JP2016161774A - Music generation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a music generation device for automatically composing a music piece which matches an intention or taste of a user.SOLUTION: A music generation system 10 generates N pieces of music data different from one another according to instruction information (that is, an instruction from a user U) sent from a terminal device 12. Specifically, N pieces of music data are generated by one piece of instruction information. In particular, a processing device 40 of the music generation system 10 generates N pieces of music data through automatic composition processing to which a plurality of variables according to instruction information are applied, and a management device 30 of the system manages the automatic composition processing performed by the processing device 40 and presentation of N pieces of music for the terminal device 12.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for generating music.

  A technique for automatically generating music (automatic music technique) has been proposed. For example, Patent Document 1 discloses a technique for generating a music melody by assigning a time series of pitches selected from a plurality of candidates to a time series of pronunciation intervals selected from a plurality of candidates. It is disclosed.

JP2013-45082A

  According to the technique of Patent Document 1, it is possible to generate one musical piece that is musically natural and diverse, but there is a possibility that the musical piece does not match the user's intention or preference. In view of the above circumstances, an object of the present invention is to increase the possibility of providing music that matches the user's intentions and preferences.

  In order to solve the above problems, a music generation device according to a preferred aspect of the present invention includes a music generation unit that generates music data of a plurality of different music according to an instruction from a user, and a music generation unit. Music presentation means for presenting a user with a plurality of music pieces for which music data has been generated. In the above aspect, music data of a plurality of music pieces generated in response to an instruction from the user is generated, and a plurality of music pieces are presented to the user as music data reproduction candidates. Therefore, compared with the structure which produces | generates one music, it is possible to increase the possibility that provision of music that matches the user's intention and preference can be provided to the user.

  In a preferred aspect of the present invention, the music generation means includes voice synthesis means for generating singing data of the singing voice for singing the music with the common lyrics designated by the user for each of the plurality of music pieces, and includes the singing voice. Generate song data for a song. In the above aspect, there exists an advantage that several different music can be shown to a user by the common lyrics designated by the user. In addition, according to the configuration in which the voice synthesizing unit generates the singing data of the singing voice having the voice quality designated by the user, there is an advantage that the music of the singing voice having the voice quality matching the user's intention and preference can be provided. On the other hand, when the user does not specify the voice quality, according to the configuration in which the voice synthesis means generates the singing data of the singing voice of the voice quality selected according to the generation history of the singing voice singing data, the voice quality of the user Compared with the case where selection is essential, there is an advantage that it is possible to select an appropriate voice quality according to the generation history of the song data while reducing the load of the user selecting the voice quality.

  In a preferred aspect of the present invention, the music generation means includes accompaniment generation means for generating accompaniment data of accompaniment sounds of a genre designated by the user for each of a plurality of music pieces, and generates music data of music pieces including accompaniment sounds. To do. In the above aspect, since a plurality of music pieces including accompaniment sounds of the genre designated by the user are generated, it is possible to provide music pieces with accompaniment sounds that match the user's intention and preferences. On the other hand, when the user does not specify a genre, the accompaniment generation means generates accompaniment data of accompaniment sounds of the genre selected according to the accompaniment data generation history. Compared with the case where it is, there exists an advantage that the appropriate genre according to the production | generation history of accompaniment data can be selected, reducing the load which a user selects a genre.

  In a preferred aspect of the present invention, the music presentation means can change the order of presentation of a plurality of music. In the above aspect, since the order of presentation of a plurality of songs is changed, there is an advantage that a plurality of songs can be preferentially presented to the user from various viewpoints. For example, the music presenting means presents a plurality of music in an order corresponding to at least one of the generation history of the music data by the music generating means, the positional information regarding the user, and the time at which the plurality of music is presented. Further, in the configuration including the above-described speech synthesizing means, the music presenting means presents a plurality of music pieces to the user in an order corresponding to the designated speed of lyrics by the user.

  In a preferred aspect of the present invention, the music generation means generates music data of a plurality of different music pieces for each of the plurality of unit sections on the time axis, and the music presentation means outputs each of the plurality of unit sections. The music generation means presents the user with a plurality of music for which music data has been generated. In the above aspect, since a plurality of pieces of music data are generated for each unit section, it is possible to configure one piece of music in which the pieces of music selected for each unit section are arranged in time series over a plurality of unit sections. .

  The music generating device according to each aspect described above is realized by a dedicated electronic circuit, or by cooperation of a general-purpose arithmetic processing device such as a CPU (Central Processing Unit) and a program. The program of the present invention can be provided in a form stored in a computer-readable recording medium and installed in the computer. The recording medium is, for example, a non-transitory recording medium, and an optical recording medium (optical disk) such as a CD-ROM is a good example, but a known arbitrary one such as a semiconductor recording medium or a magnetic recording medium This type of recording medium can be included. For example, the program of the present invention can be provided in the form of distribution via a communication network and installed in a computer.

1 is a configuration diagram of a music generation system according to a first embodiment of the present invention. It is a block diagram of a terminal device. It is a schematic diagram of a music presentation screen. It is a block diagram of a management apparatus and a processing apparatus. It is a block diagram of a music production | generation part. It is explanatory drawing of the whole operation | movement of a music production system. It is a schematic diagram of the music presentation screen in 3rd Embodiment. It is a schematic diagram of the music presentation screen in the modification of 3rd Embodiment.

