a plurality of types of sound transition probability data in a tune, a numerical value sequence generating means for generating a sequence of numerical values, and an input device among a plurality of types of transition probability data stored in the transition probability storage means Sound to be played next from each sound of the original musical score stored in the original music storage means on the basis of the transition probability data selected by the input from the numerical value string generated by the numerical value string generating means Performance sound determining means for determining the performance sound.

  According to a first automatic music composition method of the present invention, an original music storage means for storing a musical score expressing each sound of a sound string constituting the original music by a pronunciation timing, a pitch, a loudness, and a sound length; A self-composing music composition method for creating a music composition using a computer comprising transition probability storage means for storing a plurality of types of sound transition probability data in a music composition and an input device, comprising: a) the computer from the input device; A step of inputting a selection parameter; b) a step in which the computer generates a sequence of numerical values; and c) the input device among a plurality of types of transition probability data stored in the transition probability storage means. Next, based on the transition probability data selected by the selection parameter inputted from the above and the generated string of numerical values, the next performance should be performed from each sound of the original music score stored in the original music storage means. And determining the sound.

  The first program of the present invention includes an original music storage means for storing a musical score expressing each sound of a sound string constituting the original music by a sounding timing, a pitch, a loudness, and a sound length; A) a step of inputting a selection parameter from the input device to a computer provided with a transition probability storage means for storing a plurality of types of sound transition probability data and an input device; and b) generating a sequence of numerical values; c) Based on the transition probability data selected from the plurality of types of transition probability data stored in the transition probability storage means by the selection parameter input from the input device and the generated numerical sequence, Determining a sound to be played next from each sound of the score of the original music stored in the music storage means.

  ADVANTAGE OF THE INVENTION According to this invention, even if it is a user who has little knowledge about music, the automatic music apparatus and automatic music method which can enjoy the arrangement of the music (original music) used as a base, changing the degree of arrangement, and A program is obtained. The reason is that the degree of arrangement can be changed by selecting the transition probability data to be used.

  Referring to FIG. 1, an automatic composition apparatus according to an embodiment of the present invention includes an input device 101 such as a keyboard, a display device 102 such as a liquid crystal display, a printing device 103 such as a printer, and a magnetic disk. Numerical value string storage means 104, original music storage means 105, transition probability storage means 106 and arrangement music storage means 107, acoustic element 108 such as a speaker, numerical value string generation means 109 as functional means, performance sound determination means 110, The musical tone generating means 111 and the control means 112 are configured.

  The original music storage means 105 stores the score 105-1 of the music that serves as the base (original music). The score 105-1 is sound string data in which each sound of the sound string constituting the original music is expressed by four attributes of sounding timing, pitch, sound length, and sound volume.

  The transition probability storage means 106 stores a plurality of types of sound transition probabilities in the original musical score 105-1. Here, the sound transition probability means that when a sound state is defined by at least one of the three attributes (pitch, length, and loudness) excluding the sound generation timing, It means the transition probability to the next state. For example, in the original music, if the sound that appears next to the sound of Do is 3 times for Mi and 7 times for So, the transition probability from Do to Mi is 0.3, and the transition probability from Do to So is 0. .7, transition probability from sound to sound other than Mi and So is 0. This example is transition probability data that defines the state of a sound with only one attribute (pitch), but two attributes (for example, pitch and volume, pitch and pitch, or pitch and pitch) You can use transition probability data to define the state of the sound in terms of the size of the sound, and indicate the transition probability between those states, or the sound state with the three attributes (pitch, loudness, and length) You may make it use the transition probability data which define and show the transition probability between those states.

  In the case of the present embodiment, the transition probability storage means 106 is a first transition probability unit that stores transition probability data in an original song in a sound state defined by three attributes of pitch, sound volume, and sound length. 106-1 and the transition probability data in the original song in the sound state defined by any two of the pitch, the loudness, and the tone length, and the transition probability in the original song of the remaining one sound attribute The second transition probability unit 106-2 for storing data, the transition probability data for the pitch original music, the transition probability data for the original music volume, and the transition probability data for the pitch original music And a probability unit 106-3.

