JP7419768B2 - Music generation method and music generation system - Google Patents

Description

The present disclosure relates to a music generation method for generating music.

For example, various techniques for generating music have been proposed in the past. Patent Document 1 discloses a composition support system that generates music according to parameters input by a user.

Japanese Patent Application Publication No. 2007-17578

For example, if the parameters input by the user are similar, the music pieces generated by the composition support system will be similar. Therefore, it may not be possible to meet the desire to generate unique music. It is also conceivable to compare the generated song with songs generated in the past. Specifically, a configuration is envisioned in which acoustic features such as frequency characteristics and volume are compared. However, there is a problem in that comparing the acoustic features of songs requires an excessive processing load or storage capacity. Taking the above circumstances into consideration, the present invention aims to easily determine whether a piece of music to be composed is a unique piece of music.

In order to solve the above problems, a music generation method according to one aspect of the present disclosure is a music generation method that generates a music according to a parameter set including a plurality of music parameters, each of which represents a condition regarding the music. , it is determined whether the first parameter set and the second parameter set are similar.

Further, a music generation system according to one aspect of the present disclosure includes a music generation unit that generates music according to a parameter set including a plurality of music parameters, each of which represents a condition regarding the music, and a first parameter set and a second parameter. and a determination unit that determines whether or not the sets are similar.

FIG. 1 is a block diagram illustrating the configuration of a music generation system according to a first embodiment. FIG. 2 is a block diagram showing the functions of a control device. It is a schematic diagram of an operation image. It is a schematic diagram of input data. It is a schematic diagram of a parameter set. 3 is a flowchart of processing executed by the control device. FIG. 2 is a block diagram illustrating the configuration of a content generation system in a second embodiment. It is a block diagram showing the function of a control device in a 2nd embodiment. It is a flowchart of the processing which a control device performs in a modification.

A: First Embodiment FIG. 1 is a block diagram illustrating the configuration of a music generation system 100 according to a first embodiment of the present disclosure. The music generation system 100 is a computer system that generates music according to instructions from the user U. A song is generated according to a plurality of conditions regarding the song specified by the user U. The music generation system 100 is, for example, a portable information terminal such as a mobile phone or a smartphone, or a portable or stationary information terminal such as a personal computer. Specifically, the music generation system 100 includes a control device 11, a storage device 12, a display device 13, an operating device 14, and a sound emitting device 15.

The storage device 12 in FIG. 1 is one or more memories configured with a known recording medium such as a magnetic recording medium or a semiconductor recording medium. The storage device 12 stores programs executed by the control device 11 and various data used by the control device 11. Note that the storage device 12 may be configured by a combination of multiple types of recording media. Further, the storage device 12 may be a portable recording medium that is detachable from the music generation system 100, or an external recording medium (for example, online storage) with which the music generation system 100 can communicate via a communication network.

The control device 11 is, for example, one or more processors that control each element of the music generation system 100. For example, the control device 11 may include one or more types of CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), FPGA (Field Programmable Gate Array), or ASIC (Application Specific Integrated Circuit). Consists of a processor.

FIG. 2 is a block diagram illustrating the functional configuration of the control device 11. As shown in FIG. The control device 11 has a plurality of functions (data generation section 111, conversion section 112, determination section 113, music generation section 114) for generating music by executing a plurality of tasks according to a program stored in the storage device 12. , sound source section 115, and control section 116). Note that the functions of the control device 11 may be realized by a set of a plurality of devices (ie, a system), or some or all of the functions of the control device 11 may be realized by a dedicated electronic circuit.

The sound emitting device 15 emits music generated by the music generation system 100 under the control of the sound source section 115. For example, a speaker or headphones are used as the sound emitting device 15. The operating device 14 is an input device that receives instructions from the user U. The operating device 14 is, for example, a plurality of operating elements that can be operated by the user U, or a touch panel that detects contact with the display surface of the display device 13. Specifically, the operating device 14 receives from the user U an operation for specifying a plurality of parameters (hereinafter referred to as "input parameters") P1 regarding the conditions of the music.

The display device 13 (for example, a liquid crystal display panel) displays various images under the control of the control unit 116. For example, the display device 13 displays a screen (hereinafter referred to as "operation image") G for the user U to specify a plurality of input parameters P1. FIG. 3 is a schematic diagram of the operation image G. The user U specifies each of a plurality of input parameters P1 regarding a desired piece of music. Each input parameter P1 has contents that can be understood even by a user U who does not have specialized knowledge for generating music, for example. FIG. 3 exemplifies input parameters P1 representing conditions specified for "image of the song," "situation in which the song is used," "tempo of the song," and "length of the song."