<First Embodiment>
FIG. 1 is a configuration diagram of a music generation system 10 according to the first embodiment. The music generation system 10 of the first embodiment is an example of the music generation apparatus of the present invention, and generates music data D of a plurality of different (N) music pieces according to an instruction from the user U of the terminal device 12. To do. Each piece of music data D includes a time waveform of a mixed sound of a singing voice of a specific singer singing a singing part (typically main melody) of a music piece and an accompaniment sound of the accompaniment part of the music piece played by an instrument, for example. It is a music file that represents. The terminal device 12 is, for example, a portable communication terminal such as a mobile phone or a smartphone, or a portable or stationary communication terminal such as a personal computer. Note that the music generation system 10 actually communicates with a plurality of terminal devices 12, but only one arbitrary terminal device 12 is shown in FIG. 1 for convenience.

  As illustrated in FIG. 1, the music generation system 10 according to the first embodiment is realized by a computer system including a management device 30 and a processing device 40. Each of the management device 30 and the processing device 40 is a server (typically a web server) connected to the communication network 14 including, for example, a mobile communication network and the Internet. The management device 30 communicates with the terminal device 12 via the communication network 14, and the processing device 40 communicates with the management device 30 via the communication network 14. In FIG. 1, the management device 30 and the processing device 40 are illustrated as separate units for convenience, but the management device 30 and the processing device 40 may be configured as a single device.

  FIG. 2 is a configuration diagram of the terminal device 12. As illustrated in FIG. 2, the terminal device 12 according to the first embodiment includes a control device 20, a storage device 21, a communication device 22, an operation device 23, a display device 24, and a sound emitting device 25. The control device 20 comprehensively controls the operation of the terminal device 12. The communication device 22 executes communication with the management device 30 via the communication network 14. The operation device 23 is an input device that receives an instruction from the user U, and includes, for example, a plurality of operators that detect an operation by the user U and a touch panel that detects the user U's contact with the operation surface.

  The display device 24 displays an image instructed from the control device 20. Specifically, the screen of FIG. 3 (hereinafter referred to as “music presentation screen”) in which the music generation system 10 presents N pieces of music for which music data D has been generated to the user U is displayed on the display device 24. The user U can select a desired music piece among the N pieces of music presented on the music presentation screen by appropriately operating the operation device 23. The sound emitting device 25 (for example, a speaker or headphones) in FIG. 2 emits sound instructed from the control device 20. Specifically, the sound indicated by the music data D of the music (hereinafter referred to as “selected music”) selected by the user U by the operation on the operation device 23 among the N music presented on the music presentation screen is the sound emitting device. 25.

  The user U designates the condition of the music to be generated by the music generation system 10 by appropriately operating the operation device 23. Specifically, the user U can specify the lyrics L of the music and a plurality of music attributes. The lyrics L are character strings that express the pronunciation content of the singing part of the music, and are arbitrarily designated by the user U. The music attribute means the characteristics and characteristics of the music. In 1st Embodiment, the voice quality of the song voice of a music, the genre of music, and a tone are illustrated as a music attribute. Voice quality is the acoustic characteristic of the singing voice that sings the lyrics L of the music. The genre means a musical classification of music (particularly accompaniment sounds), and is selected from a plurality of candidates such as “rock”, “ballad”, “dance”, “jazz”, and the like. The melody means an emotion or feeling (music atmosphere) perceived from the music, and is selected from a plurality of candidates such as “bright”, “sad”, and “friendly”.

  The control device 20 of the terminal device 12 generates instruction information X corresponding to an instruction from the user U to the operation device 23 and transmits the instruction information X from the communication device 22 to the management device 30 of the music generation system 10. The instruction information X is information for instructing the conditions of the music to be generated by the music generation system 10, and one or more pieces of attribute information x (xA, xB, xC) specifying the lyrics L and music attributes specified by the user U. ). The attribute information xA indicates the voice quality specified by the user U, the attribute information xB indicates the genre specified by the user U, and the attribute information xC indicates the music tone specified by the user U. Note that the user U can omit specifying each of a plurality of types of music attributes (voice quality, genre, music tone). That is, it is possible to specify only one type of a plurality of music attributes or omit specification of all of the plurality of music attributes.

  The music generation system 10 in FIG. 1 generates music data D of N different music pieces according to the instruction information X (that is, an instruction from the user U) transmitted from the terminal device 12. That is, N pieces of music data D are generated from one piece of instruction information X. Specifically, the processing device 40 of the music generation system 10 generates music data D of N pieces of music by automatic music processing to which a plurality of variables according to the instruction information X are applied. It manages the automatic music processing by 40 and the presentation of N pieces of music to the terminal device 12.

  FIG. 4 is a configuration diagram of the management device 30 and the processing device 40. In FIG. 4, the communication network 14 is not shown for convenience. As illustrated in FIG. 4, the management device 30 according to the first embodiment includes a control device 32, a storage device 34, and a communication device 36. The communication device 36 communicates with each of the terminal device 12 and the processing device 40 via the communication network 14. A configuration in which the management device 30 and the processing device 40 communicate directly without going through the communication network 14 may also be employed. The storage device 34 is configured by a known recording medium such as a magnetic recording medium or a semiconductor recording medium, and stores a program executed by the control device 32 and various data used by the control device 32. The control device 32 comprehensively controls the operation of the management device 30. The control device 32 according to the first embodiment functions as the information management unit 62 and the music presentation unit 64 by executing a program stored in the storage device 34.

  The information management unit 62 generates control information Z corresponding to the instruction information X received from the terminal device 12 by the communication device 36 for each of N different pieces of music. That is, N different pieces of control information Z are generated from the common instruction information X transmitted from the terminal device 12. Each control information Z is a control file that indicates a variable that is applied to the automatic music piece processing by the processing device 40, and is described in, for example, an XML (Extensible Markup Language) format. The control information Z of the first embodiment includes the lyrics L specified by the instruction information X and attribute information z (zA, zB, zC) of each of a plurality of music attributes.