  Referring to FIG. 2, the first transition probability unit 106-1 includes a state definition table 106-11 in which the sound state is defined in H × V × L ways by a combination of pitch, sound volume, and sound length. In addition, a transition probability table 106-12 that stores transition probabilities between the states defined in the state definition table 106-1 is stored.

  Referring to FIG. 3, the second transition probability unit 106-2 includes a state definition table 106-21 in which sound states are defined in H × V combinations by combinations of pitches and loudness, and this state definition. A transition probability table 106-22 that stores transition probabilities between the states defined in the table 106-21, a state definition table 106-23 that defines a sound state by sound length, and this state definition table 106-23 The transition probability table 106-24 which memorize | stored the transition probability between each defined state is memorize | stored.

  Referring to FIG. 4, the third transition probability unit 106-3 includes a state definition table 106-31 in which the sound state is defined in H pitches and each state defined in the state definition table 106-31. Transition probability table 106-32 that stores transition probabilities between states, state definition table 106-33 that defines the state of sound as V in terms of loudness, and each of the states defined in this state definition table 106-33 Transition probability table 106-34 that stores transition probabilities between states, state definition table 106-35 that defines sound states in L lengths, and between states defined in this state definition table 106-35 The transition probability table 106-36 that stores the transition probabilities is stored.

A typical format example of the transition probability tables 106-12, 106-22, 106-24, 106-32, 106-34 and 106-36 is shown in FIG. As shown in FIG. 5, when n sound states are defined, the transition probability table includes transition probabilities x _ij (i, j = 1) for states 1 to n for each state 1 to n. ~ N) are stored. Here, in the original music, when there is at least one sound having the state i other than the final sound, the sum of the transition probabilities x _i1 , x _i2 ,..., X _in described in the row of the state i in the transition probability table. Becomes 1. In the original music, when there is no sound other than the final sound in the state i, the sum of the transition probabilities x _i1 , x _i2 ,..., X _in described in the row of the state i in the transition probability table is 0. become.

  Referring to FIG. 1 again, the numerical value sequence generating means 109 generates a numerical value sequence for selecting one candidate from a plurality of candidates for the next sound state determined by the transition probability. The numerical sequence generator 109 can be realized by a random number generator. However, in this embodiment, the numerical sequence generator 109 is configured as a unit that converts a separately provided base sequence into a numerical sequence. The base sequence is formed by polymerizing four types of nucleotides, adenine (A), thymine (T), guanine (G) and cytosine (C), and the arrangement of these four types of nucleotides is as follows. It depends on the species. The degree of difference is small between adjacent species such as humans and monkeys, and large between species that do not have a substantial common ancestor, such as humans and bacteria. In general, in the case of DNA, a base sequence is expressed by arranging four types of alphabets A, T, G, and C. As shown in FIG. 6, the numerical value string generation means 109 converts character information representing a base sequence into a three-digit numerical value (a numerical value from 0 to 63 in decimal notation) for every three characters. .

  The numerical value string storage means 104 stores the numerical value string 104-1 generated by the numerical value string generation means 109.

  The performance sound determination unit 110 is one of the three types of transition probability units 106-1 to 106-3 stored in the transition probability storage unit 106, which is selected by a selection parameter input from the user through the input device 101. Based on the transition probabilities stored in one transition probability section and the numerical value sequence generated by the numerical value sequence generation means 109, each sound of the musical score 105-1 of the original music stored in the original music storage means 105 is next This is a means for determining the pitch, loudness and length of the sound to be played. Note that the sound generation timing is the same as that of the original music.