The user U selects one of a plurality of candidates prepared in advance for each input parameter P1, for example. Examples of candidates for the input parameter P1 related to "image" include "light," "quiet," "elegant," "fashionable," "profound," or "splendid." Examples of candidates for the input parameter P1 related to "situation" include "night view," "tropical country," or "nature." Examples of candidates for the input parameter P1 related to "tempo" include "slow", "normal", and "fast". Examples of candidates for the input parameter P1 regarding "length" include "60 seconds," "90 seconds," or "180 seconds."

Note that the content that can be specified as the input parameter P1 is not limited to the above example. For example, candidates for the input parameter P1 regarding the "atmosphere of the music" include "bright," "dark," or "calm." Examples of candidates for the input parameter P1 regarding the "pitch of a song" include "high," "normal," or "low." Examples of candidates for the input parameter P1 related to "year" include "1980s," "1990s," or "2000s." Furthermore, the input parameter P1 may represent a condition specifying, for example, the attributes of the user U (for example, age or gender), the industry in which the music is used, the type of musical instrument used to play the music, or the like. As can be understood from the above description, the input parameter P1 of the first embodiment can be specified using words that are used on a daily basis rather than technical terms of music. Note that a configuration is also adopted in which the user U can specify each input parameter P1 by inputting arbitrary characters. For example, the user U specifies the input parameter P1 by inputting an arbitrary word or sentence.

The data generation unit 111 generates input data Da including a plurality of input parameters P1 specified by the user U. FIG. 4 is a schematic diagram of input data Da. Specifically, the input data Da includes a plurality of input parameters P1 regarding different conditions.

The conversion unit 112 (an example of an acquisition unit) generates a parameter set S used for generating music. Specifically, the conversion unit 112 converts the input data Da generated by the data generation unit 111 into a parameter set S. FIG. 5 is a schematic diagram of the parameter set S. The parameter set S includes a plurality of parameters (hereinafter referred to as "music parameters") P2 related to the conditions of the music. The music parameter P2 represents more specialized conditions than each input parameter P1. In other words, the conditions represented by the music parameter P2 are difficult for the user U, who does not have specialized knowledge of music generation, to understand. In FIG. 5, song parameters P2 representing conditions specified for "genre," "music composition," "chord progression," "BPM (Beats Per Minute)," and "number of measures" are illustrated. The music parameter P2 represents a more specific condition regarding the music than the input parameter P1. Note that the music parameter P2 is not limited to the above example. For example, the music parameter P2 may represent conditions such as pitch, the type of instrument used to play the music, an accompaniment pattern, or time signature (4/4, 6/8, . . . ).

Here, when generating a song, there is a desire to generate a unique song that has a musical impression that is not similar to songs that have been generated in the past. When the parameter sets S are similar, there is a tendency that the musical impressions of the generated songs are also similar. Typically, songs generated using a common parameter set S will be the same. Therefore, the music generation system 100 determines whether the parameter set S acquired by the conversion unit 112 is similar to the parameter set S that the music generation system 100 used in the past to generate music. Therefore, it is possible to determine whether the musical impression of a piece of music that is to be generated in response to an instruction from the user U is similar to the musical impression of a piece of music that has been generated in the past.

In the following description, the parameter set S generated by the conversion unit 112 will be referred to as a "first parameter set", and the parameter set S used in the past to generate music will be referred to as a "second parameter set". The storage device 12 in FIG. 1 stores a plurality of second parameter sets. Each second parameter set corresponds to a different song. Note that the second parameter set is stored in the storage device 12 every time a song is generated.

In the first embodiment, as described above, the first parameter set is generated according to the input data Da. Each input parameter P1 is directly specified by the user U. On the other hand, each music parameter P2 is set by the conversion unit 112 taking into account the input parameter P1.

For example, song parameters P2 related to "genre," "music composition," and "chord progression" are generated according to input parameters P1 related to "image" and "situation." The music parameter P2 regarding "BPM" and "number of bars" is generated according to each input parameter P1 regarding "tempo" and "length".

For example, a reference table stored in the storage device 12 is used to obtain the first parameter set. The reference table is a data table in which a first parameter set is registered for each combination of a plurality of input parameters P1 (ie, input data Da). Specifically, the conversion unit 112 generates the parameter set S corresponding to the input data Da generated by the data generation unit 111 among the plurality of input data Da of the reference table as the first parameter set. In the reference table of the first embodiment, a plurality of first parameter sets are registered for one piece of input data Da. The converter 112 probabilistically selects one of the plurality of first parameter sets corresponding to the input data Da generated by the data generator 111 using, for example, a random element such as a random number. That is, the first parameter set is generated according to the input data Da and random elements.