  The lyrics L of the control information Z is a character string designated by the user U of the terminal device 12 and is common to N pieces of control information Z. On the other hand, the attribute information zA indicates the voice quality of the singing voice, the attribute information zB indicates the genre of the song, and the attribute information zC indicates the tone of the song. The numerical value of each attribute information z can be set differently for each of the N pieces of control information Z. The generation method of each attribute information z of the control information Z is different according to the type of music attribute specified by the user U of the terminal device 12. When the user U designates only the lyrics L without designating the song attributes (designation example 1), or when the user U designates one of a plurality of song attributes together with the lyrics L (designation examples 2 to 2) A specific example of generation of attribute information z will be described in detail below for 4).

[Specification example 1] When the user U designates only the lyrics L without designating the music attribute, the storage device 34 of the management device 30 sets the voice quality of each of a plurality of voice qualities that can be used to generate the singing voice. A history (hereinafter referred to as “generation history”) HA in which songs of the singing voice have been generated in the past is stored. Specifically, the generation history HA of any one kind of voice quality is, for example, the number of times that the voice of the voice quality (single data DA described later) is generated within a past predetermined period (for example, the latest predetermined length period). It is. For example, the number of times that the voice quality is designated by the instruction information X transmitted in the past from the terminal device 12 of the user U (that is, the tendency of the voice quality that the user U prefers) or the past from any of the plurality of terminal devices 12 The number of times that the voice quality is designated by the instruction information X transmitted to (that is, the tendency of voice quality preferred by many users) is stored as the generation history HA. When the user U designates only the lyrics L without designating the music attribute, the information management unit 62 selects the voice quality with the highest number of times indicated by the generation history HA from the plurality of voice qualities, and instructs the voice quality. Attribute information zA is generated. That is, the attribute information zA indicates the voice quality frequently used in the past by the user U himself or many users.

  Further, the storage device 34 stores, for each of a plurality of genres of music that can be generated by the automatic music processing, a history (hereinafter referred to as “generation history”) HB in which music of the genre is generated in the past. Specifically, the generation history HB of any one type of genre is, for example, the number of times that a genre of music (accompaniment data DB described later) has been generated in a past predetermined period. For example, the number of times that the genre is specified by the instruction information X transmitted in the past from the terminal device 12 of the user U (that is, the tendency of the genre that the user U likes) or the past from any of the plurality of terminal devices 12. The number of times that the genre is designated by the instruction information X transmitted to (i.e., the tendency of the genre that many users prefer) is stored as the generation history HB. When the user U designates only the lyrics L without designating the music attribute, the information management unit 62 selects a predetermined number of genres (genres frequently used in the past) in descending order of the number of times indicated by the generation history HB among a plurality of genres. ) To generate attribute information zB indicating one genre randomly selected from a predetermined number of genres (hereinafter referred to as “selected genres”). In addition, the information management unit 62 randomly selects any one of a plurality of tunes suitable for the selected genre, and generates attribute information zC indicating the tune. For example, when the selected genre is “Rock”, the tone selected at random from a plurality of candidates such as “Bright” and “Intense” is indicated by the attribute information zC, and when the selected genre is “Ballade” The tone information randomly selected from a plurality of candidates such as “gentle” and “sad” is indicated by the attribute information zC. As described above, in the designation example 1, it is not necessary to specify the music attributes by the user U. Therefore, there is an advantage that the music can be easily generated while reducing the burden of the user U thinking about the desired music attributes. is there.

[Specification example 2] When user U specifies only lyrics L and voice quality (attribute information xA)
When the user U designates only the lyrics L and the voice quality without designating the genre and tone of the music, the information management unit 62 uses the voice quality designated by the attribute information xA of the instruction information X as the attribute information zA of the control information Z. But I give the same instructions. That is, the voice quality designated by the user U is indicated by the attribute information zA. When there are a plurality of candidates for the voice quality designated by the user U, the information management unit 62 selects any one of the plurality of candidates as in the above-described example by using the generation history HA (the user U itself or a number of users). Select according to voice quality selection tendency). Further, the information management unit 62 selects the genre and tune of the music in the same manner as in the specification example 1, and sets the attribute information zB and the attribute information zC.

  In the specification example 2 exemplified above, the voice quality designated by the user U is indicated by the attribute information zA, and the attribute information z (zB, zC) of the music attribute other than the voice quality is supplemented by the information management unit 62. There is an advantage that it is possible to generate a song of singing voice having a desired voice quality of the user U while reducing the burden on the user U who designates a music attribute other than the voice quality.

[Specification Example 3] When user U specifies only lyrics L and genre (attribute information xB)
When the user U designates only the lyrics L and the genre without designating the voice quality and tone, the information management unit 62 applies the genre designated by the attribute information xB of the instruction information X in the attribute information zB of the control information Z as well. To instruct. That is, the genre designated by the user U is indicated by the attribute information zB.

  In addition, the information management unit 62 refers to the generation history HA for each voice quality stored in the storage device 34, and has been frequently used in the past in the generation of music of the genre indicated by the attribute information xB of the instruction information X. A voice quality (for example, a voice quality that is used most frequently in the generation of music of the genre) is selected, and attribute information zA that indicates the voice quality is generated. The selection of the tone of music indicated by the attribute information zC is the same as in the designation example 1.

  In the specification example 3 exemplified above, the genre designated by the user U is indicated by the attribute information zB, and the attribute information z (zA, zC) of the music attribute other than the genre is supplemented by the information management unit 62. There is an advantage that music of a desired genre of the user U can be generated while reducing the burden on the user U who designates a music attribute other than the genre.