  Specifically, when the first transition probability unit 106-1 is selected, the performance sound determination means 110 is in the order of sound 1, sound 2, and sound 3 in the original music as shown in FIG. If the sounds are lined up, it should be performed next from the state specified by the pitch, the volume and the length of the sound 1, the transition probability table 106-12 in FIG. 2, and the numerical value column 104-1. The state of the sound 2 ′ (pitch, loudness and length) is determined. Similarly, the sound 3 'to be played next to the sound 2' is determined based on the sound 2 of the immediately preceding original music.

  When the second transition probability unit 106-2 is selected, the performance sound determination unit 110 determines the state specified by the pitch and the loudness of the sound 1 in the original music and the transition probability table 106 in FIG. The sound state (pitch and loudness) to be played next is determined from -22 and the numerical value column 104-1, and the state specified by the sound length of sound 1 and the transition probability in FIG. The state (sound length) of the sound to be played next is determined from the table 106-24 and the numerical value column 104-1, and the sound having the determined pitch, volume and length is determined as a sound. The state of the sound 2 ′ to be played next to 1 is determined as the state (pitch, loudness and pitch). Similarly, the sound 3 'to be played next to the sound 2' is determined based on the sound 2 of the immediately preceding original music.

  Further, when the third transition probability unit 106-3 is selected, the performance sound determination means 110 determines the state specified by the pitch of the sound 1 in the original music, the numerical value of the transition probability table 106-32 in FIG. From the column 104-1, the state (pitch) of the sound to be played next is determined, the state specified by the volume of the sound 1 and the transition probability table 106-34 in FIG. 104-1, the state of the sound to be played next (sound volume) is determined, the state specified by the sound length of sound 1, the transition probability table 106-36 in FIG. -1 is used to determine the state (sound length) of the sound to be played next, and the sound having the determined pitch, volume and length is the sound 2 to be played next to sound 1. Determined as' state (pitch, loudness and pitch). Similarly, the sound 3 'to be played next to the sound 2' is determined based on the sound 2 of the immediately preceding original music.

  Here, as the number of attributes that define the state of the sound increases, the number of state types increases, but the possibility of new states generated by transition probabilities is narrowed down, so the variety of songs to be arranged is also narrowed down It is. Conversely, the smaller the number of attributes that define the state of the sound, the less the type of state, but the greater the possibility of new state combinations generated by transition probabilities. Will also expand. For example, in a certain original music, a sound that appears after a sound A having “pitch 1, sound volume 1, sound length 1” is a sound having “pitch 2, sound volume 2, sound length 2”. If there are only two of the sound C of B and the sound C of “pitch 3, sound volume 3, sound length 3”, when the first transition probability unit 106-1 is selected, the sound to be played next to the sound A Can be selected only from the two types of sound B and sound C. On the other hand, in the original music, the pitch of the sound that appears next to the sound of pitch 1 has three types, pitch 1, pitch 2, and pitch 3, and the sound that appears next to the sound of volume 1. Suppose that the length of the sound that appears next to the sound of size 1, size 2, size 3, and length 1 is three types, length 1, length 2, and length 3. When the third transition probability unit 106-3 is selected, the sound to be played next to the sound A is selected from 3 × 3 × 3 = 27 types of sounds.

  The arrangement storage means 107 stores the performance sound data string determined by the performance sound determination means 110 as an arrangement score 107-1.

  The musical sound generating means 111 is a means for generating a musical sound corresponding to the score from the acoustic element 108, that is, an automatic performance means.

  The input device 101 is used by a user to input data such as selection parameters that specify a transition probability part to be used. The display device 102 and the printing device 103 are used when the generated score is displayed to the user or printed on paper.

  The control unit 112 is a unit that controls the entire apparatus.

  Next, the overall operation of the automatic musical composition apparatus of the present embodiment will be described.