Therefore, in the first embodiment, even if the input parameters P1 designated by the user U are common, various first parameter sets that are different from each other can be generated. Note that the method for generating the first parameter set according to the input data Da is arbitrary. For example, a configuration may be adopted in which the first parameter set is generated by inputting the input data Da to a statistical model (for example, a neural network) that has learned the relationship between the input data Da and the first parameter set.

The determining unit 113 determines whether the first parameter set and the second parameter set are similar. Specifically, the determination unit 113 includes a calculation unit 133 and a processing unit 135. The calculation unit 133 calculates an index (hereinafter referred to as "evaluation value") representing the degree of similarity between the first parameter set and the second parameter set. The evaluation value is calculated according to the degree of similarity for each music parameter P2 between the first parameter set and the second parameter set. The degree of similarity between music parameters P2 of the same type is calculated between the first parameter set and the second parameter set. For example, the total value of similarities over a plurality of song parameters P2 is calculated as the evaluation value. Therefore, the more similar the first parameter set and the second parameter set are as a whole, the larger the evaluation value becomes.

The processing unit 135 determines whether the first parameter set and the second parameter set are similar according to the evaluation value calculated by the calculation unit 133. Specifically, the processing unit 135 determines that the first parameter set and the second parameter set are not similar when the evaluation value is less than the threshold, and when the evaluation value exceeds the threshold, the processing unit 135 determines that the first parameter set and the second parameter set are not similar. It is determined that the set and the second parameter set are similar. If the evaluation value exceeds the threshold, the music to be generated according to the first parameter set is similar to music generated in the past (that is, the music according to the second parameter set). On the other hand, if the evaluation value is less than the threshold, the song to be generated is not similar to songs generated in the past. The threshold value is set experimentally or statistically.

Note that the determination unit 113 determines, for each of the plurality of second parameter sets stored in the storage device 12, whether or not the second parameter set and the first parameter set are similar. If it is determined that the first parameter set is not similar to any of the plurality of second parameter sets, the music to be generated according to the first parameter set is based on the plurality of songs generated by the music generation system 100 in the past. Not similar to any of the songs.

The music generation unit 114 generates music according to the first parameter set. That is, a piece of music is generated according to the plurality of input parameters P1 specified by the user U. Specifically, music data M representing music according to the first parameter set is generated. The music data M is, for example, in a MIDI (Musical Instrument Digital Interface) format in which instruction data that specifies the intensity and instructs pronunciation, and time data that specifies the point of pronunciation specified by each instruction data are arranged in chronological order. This is a file (SMF: Standard MIDI File).

Specifically, if it is determined that the first parameter set and the second parameter set are not similar, the music generation unit 114 generates the music according to the first parameter set. In the first embodiment, when it is determined that the first parameter set is not similar to any of the plurality of second parameter sets, a music piece is generated according to the first parameter set. Therefore, a unique song that is not similar to any of the songs that the song generation system 100 has generated in the past is generated.

Known automatic composition is optionally employed to generate the music. For example, a plurality of material data representing parts of a musical piece generated in advance are stored in the storage device 12. The music generation unit 114 arranges material data that matches one or more music parameters P2 specified by the first parameter set in chronological order. For example, the material data is selected according to the music parameter P2 specified regarding "genre", "music composition", or "chord progression". Then, the music generation unit 114 generates music data M by adjusting each material data according to one or more music parameters P2 specified by the first parameter set. For example, each material data is adjusted according to music parameters P2 specified for "BPM" and "number of bars." In the first embodiment, a configuration is assumed in which if the first parameter set is common, the music generated by the music generation unit 114 is also common. Note that, for example, a configuration may also be adopted in which the music data M is generated by inputting the first parameter set to a statistical model (for example, a neural network) that has learned the relationship between the first parameter set and the music data M.

The control unit 116 presents the music generated by the music generation unit 114 to the user U. Specifically, the control unit 116 causes the display device 13 to display that the music has been generated. After viewing the display on the display device 13, the user U instructs the playback of the music by operating the operating device 14. The control unit 116 plays music according to instructions from the user U. Specifically, the control unit 116 supplies the instruction data to the sound source unit 115 at the timing specified by the time data of the music data M. The sound source section 115 generates an acoustic signal according to each instruction data of the music data M, and supplies the acoustic signal to the sound emitting device 15. The first parameter set used for music generation by the music generation unit 114 is stored in the storage device 12 as a second parameter set used for subsequent determination by the determination unit 113.

On the other hand, if it is determined that the first parameter set and the second parameter set are similar, the music generation unit 114 does not generate a music using the first parameter set. In the first embodiment, if there is even one second parameter set similar to the first parameter set, no music is generated. If the music according to the first parameter set is not generated, the control unit 116 proposes to the user U a change in the input parameter P1. Specifically, the control unit 116 causes the display device 13 to display information that the music cannot be generated because it is similar to music that has been generated in the past, and that prompts the user U to change the input parameter P1.