[Specification Example 4] When user U specifies only lyrics L and tone (attribute information xC)
When the user U designates only the lyrics L and the tone without designating the voice quality and genre, the information management unit 62 applies the tone specified by the attribute information xC of the instruction information X in the attribute information zC of the control information Z as well. Is specified. That is, the music tone designated by the user U is designated by the attribute information zC.

  In addition, the information management unit 62 selects a genre frequently used in the past (for example, a genre that has been used most frequently in the generation of past music) among a plurality of genres suitable for the tone specified by the attribute information zC, The genre is designated by the attribute information zB of the control information Z. For example, when the tune specified by the attribute information zC is “violent”, any of a plurality of genres corresponding to “rock” is selected, and when the tune is “friendly”, a plurality of tunes corresponding to “ballad” are selected. One of the genres is selected. The selection of the voice quality designated by the attribute information zA is the same as in the designation example 1.

  In the designation example 4 exemplified above, the music tone designated by the user U is designated by the attribute information zC, and the attribute information z (zA, zB) of the music attribute other than the music tone is supplemented by the information management unit 62. There is an advantage that it is possible to generate a musical piece having a desired musical tone of the user U while reducing a burden on the user U who designates a musical piece attribute other than the musical tone.

  N pieces of control information Z generated by the information management unit 62 in the procedure illustrated above (Specification Example 1 to Specification Example 4) are transmitted from the communication device 36 to the processing device 40. As described above, since each music attribute specified by the control information Z can be selected at random, the lyrics L are common to N pieces of control information Z, while the music attributes (voice quality, genre, music tone) specified by the attribute information z. ) May differ for each control information Z. That is, when the information management unit 62 of the management apparatus 30 transmits N pieces of control information Z, the processing apparatus 40 is instructed to generate N pieces of music having the same lyrics L but different music attributes. .

  As illustrated in FIG. 4, the processing device 40 includes a control device 42, a storage device 44, and a communication device 46. The communication device 46 communicates with the management device 30 via the communication network 14 (or directly). The storage device 44 is composed of a known recording medium such as a magnetic recording medium or a semiconductor recording medium, and stores a program executed by the control device 42 and various data used by the control device 42. The control device 42 comprehensively controls the operation of the processing device 40. The control device 42 according to the first embodiment functions as the music generation unit 66 by executing a program stored in the storage device 44.

  FIG. 5 is a configuration diagram of the music generation unit 66. The music generation unit 66 generates the music data D of N pieces of music corresponding to the different control information Z by the automatic music processing to which each of the N pieces of control information Z received by the communication device 46 from the management device 30 is applied. To do. That is, the music generation unit 66 is an example of an element (music generation means) that generates music data of a plurality of music. As illustrated in FIG. 5, the music generation unit 66 of the first embodiment includes a melody generation unit 72, a voice synthesis unit 74, an accompaniment generation unit 76, and an acoustic processing unit 78.

  The melody generation unit 72 generates a melody (melody) M for each of the N pieces of music indicated by the control information Z. The melody M is a time series of a plurality of notes constituting the singing part, and is described by time series data conforming to a MIDI (Musical Instrument Digital Interface) format, for example. For each note constituting the melody M, a pitch and a sound generation period (for example, a start point and a continuation length) are designated. The melody generation unit 72 of the first embodiment generates a melody M of melody indicated by the attribute information zC of the control information Z for each of the N pieces of control information Z. For generating the melody M, a known technique can be arbitrarily adopted. For example, the method disclosed in Patent Document 1 is suitable. Specifically, a plurality of different pitch sequences (time series of pitches) are prepared for each tune, and the pitch sequence selected at random from a plurality of pitch sequences corresponding to the tune specified by the attribute information zC. The melody M is generated by sequentially assigning (placing) each of the pitches to the sound generation sections. Therefore, the melody M may be different among N pieces of music.

  For each of the N melodies M generated by the melody generator 72, the voice synthesizer 74 sings the singing voice in which the lyrics L of the control information Z is sung by the melody M according to the voice quality specified by the attribute information zA of the control information Z. Singing data DA is generated. That is, the speech synthesizer 74 is an example of an element (speech synthesizer) that generates singing data of singing voice for each of a plurality of music pieces. The singing data DA of the first embodiment represents the waveform of the singing voice.

  In the first embodiment, a plurality of audio libraries P corresponding to different voice qualities are stored in the storage device 44. The speech library P corresponding to any one kind of voice quality is composed of a plurality of speech segments (for example, a single phoneme or a phoneme chain) collected from the speech of the voice quality. The speech synthesizer 74 generates singing data DA for each of the N pieces of music by a unit connection type speech synthesis process in which speech units of the speech library P are sequentially connected.

  Specifically, the speech synthesizer 74 selects the speech library P having the voice quality specified by the attribute information zA from the plurality of speech libraries P in the storage device 44, and selects the speech unit corresponding to the lyrics L from the speech library. Select sequentially from P. Then, the speech synthesizer 74 adjusts each speech unit to the pitch of each note constituting the melody M and the pronunciation period, and generates the song data DA of the singing speech by connecting them on the time axis. . As understood from the above description, the attribute information zA of the control information Z (attribute information xA of the instruction information X) can be rephrased as information designating any one of the plurality of audio libraries P. Note that the content of the speech synthesis process is not limited to the above example (unit connection type). For example, the singing data DA can be generated by speech synthesis processing using a state transition model such as HMM (Hidden Markov Model).