  Referring to FIG. 8, the control unit 112 first inputs a selection parameter from the user through the input device 101 (S101). By this selection parameter, any one transition probability part is designated among the plurality of transition probability parts 106-1 to 106-3 stored in the transition probability storage means 106. As a specific selection method, a selection screen for the arrangement degree of the music to be arranged is displayed on the display device 102. When the user selects the arrangement degree “small” from the selection screen, the first transition probability section When the arrangement degree “medium” is selected, the second transition probability unit 106-2 is selected. When the arrangement degree “large” is selected, the third transition probability unit 106-3 is set. A method of selecting is conceivable. At this time, the control unit 112 sets the selected arrangement degree and the corresponding transition probability unit in the performance sound determination unit 110.

  Next, the control unit 112 generates a numerical sequence from the base sequence separately given by the numerical sequence generation unit 109 and stores it in the numerical sequence storage unit 104 (S102). In this case, a plurality of base sequences are stored in advance in a base sequence storage means (not shown) in association with the name of the organism, a list of organism names is displayed on the display device 102, and the user can select any of the organism names from the list of organism names. It may be configured such that the organism name is selected and the base sequence corresponding to the selected organism name is given to the numerical value string generation means 109.

  Next, the control means 112 outputs the first sound of the original music score 105-1 from the original music storage means 105 to the arrangement music score 107-1 of the arrangement memory means 107 as the first performance sound of the arrangement, and the first The state of the previous sound is set based on the sound (S103). Here, the state of the previous sound is set according to the selected degree of arrangement. Specifically, when the degree of arrangement is “Large”, the pitch, volume, and length of the first sound are independently set as the state of the previous sound, and when the degree of arrangement is “Middle”, the first sound is set. If the combination of pitch and loudness is set as one state of the previous sound, the length is set as another state of the previous sound, and if the degree of arrangement is “low”, the pitch of the first sound The combination of the sound volume and the sound length is set as one state of the immediately preceding sound.

  Next, the control means 112 reads one numerical value from the head of the numerical value string 104-1 stored in the numerical value string storage means 104 (S104), and uses the read numerical value and the set state of the immediately preceding sound as performance sound determining means. 110, the performance sound (pitch, loudness, and length) determined by the performance sound determination means 110 from these information and the information of the set transition probability portion is used to constitute the musical score 107-1. The next sound is stored in the arrangement storage means 107 (S105). Here, the sound generation timing is the same as that of the original music.

  Next, the control means 112 determines whether or not there is an uninput numerical value in the numerical value string 104-1 (S106), and if an uninput numerical value remains, the pitch and the tone of the next sound of the original music are recorded. The state of the immediately preceding sound is set based on the loudness and the sound length (S107). Again, the setting of the state of the immediately preceding sound is performed according to the selected degree of arrangement. If the next sound does not exist in the original music, for example, the process returns to the first sound of the original music and the process is continued. Then, the process returns to step S104 to read a non-input numerical value from the numerical value string 104-1, and based on the read numerical value and the set state of the immediately preceding sound, the processing sound determining means 110 determines the next performance sound. repeat. On the other hand, when all of the numerical value string 104-1 has been input (NO in S106), the control means 112 reads the arrangement score 107-1 stored in the arrangement storage means 107 and inputs it to the musical sound generation means 111. Then, a musical sound corresponding to the score 107-1 is generated from the acoustic element 108 by the musical sound generating means 111 (S108). At the same time, or instead of this, the control means 112 may display the arrangement score 107-1 on the display device 102 or output it from the printing device 103.

  Next, details of step S105 will be described. The performance sound determining means 110 executes the process shown in the flowchart of FIG. 9 in step S105 when the degree of arrangement is “low”, and executes the process shown in the flowchart of FIG. 12 when the degree of arrangement is “medium”. If the degree of arrangement is “large”, the processing shown in the flowchart of FIG. 13 is executed.