FIG. 6 is a flowchart of processing executed by the control device 11. The process in FIG. 6 is started, for example, in response to an instruction from the user U. The data generation unit 111 receives designation of a plurality of input parameters P1 from the user U (Sa1). The data generation unit 111 generates input data Da including a plurality of input parameters P1 specified by the user U (Sa2). The conversion unit 112 converts the input data Da into a first parameter set including a plurality of music parameters P2 (Sa3). The calculation unit 133 sequentially selects a plurality of second parameter sets stored in the storage device 12 (Sa4). Then, the calculation unit 133 calculates the evaluation value between the selected second parameter set and the first parameter set (Sa5).

If there is a second parameter set for which an evaluation value has not been calculated among the plurality of second parameter sets (Sa6: NO), the process moves to step Sa4. That is, the processes from step Sa4 to step Sa6 are repeatedly executed until evaluation values are calculated for all the second parameter sets.

If evaluation values have been calculated for all second parameter sets (Sa6: YES), the processing unit 135 determines whether there is a second parameter set similar to the first parameter set (Sa7). It is determined whether each of the plurality of second parameter sets is similar to the first parameter set. Specifically, the processing unit 135 determines that the first parameter set and the second parameter set are not similar when the evaluation value is less than the threshold, and when the evaluation value exceeds the threshold, the processing unit 135 determines that the first parameter set and the second parameter set are not similar. It is determined that the set and the second parameter set are similar.

If there is no second parameter set similar to the first parameter set (Sa7: NO), the music generation unit 114 generates music according to the first parameter set (Sa8). That is, music data M is generated according to the first parameter set. The control unit 116 stores the first parameter set used to generate the current music piece in the storage device 12 as a second parameter set that the determination unit 113 uses for subsequent determination (Sa9).

The control unit 116 presents the music generated by the music generation unit 114 to the user U (Sa10). For example, the control unit 116 displays that a song has been generated, and plays the song in response to an instruction from the user U. Specifically, the control unit 116 generates an acoustic signal by supplying each instruction data of the music data M to the sound source unit 115. That is, ultimately, the user U is provided with the audio signal of the music according to the first parameter set.

On the other hand, if any one of the plurality of second parameter sets is similar to the first parameter set (Sa7: YES), the control unit 116 displays, for example, on the display device 13 that similar songs have been generated in the past. The user U is notified by displaying the image (Sa11). If a second parameter set similar to the first parameter set exists (Sa7: YES, Sa11), the process moves to step Sa3. In step Sa3, the conversion unit 112 generates a first parameter set different from the first parameter set already generated in the process of FIG. The control unit 116 calculates the evaluation value between the updated first parameter set and each second parameter set (Sa4-Sa6), and determines whether there is a second parameter set similar to the first parameter set. The following judgment (Sa7) is executed. That is, the processes from step Sa3 to step Sa7 are repeatedly executed until a music piece that is not similar to music that has been generated in the past is generated. Through the process shown in FIG. 6, a song that is not similar to any of the plurality of songs generated in the past is generated.

As can be understood from the above description, it is determined whether the first parameter set related to the song to be generated is similar to the second parameter set that was used in the past to generate the song. Therefore, since it is not necessary to compare the acoustic features of songs, it is necessary to check whether the song to be generated is similar to songs generated in the past (i.e., whether the song to be generated is a unique song). It is possible to easily determine whether the In other words, it is possible to generate unique music.

Since a song is generated only when the first parameter set and the second parameter set are not similar, it is possible to generate a song that is dissimilar to previously generated songs. Further, in the first embodiment, the first parameter set and the second parameter set are similar according to the evaluation value according to the degree of similarity for each song parameter P2 between the first parameter set and the second parameter set. It is calculated whether or not. Therefore, it can be determined whether the first parameter set and the second parameter set are similar as a whole.

If specialized knowledge about music is required to understand the music parameter P2, for example, in a configuration where the user U directly specifies the music parameter P2, it is difficult for the user U who does not have specialized knowledge to understand the music parameter P2. Difficult to specify. In contrast, in the first embodiment, the music parameter P2 is generated from the input parameter P1 designated by the user U, so the user U does not have to directly designate the music parameter P2. Therefore, even the user U who does not have specialized knowledge regarding music can easily generate a desired music.

B: Second Embodiment The second embodiment will be described below. In addition, in each configuration illustrated below, for elements whose functions are similar to those in the first embodiment, the reference numerals used in the description of the first embodiment are used, and the detailed description of each is omitted as appropriate.