  The accompaniment generation unit 76 of FIG. 5 generates accompaniment data DB for each of the N pieces of music indicated by the control information Z. In other words, the accompaniment generation unit 76 is an example of an element (accompaniment generation means) that generates accompaniment sound accompaniment data for each of a plurality of music pieces. The accompaniment data DB of the first embodiment represents the waveform of the accompaniment sound of the music. The storage device 44 stores, for each of a plurality of different genres, a plurality of candidate data representing the time waveform of the accompaniment sound of the genre. For each of the N pieces of control information Z, the accompaniment generator 76 of the first embodiment, for example, randomly selects one of a plurality of candidate data items of the genre indicated by the attribute information zB of the control information Z as the accompaniment data DB. select. That is, accompaniment data DB is individually generated for each of the N pieces of music.

  The sound processing unit 78 in FIG. 5 generates music data D for each of the N music pieces according to the singing data DA generated by the voice synthesis unit 74 and the accompaniment data DB generated by the accompaniment generation unit 76. That is, music data D representing a mixed sound of the singing voice expressed by the singing data DA and the accompaniment sound expressed by the accompaniment data DB is generated. Specifically, the acoustic processing unit 78 generates, for each of the N pieces of control information Z received from the management device 30, the song data DA generated according to the control information Z (attribute information zA, attribute information zC). Music data D representing a mixed sound of the singing voice and the accompaniment sound of the accompaniment data DB generated according to the control information Z (attribute information zB) is generated. Music data D of N pieces of music generated by the processing device 40 is transferred from the communication device 46 to the management device 30 and stored in the storage device 34. As understood from the above description, the music generation unit 66 of the processing device 40 generates music data D of N different music by the automatic music processing according to the instruction (instruction information X) from the user U. It is expressed comprehensively as an element.

  The music presentation unit 64 of the management device 30 illustrated in FIG. 4 presents the user U with N pieces of music for which the processing device 40 (music generation unit 66) has generated music data D. That is, the music presentation unit 64 is an example of an element (music presentation means) that presents a plurality of music to the user. Specifically, the music presentation unit 64 generates image data G that represents the music presentation screen of FIG. 2 and transmits the image data G to the terminal device 12 from the communication device 36, thereby presenting the music on the display device 24 of the terminal device 12. Display the screen. As illustrated in FIG. 2, the music presentation screen is an image including an area A1 and an area A2. In the area A1, the lyrics L designated by the user U are displayed.

  The area A2 includes N unit areas AU corresponding to different music pieces. The unit area AU of an arbitrary piece of music includes a music number of the music (a code for the user U to distinguish each music), a genre and tone of the music, and an operation image that receives an operation by the user U. And C1. The user U can use the operation device 23 to operate the operation image C1 of the desired music, thereby selecting the music and instructing the reproduction of the music data D. As understood from the above description, the music presentation unit 64 of the management device 30 functions as an element that presents N songs to the user U as reproduction candidates for the music data D. The music presentation unit 64 transmits the music data D of the selected music selected by the user U among the music data D of the N music stored in the storage device 34 from the communication device 36 to the terminal device 12. The control device 20 of the terminal device 12 causes the sound emitting device 25 to reproduce sound corresponding to the music data D by supplying the music data D received by the communication device 22 from the management device 30 to the sound emitting device 25.

  FIG. 6 is an explanatory diagram of the operation of the first embodiment. When the user U of the terminal device 12 designates the lyrics L and the music attributes (voice quality, genre, tone) of the music by operating the operation device 23 (SA1), the control device 20 gives an instruction according to the instruction from the user U. Information X is transmitted from the communication device 22 to the management device 30 (SA2). The instruction information X includes lyrics L designated by the user U and attribute information x (xA, xB, xC) of the music attributes designated by the user.

  When the communication device 36 receives the instruction information X transmitted from the terminal device 12, the information management unit 62 of the management device 30 generates N pieces of control information Z corresponding to different music pieces from the instruction information X (SA3). . Each control information Z includes lyrics L common to N pieces of control information Z, and attribute information z (zA, zB, zC) individually set for each control information Z according to the instruction information X. The information management unit 62 sequentially transmits N pieces of control information Z from the communication device 36 to the processing device 40 (SA4).

  When the communication device 46 receives the N pieces of control information Z transmitted from the management device 30, the music piece generating unit 66 of the processing device 40 generates music piece data D for each of the N pieces of control information Z (SA5). Specifically, as described above, a mixture of the voice of the attribute information zA voiced with the melody L of the melody specified by the attribute information zC and the accompaniment sound of the genre specified by the attribute information zB Sound music data D is generated.

  On the other hand, the control device 32 of the management device 30 periodically inquires of the processing device 40 whether or not the generation of the N pieces of music data D has been completed (that is, the progress of the automatic music processing) (SA6). When the generation of N music data D is completed, the control device 42 of the processing device 40 transmits N music data D from the communication device 46 to the management device 30 as a response to the inquiry from the management device 30 (SA7). . The music presentation unit 64 of the management device 30 stores the N pieces of music data D generated by the processing device 40 in the storage device 34 (SA8), and the image data G on the music presentation screen for presenting the N pieces of music. N pieces of music are presented to the user U by being generated and transmitted to the terminal device 12 (SA9). The control device 32 of the terminal device 12 causes the display device 24 to display the music presentation screen of FIG. 2 expressed by the image data G received by the communication device 22 from the management device 30 (SA10).

  When the user U designates an arbitrary piece of music (selected music) by operating the operation device 23, the control device 20 of the terminal device 12 sends an information request R including the designation of the selected music from the communication device 22 to the management device 30. (SA11). When the communication device 36 receives the information request R, the music presentation unit 64 of the management device 30 acquires the music data D of the selected music specified by the information request R from the storage device 34, and the music data D is transmitted to the communication device 36. To the requesting terminal device 12 (SA12). The control device 32 of the terminal device 12 reproduces the selected song by supplying the music data D received by the communication device 22 from the management device 30 to the sound emitting device 25 (SA13). That is, the singing sound that pronounces the lyrics L designated by the user is reproduced in parallel with the accompaniment sound.