First, the operation of the performance sound determining means 110 when the degree of arrangement is “low” will be described with reference to FIG. The performance sound determination means 110 first substitutes the quotient obtained by dividing the input base sequence value by 63 for the variable x _max (S201). The numerical value of the base sequence is a numerical value of 3 digits in quaternary and its value range is 0 to 63 in decimal number, so the value range of the variable x _max is 0 to 1. Next, the performance sound determination means 110 substitutes the number of the set state (pitch, volume, and length) of the immediately preceding sound into the variable i (S202). For example, if the set sound state is state 2, the state number is 2, so i = 2. Then, the performance sound determination means 110 obtains the state of the next performance sound (pitch, volume, length) by calculation (S203) and outputs it (S204).

Details of the processing in step S203 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j that holds the number of the next sound state (pitch, pitch, and length) to 1 (S301), and sets a variable T that holds the cumulative probability value. It is initialized to 0 (S302). Next, referring to the transition probability table 106-12 in FIG. 2, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is equal to or greater than x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The above operation is repeated until the value of the variable T is equal to or greater than x _max, the size of the pitch and the state j the value of the variable T is identified by the value of j at the time of equal to or greater than x _max The sound length is acquired from the state definition table 106-11 in FIG. 2 (step S306). This operation will be described with a simplified example as follows.

Now, assume that the number of states is 8, and the transition probability from state 1 to another state is a value as shown in FIG. At this time, assuming that i = 1, the input value of the base sequence is 20, and x _max = 0.32, when j = 4, since 0.125 + 0.0 + 0.125 + 0.125 ≧ 0.32, the next sound state is state 4 Is determined. If the input value of the base sequence is 45 and x _max = 0.72, when j = 5, 0.125 + 0.0 + 0.125 + 0.125 + 0.5 ≧ 0.72, so the next sound state is determined as state 5. The

Next, the operation of the performance sound determination means 110 when the arrangement degree is “medium” will be described with reference to FIG. The performance sound determination means 110 first substitutes a quotient obtained by dividing the input base sequence value by 63 for a variable x _max (S401). Next, the performance sound determination means 110 substitutes the number of the set state of the previous sound (pitch, volume) into the variable i (S402). Then, the performance sound determination means 110 obtains the state (pitch, volume) of the next performance sound by calculation (S403).

Details of the processing in step S403 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j that holds the number of the next sound state (pitch, volume) to 1 (S301), and initializes a variable T that holds the cumulative probability value to 0. (S302). Next, referring to the transition probability table 106-22 in FIG. 3, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is equal to or greater than x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The size of the above-described operation is repeated until the value of the variable T is equal to or greater than x _max, the state j the value of the variable T is identified by the value of j at the time of equal to or greater than x _max pitch and Is obtained from the state definition table 106-21 in FIG. 3 (step S306).

  Next, the performance sound determination means 110 substitutes the number of the set state (sound length) of the immediately preceding sound for the variable i (S404 in FIG. 12). Then, the performance sound determination means 110 obtains the state (sound length) of the next performance sound by calculation (S405).

Details of the processing in step S405 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j holding the number of the next sound state (sound length) to 1 (S301), and initializes a variable T holding a cumulative probability value to 0 (S302). . Next, referring to the transition probability table 106-24 in FIG. 3, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is equal to or greater than x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The above operation is repeated until the value of the variable T is equal to or greater than x _max, the state of FIG. 3 the durations of the state j the value of the variable T is identified by the value of j at the time of equal to or greater than x _max Obtained from the definition table 106-23 (step S306).

  Finally, the performance sound determination means 110 outputs the pitch and loudness determined in step S403 and the pitch determined in step S405 as the pitch, loudness and length of the sound to be played next. (S406 in FIG. 12).

Next, the operation of the performance sound determining means 110 when the degree of arrangement is “large” will be described with reference to FIG. The performance sound determination means 110 first substitutes a quotient obtained by dividing the input base sequence value by 63 for a variable x _max (S501). Next, the performance sound determination means 110 substitutes the set number of the state (pitch) of the immediately preceding sound for the variable i (S502). Then, the performance sound determination means 110 obtains the state (pitch) of the next performance sound by calculation (S503).