FIG. 7 is a block diagram illustrating the configuration of a content generation system 300 according to the second embodiment. The content generation system 300 generates content C by adding music generated by the music generation system 100 to a video V captured by the user U using the terminal device 200. For example, a video V used for product introduction, company introduction, tourism PR, etc. is captured by user U's terminal device 200. Specifically, the content generation system 300 includes a music generation system 100 and a terminal device 200. The music generation system 100 and the terminal device 200 can communicate via a communication network 30 such as a mobile communication network or the Internet.

The terminal device 200 images a moving image V using an imaging device. User U inputs information for editing video V (hereinafter referred to as “editing information”) from terminal device 200. The editing information is, for example, information specifying the character string and position of a subtitle to be displayed on the video V, or information specifying processing of the video V, or the like. Further, the user U specifies an input parameter P1 regarding the music to be added to the video V. The contents of the input parameter P1 and the method of specifying the input parameter P1 are the same as in the first embodiment.

The terminal device 200 generates input data Da including a plurality of input parameters P1, and data Db representing editing information (hereinafter referred to as "editing data"). Then, the terminal device 200 transmits the video V, input data Da, and editing data Db to the music generation system 100. That is, in the second embodiment, an element that generates input data Da is installed in the terminal device 200.

The music generation system 100 of the second embodiment includes a communication device 16 in addition to the same elements as the first embodiment (control device 11, storage device 12, display device 13, operation device 14, and sound emitting device 15). . The communication device 16 communicates with the terminal device 200 via the communication network 30.

FIG. 8 is a block diagram illustrating the functional configuration of the music generation system 100 in the second embodiment. The control device 11 of the music generation system 100 of the second embodiment functions as the editing section 117 in addition to the conversion section 112, determination section 113, music generation section 114, and sound source section 115 similar to those of the first embodiment. , the music generation system 100 generates content C according to the video V, input data Da, and editing data Db transmitted from the terminal device 200.

The editing unit 117 edits the video V according to the editing data Db. Any known image processing technology may be used to edit the video V. Note that the administrator of the music generation system 100 may instruct the music generation system 100 about specific editing contents of the video V according to the editing data Db. The music generation system 100 generates content C by adding music according to the input data Da to the edited video V.

The conversion unit 112 generates a first parameter set according to the input data Da transmitted from the terminal device 200. The first parameter set is generated in the same manner as in the first embodiment. Similar to the first embodiment, the determining unit 113 determines whether each of the plurality of second parameter sets and the first parameter set are similar. In other words, each second parameter set stored in the storage device 12 of the second embodiment is a parameter set S that was used to generate music included in content C that was generated in the past by the content generation system 300.

Similar to the first embodiment, the music generation unit 114 generates music according to the first parameter set when it is determined that the first parameter set and the second parameter set are not similar. The sound source section 115 generates an acoustic signal according to the music data M generated by the music generation section 114. Then, the editing unit 117 generates content C by adding the audio signal generated by the sound source unit 115 to the edited video V. The generated content C is transmitted to the terminal device 200.

The second embodiment also achieves the same effects as the first embodiment. Note that it is not essential for the user U of the terminal device 200 to specify the input parameter P1. For example, the administrator of the music generation system 100 may specify the input parameter P1 according to the video V transmitted from the terminal device 200.

Further, the video V may be composed of a plurality of scenes. In the above configuration, the input parameter P1 is specified for each of a plurality of scenes that make up the video V. That is, content C to which music is added for each of a plurality of scenes is generated.

C: Modifications Specific modifications added to each of the above-mentioned embodiments will be exemplified below. Two or more aspects arbitrarily selected from the examples below may be combined as appropriate to the extent that they do not contradict each other.

(1) In each of the above-mentioned embodiments, a configuration is illustrated in which the first parameter set including the plurality of input parameters P1 specified by the user U is obtained, but the method for obtaining the first parameter set is not limited to the above examples. . For example, a configuration in which the user U can directly specify the music parameter P2 is also envisioned.

Further, the conversion unit 112 may obtain the first parameter set according to an image (still image or video) captured by the user U, for example. Specifically, an image captured by the user U is analyzed using a known image analysis technique, and data representing the characteristics of the image (hereinafter referred to as "feature data") is generated. The storage device 12 stores a feature table in which a parameter set S is associated with each of a plurality of different feature data. Then, the conversion unit 112 obtains a parameter set S corresponding to the generated feature data as a first parameter set, from among the plurality of parameter sets S registered in the feature table. That is, the data used to generate the first parameter set is not limited to the input data Da. Input data Da including a plurality of input parameters P1 and various other data (for example, data representing an image) are used to generate the first parameter set. As understood from the above explanation, it is not essential for the user U to specify the input parameter P1. Further, the data generation unit 111 may set the input parameter P1 by analyzing an image captured by the user U, and may generate input data Da including the input parameter P1.