  As described above, in the first embodiment, the music data D of N different music pieces are generated by the automatic music processing according to the instruction from the user U, and the reproduction candidates of the music data D (by the user U) N pieces of music are presented to the user U as selection candidates). Therefore, it is possible to increase the possibility that the user U can be provided with music that matches the intention and preference of the user U, as compared with the technique of Patent Document 1 that generates one piece of music.

  In the first embodiment, since N pieces of music including the singing voice of the common lyrics L specified by the user U are generated, the music attributes (voice quality, genre, tone) are expressed by the common lyrics specified by the user U. ) Can be presented to the user in N pieces of music. In the first embodiment, in particular, since the singing voice having the voice quality designated by the user U is generated, it is possible to provide the song having the voice quality that matches the user's intention and preference. On the other hand, when the user U does not specify the voice quality (for example, designation example 1), since the voice of the voice quality corresponding to the singing voice generation history HA is generated, the selection of the voice quality by the user U is essential. There is an advantage that it is possible to select an appropriate voice quality according to the generation history HA while reducing the load on the user U as compared with the above.

  In the first embodiment, since N pieces of music including the accompaniment sound of the genre designated by the user U are generated, it is possible to provide the music of the accompaniment sound that matches the user's intention and preference. It is. On the other hand, if the user does not specify a genre, an accompaniment sound of a genre corresponding to the generation history HB is generated, so that the load on the user U is reduced compared to a configuration in which the user U must select a genre. There is an advantage that an appropriate genre can be selected according to the generation history HB while reducing.

Second Embodiment
A second embodiment of the present invention will be described below. In addition, about the element which an effect | action and function are the same as that of 1st Embodiment in each form illustrated below, the code | symbol used in 1st Embodiment is diverted and each detailed description is abbreviate | omitted suitably.

  The music presentation unit 64 of the second embodiment can change the order of presentation of N pieces of music presented on the music presentation screen (that is, the order of arrangement of the unit areas AU for each music on the music presentation screen). Specifically, the music presenting unit 64 presents N pieces of music in the order corresponding to various information selected by the user U from a plurality of options by operating the operation device 23. In the second embodiment, it is assumed that the user U can select any of a plurality of pieces of information including a music generation history, position information, user information, presentation time, and lyrics designation speed.

(1) Music generation history When the user U selects the presentation of N pieces of music in the order corresponding to the music generation history, the music presentation unit 64 has a history in which the music generation unit 66 has generated the music data D in the past. The order of presentation of N pieces of music is determined according to the above. Specifically, a song having a higher number of times indicated by the generation history HB stored in the storage device 34 (that is, a song of a genre that is frequently selected by the user U or a large number of users U in the past) is ranked higher. The presentation order of N pieces of music is determined so as to be positioned. For example, when the frequency at which the user U designates the “Rock” genre in the past is high, the music of the “Rock” genre is presented at the top. It is also possible to present a genre of music selected by a large number of users within a predetermined period (for example, within the last week) at the top.

(2) Position information When the user U selects presentation of N pieces of music in the order according to the position information, the music presentation unit 64 is generated by a position detection function using, for example, GPS (Global Positioning System). The user U's location information (route, latitude) is acquired from the terminal device 12, and the presentation order of the N pieces of music is determined according to the location information. Specifically, a correlation with a geographical position is set in advance for each of a plurality of genres that can be selected by the user U, and a genre having a higher correlation with the position of the user U is positioned higher. The order in which N pieces of music are presented is determined. For example, when the user U is located in South America, the music of the “samba” genre is presented at the top, and when the user U is located in Europe, the music of the “country” genre is presented at the top. . In addition, the method of specifying the position of the user U is not limited to the above illustration (GPS). For example, a configuration in which the approximate position of the user U is estimated from the IP (Internet Protocol) address of the terminal device 12 may be employed.

(3) User information When the user U selects presentation of N pieces of music in the order corresponding to the user information, the music presentation unit 64 responds to the user information stored in advance in the storage device 34. To determine the presentation order of the N pieces of music. The user information includes, for example, location information regarding the location of the user U's birthplace (for example, nationality) or residence. Similar to the above-described example, the music presentation unit 64 determines the order of presentation of the N pieces of music so that the genre of music having a higher correlation with the position indicated by the position information registered in the user information is positioned higher. .

(4) Presentation time When the user U selects the presentation of N pieces of music in the order corresponding to the presentation time, the music presentation unit 64 sets the music presentation screen to the user U at the time (presentation time). Accordingly, the order of presentation of N pieces of music is determined. Specifically, when the presentation time is included in the morning time zone, a song with a “bright” tone is positioned at the top, and when the presentation time is included in the daytime afternoon time zone, The order of presentation of the N songs is determined so that the song with the tone of "" is positioned at the top and the song with the tone of "gentle" is positioned at the top when the presentation time is included in the night time zone. The

(5) Lyric designation speed When the user U selects presentation of N pieces of music in the order corresponding to the lyrics designation speed, the music presentation section 64 causes the user U to specify the lyrics L by an operation on the operation device 23. The order of presentation of N pieces of music is determined in accordance with the speed of performing (lyric designation speed). The lyric designation speed is the number of characters of the lyrics L (for example, typing speed with respect to the keyboard) input by the user U within a predetermined time, and is measured by the terminal device 12 and then notified to the music presentation unit 64. It is estimated that the user U who has a high lyric designation speed is likely to desire a song with a fast tempo. In consideration of the above tendency, the music presentation unit 64 is arranged such that a song with a fast tempo is positioned higher when the lyrics designation speed is high, and a song with a slower tempo is ranked higher when the lyrics designation speed is low. The order of presentation of N pieces of music is determined. However, contrary to the above example, when the lyrics designation speed is high, it is also possible to place a song with a low tempo at the top.