Details of the processing in step S503 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j that holds the number of the next sound state (pitch) to 1 (S301), and initializes a variable T that holds the cumulative probability value to 0 (S302). . Next, referring to the transition probability table 106-32 in FIG. 4, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is equal to or greater than x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The above operation is repeated until the value of the variable T is equal to or greater than x _max, the state of FIG. 4 the pitch of the state j the value of the variable T is identified by the value of j at the time of equal to or greater than x _max Obtained from the definition table 106-31 (step S306).

  Next, the performance sound determination means 110 substitutes the number of the set state of the previous sound (sound volume) into the variable i (S504 in FIG. 13). Then, the performance sound determination means 110 obtains the state (sound volume) of the next performance sound by calculation (S505).

Details of the processing in step S505 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j holding the number of the next sound state (sound volume) to 1 (S301), and initializes a variable T holding the cumulative probability value to 0 (S301). S302). Next, referring to the transition probability table 106-34 in FIG. 4, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is greater than or equal to x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The above operation is repeated until the value of the variable T is equal to or greater than x _max, 4 the size of the state j sounds the value of the variable T is identified by the value of j at the time of equal to or greater than x _max From the state definition table 106-33 (step S306).

  Next, the performance sound determination means 110 substitutes the number of the set state (sound length) of the immediately preceding sound for the variable i (S506 in FIG. 13). Then, the performance sound determining means 110 obtains the state (sound length) of the next performance sound by calculation (S507).

Details of the processing in step S507 will be described with reference to FIG. The performance sound determination means 110 initializes a variable j holding the number of the next sound state (sound length) to 1 (S301), and initializes a variable T holding a cumulative probability value to 0 (S302). . Next, referring to the transition probability table 106-36 in FIG. 4, the value of the transition probability x _i1 is added to the variable T (S303), and it is determined whether or not the value of the variable T is equal to or greater than x _max (S304). ). If the value of variable T is not greater than or equal to x _max , the value of variable j is incremented by 1 to 2 (S305), the value of transition probability x _i2 is added to variable T (S303), and the value of variable T is x _max It is again determined whether or not the above has been reached (S304). The above operation is repeated until the value of the variable T is equal to or greater than x _max, the state of FIG. 4 the tone length state j the value of the variable T is identified by the value of j at the time of equal to or greater than x _max Obtained from the definition table 106-35 (step S306).

  Finally, the performance sound determination means 110 uses the pitch determined in step S503, the sound volume determined in step S505, and the sound length determined in step S507 as the pitch and volume of the sound to be played next. Is output as a sound length (S508).

  Next, the effect of this embodiment will be described.

  According to the present embodiment, since the degree of arrangement can be changed by selecting the transition probability data to be used, even a user who has little knowledge about music can change the original degree while changing the degree of arrangement. A song arrangement can be created.

  Further, according to the present embodiment, the character information representing the base sequence is converted into a numerical value for every three characters, and the next performance sound is determined depending on the numerical value. It can be converted into a meaningful score. The reason is that an amino acid sequence (protein) is synthesized in the living body based on the information on the base sequence, but the type of amino acid to be synthesized is determined by the combination of the three-letter bases in the base sequence. is there.

  Further, according to the present embodiment, the character information representing the base sequence is converted into a three-digit numerical value for every three characters, so that the first digit in the three-digit numerical value has less influence on the numerical value. This is because the influence of the third character base on the amino acid is small, so the third character base is susceptible to mutation, but the biological effect of this is small. It can be said that this is a scientifically meaningful numerical conversion method.