(2) In each of the above embodiments, the weight of the similarity of each music parameter P2 to the evaluation value may be set for each music parameter P2. Specifically, the calculation unit 133 calculates a weighted sum of similarities over a plurality of song parameters P2 as an evaluation value. For example, the weight of the song parameter P2, which has a large influence on the impression of the ultimately generated song, is set to a large numerical value. For example, the song parameters P2 related to "genre," "music composition," and "chord progression" tend to have a greater influence on the impression of the song than the song parameters P2 related to "BPM" and "number of measures." According to the above configuration, the weight of the similarity of the music parameter P2 can be set according to the nature of the music parameter P2. Therefore, it is possible to calculate an evaluation value that accurately reflects the degree of similarity in musical impression between the music corresponding to the first parameter set and the music corresponding to the second parameter set.

Further, among the plurality of music parameters P2, there may be music parameters P2 that are not taken into consideration in calculating the evaluation value. For example, it is not necessary to use the degree of similarity of the music parameter P2 regarding "BPM" and "number of bars" to calculate the evaluation value.

(3) In each of the above embodiments, the method of determining whether the first parameter set and the second parameter set are similar is not limited to the above examples. For example, the determining unit 113 may determine whether the first parameter set and the second parameter set are similar depending on whether or not the respective song parameters P2 match. For example, if the song parameters P2 of the same type do not match, it is determined that the first parameter set and the second parameter set are not similar (that is, different). Specifically, if there is a music parameter P2 that does not match even one of the plurality of music parameters P2 between the first parameter set and the second parameter set, the first parameter set and the second parameter set are determined to be dissimilar. On the other hand, if all the music parameters P2 match between the first parameter set and the second parameter set, it is determined that the first parameter set and the second parameter set are similar (that is, common). Ru. When the first parameter set and the second parameter set are common, typically the generated music is also the same. As illustrated above, it is not essential to calculate the evaluation value.

In addition, in the above configuration, it was determined whether all one music parameter P2 of the first parameter set and all one music parameter P2 of the second parameter set matched/mismatched. However, for example, it may be determined whether a part of one music parameter P2 of the first parameter set matches/mismatches a part of one music parameter P2 of the second parameter set. Further, the determination unit 113 determines whether the first parameter set and the second parameter set are similar, taking into account the match/mismatch of one or more specific music parameters P2 among the plurality of music parameters P2. You may. For example, the match/mismatch of the song parameters P2 regarding "genre", "music composition", and "chord progression" is taken into consideration in determining whether the first parameter set and the second parameter set are similar.

In addition, in a configuration in which each of the first parameter set and the second parameter set is expressed as a vector, the distance (for example, Euclidean distance) or correlation between the first parameter set and the second parameter set may be calculated as the evaluation value. good.

(4) In each of the above-mentioned embodiments, a configuration was exemplified in which different first parameter sets are generated even when the input parameter P1 specified by the user U is common, but the input data Da and the first parameter set are A one-to-one correspondence may also be provided. For example, in the reference table, one first parameter set may be registered for one piece of input data Da.

(5) In each of the above-mentioned embodiments, if the first parameter set is common, the music generated by the music generation unit 114 is also common, but a plurality of different music with a common first parameter set A configuration that generates is also adopted. For example, the music data M may be generated according to the first parameter set including the music parameter P2 and a randomly set element such as a random number. According to the above configuration, various music data M can be generated even when the music parameter P2 is common.

(6) In each of the above embodiments, a configuration in which a plurality of different songs are generated is also adopted. For example, the conversion unit 112 generates a plurality of different first parameter sets corresponding to one piece of input data Da. Then, for each of the plurality of first parameter sets, determination of similarity with the second parameter set and generation of music are performed. The control unit 116 presents the generated plural pieces of music data M to the user U. User U adopts any one of the plurality of songs presented by control unit 116.

(7) In the second embodiment, the input data Da is transmitted from the terminal device 200 to the music generation system 100, but the first parameter set may be generated from the input data Da in the terminal device 200. That is, the first parameter set is transmitted from the terminal device 200 to the music generation system 100. The communication device 16 of the music generation system 100 receives the first parameter set from the terminal device 200. That is, the communication device 16 of the music generation system 100 functions as an element that acquires the first parameter set. In a configuration in which the music generation system 100 receives the first parameter set from the terminal device 200, the data generation unit 111 and the conversion unit 112 are omitted from the music generation system 100.

As understood from the above example, the element that acquires the first parameter set in the music generation system 100 is comprehensively expressed as an "acquisition unit." For example, the conversion unit 112 of the first embodiment and the second embodiment is an example of an acquisition unit. In a configuration in which the first parameter set is transmitted from the terminal device 200 to the music generation system 100, the communication device 16 that receives the first parameter set corresponds to the acquisition unit. Further, for example, in a configuration in which the user U directly specifies the music parameter P2, the element that generates the first parameter set representing the plurality of music parameters P2 corresponds to the acquisition unit.