  In the second embodiment, the same effect as in the first embodiment is realized. Moreover, in 2nd Embodiment, since the order which shows N music is changed suitably, there exists an advantage that N music can be preferentially shown to the user U from various viewpoints. Note that the information reflected in the order of presentation of the N pieces of music is not limited to the above examples (music generation history, position information, user information, presentation time, lyrics designation speed).

<Third Embodiment>
FIG. 7 is an explanatory diagram of a music presentation screen in the third embodiment. As illustrated in FIG. 7, the music presentation screen of the third embodiment includes a plurality of unit regions B corresponding to different sections (hereinafter referred to as “unit sections”) on the time axis. The unit section is a section having a time length corresponding to a predetermined number of bars, for example. A unit region B corresponding to an arbitrary unit section includes a region B1 and a region B2. As illustrated in FIG. 7, the lyrics L designated by the user U are divided into a plurality of parts (character strings) λ for each unit section, and are divided into a region B1 of the unit area B corresponding to any one unit section. Indicates the part λ of the lyrics L corresponding to the unit section. The method of dividing the lyrics L into a plurality of parts λ is arbitrary. For example, a method in which the user U individually designates each part λ or a method in which the user U designates a boundary between successive parts λ is adopted. It is also possible to classify the lyrics L in units of a predetermined number of characters, or to classify the lyrics L in units of a semantic group specified by known natural language processing such as morphological analysis.

  The music generation unit 66 of the processing device 40 of the third embodiment generates music data D of N different music pieces for each of a plurality of unit sections on the time axis. Specifically, the music generation unit 66 executes the process of generating the music data D of N pieces of music in the same manner as in the first embodiment for each of the unit sections by sequentially executing the process for each of the plurality of unit sections. N pieces of music data D are generated. However, the part corresponding to the unit section of the lyrics L is applied to the generation (speech synthesis process) of the song data DA in any one unit section. In addition, although the case where the total number N of the music for every unit section is common over several unit sections is illustrated in FIG. 8, it is also possible to make the total number N of music differ for every unit section.

  As illustrated in FIG. 7, each of the N pieces of music generated by the music generation unit 66 for the unit section is stored in the area B <b> 2 of the unit area B corresponding to any one unit section of the music presentation screen. A music image C2 to be represented is displayed. The user U can instruct the reproduction of the music data D of the music by operating the music image C2 of the desired music among the N music for each unit section using the operation device 23. is there. The music presentation unit 64 of the management device 30 transmits the music data D of the music selected by the user U among the music data D of N pieces of music stored in the storage device 34 from the communication device 36 to the terminal device 12 for each unit section. The sound emission device 25 is made to reproduce by transmitting. That is, one piece of music in which selected songs for each unit section are arranged over a plurality of unit sections is configured. As understood from the above description, the music generation unit 66 of the third embodiment presents N pieces of music to the user U as reproduction candidates for the music data D in each unit section for each of the plurality of unit sections. Acts as an element.

  In the third embodiment, the same effect as in the first embodiment is realized. In the third embodiment, since music data D of N pieces of music is generated for each unit section, one piece of music in which the selected pieces of each unit section are connected in time series over a plurality of unit sections is formed. It is possible.

  Note that the same method as in the second embodiment can be adopted for selecting the presentation order of N pieces of music for each unit section. In addition, according to the contents of the part λ of each unit section in the lyrics L, it is also possible to determine the order of presentation of N pieces of music in the unit section. Specifically, a configuration in which music of a genre or a melody that correlates with the meaning or feeling of the part λ of any one unit section in the lyrics L is preferred. For example, as illustrated in FIG. 8, for a unit section in which the part λ of the lyrics L is “sunny”, a song with a “bright” tone that correlates with the meaning or feeling of the part λ is “lonely”. For a unit section that is presented at a higher rank than the music in the melody and the part λ of the lyrics L is “rainy”, the music in the “lonely” melody that correlates with the meaning or feeling of the part λ is “bright” "Is presented at a higher rank than the music with the tone of". " According to the above configuration, there is an advantage that it is easy for the user U to select a music piece in tune that harmonizes with each part λ of the lyrics L.

<Modification>
Each form illustrated above can be variously modified. Specific modifications are exemplified below. Two or more aspects arbitrarily selected from the following examples can be appropriately combined as long as they do not contradict each other.

(1) The type of music attribute indicated by the control information Z is not limited to the above-described examples (voice quality, genre, music tone). For example, the tempo of the music can be indicated by the attribute information z of the control information Z. The music generation unit 66 generates music data D of the music with the tempo designated by the attribute information z. When the user U designates the tempo, the tempo is designated by the attribute information z. When the user U does not designate the tempo, the tempo corresponding to the genre designated by the attribute information zB is displayed in the attribute information z. Is instructed. That is, a suitable tempo is determined in advance for each genre of music, and a tempo corresponding to the genre specified by the attribute information zB is selected. For example, when the music is in the genre of “rock”, a faster tempo is set compared to “ballad”.

(2) A configuration (hereinafter referred to as “configuration A”) in which the music data D of a plurality of music generated by the music generation unit 66 can be reproduced by any terminal device 12 other than the terminal device 12 of the user U may be employed. . Specifically, the management device 30 sequentially accumulates the N pieces of music data D generated by the music generation unit 66 in the storage device 34, and the music data D requested from any terminal device 12 is the requesting terminal. Transmit to device 12.