  In the present embodiment, the base sequence is converted into a numerical value for every three characters, but can generally be converted into a numerical value for every n (a positive integer of 1 or more) characters.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above examples, and various other additions and modifications can be made. For example, an example in which there is only one base music as a base is shown, but a plurality of original music scores and a plurality of types of transition probabilities corresponding to each original music are stored, and a user can select a favorite original music May be selected. Further, it may have a function of performing according to the score of the original music as a base. Furthermore, the self-composing device of the present invention can be realized by a computer and a program such as a personal computer or a computer constituting a mobile phone as well as realizing the functions of the automatic composition device by hardware. The program is provided by being recorded on a computer-readable recording medium such as a magnetic disk or a semiconductor memory, and is read by the computer when the computer is started up, and the computer is controlled by controlling the operation of the computer. It functions as the numerical string generation means, performance sound determination means, and musical sound generation means in the embodiment, and performs the above-described processing.

  As described above, the automatic song apparatus and the automatic song method according to the present invention are useful as an apparatus and a method for easily creating and enjoying an arrangement while the user selects an arrangement degree of a base song. It is suitable for use in an apparatus and method for arranging and enjoying music unique to the base sequence. The automatic music composition apparatus and automatic music composition method according to the present invention can be used, for example, for arrangement of incoming melody of a mobile phone.

It is a block diagram of the automatic musical composition apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the memory content of the 1st transition probability part in a transition probability memory | storage means. It is a figure which shows an example of the memory content of the 2nd transition probability part in a transition probability memory | storage means. It is a figure which shows an example of the memory content of the 3rd transition probability part in a transition probability memory | storage means. It is a figure which shows the general structure of a transition probability table. It is a figure which shows the example which digitizes a base sequence. It is function explanatory drawing of a performance sound determination means. It is a flowchart which shows the process example of the automatic musical composition apparatus which concerns on embodiment of this invention. It is a flowchart which shows the process example of the performance sound determination means in the case of arrangement degree "small". It is a flowchart which shows the example of a calculation process of the state of the following performance sound in a performance sound determination means. It is a figure which shows the simple example of the transition probability used for description of the performance sound determination method of a performance sound determination means. It is a flowchart which shows the process example of the performance sound determination means in the case of arrangement degree "medium". It is a flowchart which shows the example of a process of the performance sound determination means in the case of arrangement degree "Large".

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Input device 102 ... Display device 103 ... Printing device 104 ... Numeric string storage means 105 ... Original song storage means 106 ... Transition probability storage means 107 ... Arrangement storage means 108 ... Acoustic element 109 ... Numeric string generation means 110 ... Performance sound determination Means 111 ... Musical sound generation means 112 ... Control means

Claims (15)