(8) In each of the above embodiments, the music generation system 100 includes the music generation unit 114, but the music generation unit 114 may be omitted from the music generation system 100. For example, in the second embodiment, the determination unit 113 of the music generation system 100 determines whether the first parameter set and the second parameter set are similar. If the first parameter set is not similar to the second parameter set, the control unit 116 notifies the terminal device 200 that music can be generated. Then, the terminal device 200 generates music data M according to the first parameter set generated by itself. Note that if the first parameter set is similar to the second parameter set, the control unit 116 notifies the terminal device 200 that the music cannot be generated. The user U then instructs the terminal device 200 to change the input parameter P1.

As understood from the examples of modification (7) and modification (8), the function of generating the first parameter set (data generation unit 111 and conversion unit 112) and the function of generating music data M (music generation unit 114) ) may be omitted from the music generation system 100. That is, a configuration in which the music generation system 100 includes only the determination unit 113 that determines whether the first parameter set and the second parameter set are similar is also assumed.

(9) In the configuration of the first embodiment in which the music generation system 100 is realized by an information terminal used by the user U, the music data M may be generated by a server device that can communicate with the music generation system 100. That is, the music generation unit 114 is installed in the server device and is omitted from the music generation system 100.

(10) In each of the above embodiments, the degree of similarity between the music data M generated by the music generation unit 114 according to the first parameter set and the music data M generated by the music generation unit 114 in the past is calculated. Good too. The music data M generated in the past is stored in the storage device 12 in association with the second parameter set used to generate the music data M. Moreover, the music data M generated by the music generation unit 114 according to the first parameter set may be processed and presented to the user U.

(11) In each of the above embodiments, the input data Da used to generate the music may be stored in the storage device 12 together with the parameter set S. In the above configuration, the degree of similarity between the input data Da of the music to be generated and the input data Da used for composing music in the past may be taken into consideration in calculating the evaluation value.

(12) In each of the above-mentioned embodiments, the user U edits a plurality of input parameters P1 represented by input data Da used in the past to generate music, and uses the edited input parameters P1 to generate input data Da may be generated. Input data Da used in the past to generate music is stored in the storage device 12. For example, the user U selects arbitrary input data Da from a plurality of input data Da stored in the storage device 12. A plurality of input parameters P1 represented by the input data Da selected by the user U are displayed on the operation image G. For example, the user U changes only the input parameter P1 specified for "situation" among the plurality of input parameters P1 displayed on the operation image G. Then, input data Da including the changed input parameter P1 is generated.

Similarly, the user U may edit a plurality of music parameters P2 represented by the second parameter set that has been used to generate music in the past, and the music may be generated using the edited music parameters P2.

(13) In each of the above embodiments, in steps Sa5 to Sa7 of FIG. 6, after calculating the evaluation values for all the second parameter sets, it is determined whether there is a second parameter set similar to the first parameter set. However, for each second parameter set, the process of calculating the evaluation value and the process of determining similarity with the first parameter set may be sequentially performed. In the above configuration, a change in the input parameter P1 may be presented to the user U at the time when a second parameter set similar to the first parameter set is detected.

(14) In each of the above embodiments, when there is a second parameter set similar to the first parameter set (Sa7: YES), the conversion unit 112 converts the first parameter set to a first parameter set that is different from the generated first parameter set. was generated. However, as illustrated in FIG. 9, when there is a second parameter set similar to the first parameter set (Sa7: YES), the user U is suggested to change the input parameter P1 by displaying the image on the display device 13. After doing so (Sa11), the input of the input parameter P1 may be accepted again from the user U (Sa1). The conversion unit 112 generates a first parameter set from the input parameters P1 updated by the user U (Sa2, Sa3). However, the configuration of the first embodiment in which the first parameter set is changed without requiring the user U to update the input parameter P1 has the advantage that the load on the user U is reduced.

(15) In each of the above-described embodiments, when a change in the input parameter P1 is proposed to the user U, the input parameter P1 to be changed among the plurality of input parameters P1 may be suggested to the user U. For example, if the input parameter P1 related to "image" is changed and the first parameter set is dissimilar to the second parameter set (the second parameter set determined to be similar to the first parameter set), the corresponding " You may propose to the user U a change in the input parameter P1 regarding "Image".

(16) In each of the above embodiments, the configuration in which the evaluation value increases as the first parameter set and the second parameter set become more similar is illustrated, but the relationship between the degree of similarity and the evaluation value is limited to the above examples. Not done. For example, the calculation unit 133 may calculate the evaluation value such that the evaluation value decreases as the first parameter set and the second parameter set become more similar. In the above configuration, the processing unit 135 determines that the first parameter set and the second parameter set are similar when the evaluation value is less than the threshold value, and determines that the first parameter set and the second parameter set are similar when the evaluation value exceeds the threshold value. It is determined that the sets are not similar.