(3) The method by which the information management unit 62 determines the genre of music is not limited to the above examples (designation examples 1, 2, and 4). For example, in the above-described configuration A, a configuration in which a genre of a song that has been played back in response to a request from a large number of terminal devices 12 among a plurality of genres (a genre of a song that is ranked higher in the playback count ranking) is selected. Can be employed. It is also possible to determine the genre according to the time (for example, the aforementioned presentation time) when the music data D of the music is provided to the user U according to the instruction information X among a plurality of genres. For example, if the presentation time is included in the daytime period, the “lock” genre is selected, and if the presentation time is included in the nighttime period, the “ballad” genre is selected.

(4) In each of the above-described embodiments, the music generation system 10 including the management device 30 and the processing device 40 has been exemplified as an embodiment of the music generation device. However, the management device 30 and the processing device 40 are integrally configured. It is also possible to use the device as a music generation device. Moreover, it is also possible to implement | achieve one or both of the management apparatus 30 and the processing apparatus 40 with the some apparatus comprised mutually separately.

(5) In each of the above-described embodiments, the voice quality with the highest number of times indicated by the generation history HA is selected from among a plurality of voice qualities. However, the method for selecting the voice quality according to the generation history HA is not limited to the above examples. For example, singing data by synthesizing each speech unit over a configuration in which one kind of voice quality with the minimum number of times indicated by the generation history HA is selected, or over a plurality of voice quality voice libraries P selected in descending or ascending order of the number of times. A configuration for generating DA may also be employed. Further, in each of the above-described embodiments, the genre positioned higher in the descending order of the number of times indicated by the generation history HB is selected from among the plurality of genres, but the method of selecting the genre according to the generation history HB is not limited to the above examples. . For example, it is possible to select a genre located at the top in ascending order of the number of times indicated by the generation history HB.

(6) The generation history HA and the generation history HB in each of the above-described forms are cumulatively reflected in the generation of the past music data D, but the significance as the generation history HA and the generation history HB is ensured. In the initial stage where the music data D has not been generated a sufficient number of times (for example, the initial shipping stage), for example, the average history of the results specified in the past by a large number of users throughout the country and the trends of that era are displayed. The reflected history can be set as the generation history HA and the generation history HB.

(7) In each of the above-described embodiments, the music data D representing the time waveform of the mixed sound of the singing voice and the accompaniment sound of the music is illustrated, but the format of the music data D is arbitrary. For example, a voice synthesis file (for example, a VSQ format file) that specifies pronunciation content (lyrics), pitch, and duration for each of a plurality of musical notes of a musical piece, and the pronunciation of each musical accompaniment note. It is also possible for the music generation unit 66 to generate music data D including a file for automatic performance (for example, a file in MIDI format) designated for the series.

(8) In each embodiment described above, N pieces of music are presented to the user U as reproduction candidates for the music data D, but the purpose of presenting the N pieces of music (use of the music data D) is to reproduce the music data D. It is not limited to. For example, a configuration in which N pieces of music are presented as music data D distribution candidates (not required to be playback candidates), or music data D is converted to another format (for example, wav format) to the user U. A configuration in which N pieces of music are presented as conversion candidates for distribution may be employed. As can be understood from the above description, the music presentation unit 64 in each of the above-described forms is comprehensively expressed as an element that the music generation unit 66 presents N pieces of music for which the music data D is generated to the user U. .

DESCRIPTION OF SYMBOLS 10 ... Music generation system (music generation device), 12 ... Terminal device, 14 ... Communication network, 20 ... Control device, 21 ... Storage device, 22 ... Communication device, 23 ... Operation device, 24 ... ... Display device, 25 ... Sound emitting device, 30 ... Management device, 32 ... Control device, 34 ... Storage device, 36 ... Communication device, 40 ... Processing device, 42 ... Control device, 44 ... Storage device 46 .. communication device 62 .. information management unit 64 .. music presentation unit 66 .. music generation unit 72 .. melody generation unit 74 .. speech synthesis unit 76 .. accompaniment generation unit 78. Acoustic processing unit.

Claims (5)

Music generation means for generating music data of a plurality of different music according to instructions from the user;
A music generating apparatus comprising: music presenting means for presenting the plurality of music generated by the music generating means to the user. The music generating means includes voice synthesizing means for generating, for each of the plurality of music pieces, singing data of a singing voice that sings the music with the common lyrics designated by the user, and music data of the music containing the singing voice Produces
The voice synthesis means generates the singing data of the singing voice of the voice quality specified by the user, while the voice quality selected according to the generation history of the singing voice singing data when the user does not specify the voice quality The song generation device according to claim 1, wherein the song data of the singing voice is generated. The music generation means generates accompaniment data of the accompaniment sound of the genre designated by the user for each of the plurality of music pieces, and when the user does not designate a genre, according to the accompaniment data generation history The music generation device according to claim 1, further comprising accompaniment generation means for generating accompaniment data of the accompaniment sound of the selected genre, and generating music data of the music including the accompaniment sound. The music presentation means displays the music pieces in a variable order according to at least one of a generation history of music data by the music generation means, position information about the user, and a time at which the music pieces are presented. The music generating device according to any one of claims 1 to 3. The music generation means generates music data of a plurality of different music for each of the plurality of unit sections on the time axis,
The music generation device according to any one of claims 1 to 4, wherein the music presentation unit presents the plurality of pieces of music generated by the music generation unit for each of the plurality of unit sections to the user. .

Images