Original music storage means for storing a musical score expressing each sound of the sound sequence constituting the original music in terms of pronunciation timing, pitch, volume, and length;
Transition probability storage means for storing a plurality of types of sound transition probability data in the original music;
Numeric sequence generation means for generating a sequence of numeric values;
Of the plurality of types of transition probability data stored in the transition probability storage means, based on the transition probability data selected by the input from the input device and the numerical value sequence generated by the numerical value sequence generation means, the original music An automatic musical composition apparatus comprising performance sound determining means for determining a sound to be played next from each sound of the original musical score stored in the storage means.   The transition probability storage means includes first transition probability data including data indicating transition probability in the original music in a sound state defined by three attributes of pitch, sound volume, and sound length, pitch and sound. Data indicating the transition probability in the original song of the sound state defined by any two attributes of the size and length of the sound and data indicating the transition probability in the original song of the remaining one sound attribute The third transition probability, including the transition probability data of 2, the data indicating the transition probability in the original musical piece of pitch, the data indicating the transition probability in the original musical piece of the loudness, and the data indicating the transition probability in the original musical piece of the sound length 2. The automatic composition apparatus according to claim 1, wherein at least two pieces of transition probability data among the data are stored.   The automatic musical composition apparatus according to claim 1 or 2, wherein the numerical value string generating means generates a numerical value string obtained by converting character information representing a base sequence into a numerical value for every n characters.   4. The automatic musical composition apparatus according to claim 3, wherein the numerical value string generating means generates a numerical value string obtained by converting character information representing a base sequence into a three-digit numerical value for every three characters. Arrangement storage means for storing a score composed of sounds to be played determined by the performance sound determination means;
The automatic musical composition apparatus according to any one of claims 1 to 4, further comprising a musical sound generating means for generating a musical sound corresponding to a musical score stored in the arrangement storage means. Original music storage means for storing a musical score expressing each sound of the sound sequence constituting the original music in terms of pronunciation timing, pitch, loudness, and length, and plural types of sound transition probability data in the original music An automatic music composition method for creating music using a computer having a transition probability storage means and an input device,
a) the computer inputting selection parameters from the input device;
b) the computer generating a sequence of numerical values;
c) Among the plurality of types of transition probability data stored in the transition probability storage means, the computer selects transition probability data selected by a selection parameter input from the input device and the generated numerical value sequence. And a step of determining a sound to be played next from each sound of the score of the original music stored in the original music storage means.   The transition probability storage means includes first transition probability data including data indicating transition probability in the original music in a sound state defined by three attributes of pitch, sound volume, and sound length, pitch and sound. Data indicating the transition probability in the original song of the sound state defined by any two attributes of the size and length of the sound and data indicating the transition probability in the original song of the remaining one sound attribute The third transition probability, including the transition probability data of 2, the data indicating the transition probability in the original musical piece of pitch, the data indicating the transition probability in the original musical piece of the loudness, and the data indicating the transition probability in the original musical piece of the sound length 7. The automatic composition method according to claim 6, wherein at least two pieces of transition probability data among the data are stored.   8. The automatic music composition method according to claim 6 or 7, wherein in step b, a numerical value sequence is generated by converting character information representing a base sequence into numerical values for every n characters.   9. The automatic composition method according to claim 8, wherein in step b, a numerical value sequence is generated by converting character information representing a base sequence into a three-digit numerical value for every three characters. d) the computer storing a score composed of the sound to be played determined in step c in an arrangement storage means;
The automatic composition method according to any one of claims 6 to 9, further comprising the step of: e) generating a musical tone corresponding to the score stored in the arrangement storage means. Original music storage means for storing a musical score expressing each sound of the sound sequence constituting the original music in terms of pronunciation timing, pitch, loudness, and length, and plural types of sound transition probability data in the original music A computer having a transition probability storage means and an input device,
a) inputting a selection parameter from the input device;
b) generating a sequence of numerical values;
c) Based on the transition probability data selected from the plurality of types of transition probability data stored in the transition probability storage means by the selection parameter input from the input device and the generated numerical sequence, A program for determining a sound to be played next from each sound of the score of the original music stored in the music storage means.   The transition probability storage means includes first transition probability data including data indicating transition probability in the original music in a sound state defined by three attributes of pitch, sound volume, and sound length, pitch and sound. Data indicating the transition probability in the original song of the sound state defined by any two attributes of the size and length of the sound and data indicating the transition probability in the original song of the remaining one sound attribute The third transition probability, including the transition probability data of 2, the data indicating the transition probability in the original musical piece of pitch, the data indicating the transition probability in the original musical piece of the loudness, and the data indicating the transition probability in the original musical piece of the sound length The program according to claim 11, wherein at least two pieces of transition probability data among the data are stored.   13. The program according to claim 11 or 12, wherein in step b, a numerical value sequence is generated by converting character information representing a base sequence into numerical values for every n characters.   14. The program according to claim 13, wherein in step b, a numerical value sequence is generated by converting character information representing a base sequence into a numerical value of three digits of quaternary for every three characters. In addition to the computer,
d) storing a score composed of the sound to be played determined in step c in the arrangement storage means;
The program according to claim 11, further comprising: e) generating a musical sound according to a score stored in the arrangement storage unit.

Images