As can be understood from the above examples, when the evaluation value is on the dissimilar side with respect to the threshold, the determination unit 113 determines that the first parameter set and the second parameter set are not similar, and the evaluation value is the threshold. , it is determined that the first parameter set and the second parameter set are similar. "Similar side with respect to the threshold value" means a range on the positive side of the threshold value in a configuration where the evaluation value increases as parameter sets become more similar, and a range on the positive side with respect to the threshold value in a configuration where the evaluation value decreases as the parameter sets become more similar. means the range on the negative side with respect to the threshold value. Furthermore, in a configuration where the evaluation value increases as parameter sets become more similar, "on the dissimilar side with respect to the threshold value" means a range on the negative side with respect to the threshold value, and the more similar the parameter sets become, the evaluation value increases. In a decreasing configuration, it means a range on the positive side with respect to the threshold value.

(17) In each of the above embodiments, the storage location of the second parameter set is not limited to the storage device 12 of the music generation system 100. For example, a storage device (for example, cloud storage) is prepared separately from the music generation system 100, and the control device 11 reads and writes to the storage device via a communication network such as a mobile communication network or the Internet. It is also possible. When a storage device is prepared separately from the music generation system 100, a configuration is also adopted in which a second parameter set that each of the plurality of music generation systems 100 used in the past to generate music is stored in the storage device. In the above configuration, for example, in each music generation system 100, it is determined whether the second parameter set that the plurality of music generation systems 100 used in the past to generate music is similar to the first parameter set. Therefore, it is possible to easily determine whether or not a song to be generated is similar to a song that has been generated in the past by the plurality of song generation systems 100.

(18) In each of the above embodiments, the type of music generated by the music generation system 100 is arbitrary. For example, the music generation system 100 generates various types of music, such as music consisting only of melodies without singing, music consisting only of sounds of musical instruments, music including singing and performance sounds, and background music (BGM). be done.

(19) In each of the above embodiments, the threshold value to be compared with the evaluation value may be a variable value. For example, the user U specifies the threshold value by operating the operating device 14. Specifically, the user U specifies a threshold value depending on how much he/she wishes to generate a music piece that is different from music that has been generated in the past. For example, if the user U wants to generate a song that is significantly different from songs generated in the past, the user U should set a relatively large threshold value, and if there is no need to make the song different from the songs generated in the past, Make the threshold relatively small.

(20) In each of the above embodiments, the first parameter set related to the song to be generated and the second parameter set used in the past to generate the song are compared, but the comparison by the determination unit 113 is as follows. The invention is not limited to this example. For example, when generating music data M from each of first input data Da1 and second input data Da2 according to an instruction from user U, the first parameter set corresponding to the first input data Da1 and the second input It may also be compared with the second parameter set corresponding to the data Da2. That is, the second parameter set does not necessarily have to be a parameter set used to generate music in the past.

(21) As described above, the functions of the music generation system 100 illustrated above are realized through cooperation between one or more processors configuring the control device 11 and a program stored in the storage device 12. A program according to the present disclosure may be provided in a form stored in a computer-readable recording medium and installed on a 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 any known recording medium such as a semiconductor recording medium or a magnetic recording medium is used. Also included are recording media in the form of. Note that the non-transitory recording medium includes any recording medium except for transitory, propagating signals, and does not exclude volatile recording media. Furthermore, in a configuration in which a distribution device distributes a program via a communication network, the storage device 12 that stores the program in the distribution device corresponds to the above-mentioned non-transitory recording medium. Note that the present disclosure may also be specified as an operating method (music generation method) of the music generation system 100 according to each of the above-described embodiments.

D: Supplementary Note From the configurations exemplified above, the following configurations can be understood, for example.

A music generation method according to one aspect (aspect 1) of the present disclosure is a music generation method that generates a music according to a parameter set including a plurality of music parameters, each of which represents a condition regarding the music, the music generation method comprising: a first parameter set; and the second parameter set are similar. In the above aspect, since it is determined whether the first parameter set and the second parameter set are similar, there is an advantage that it is not necessary to compare the acoustic features of the songs.

The music generation method according to the specific example of aspect 1 (aspect 2) further acquires the first parameter set regarding the music to be generated, and in the determination, the acquired parameter set and the past music It is determined whether the second parameter set used for generation is similar. In the above aspect, it is determined whether or not the first parameter set related to the music to be generated is similar to the second parameter set that was used in the past to generate the music. Since there is no need to compare songs, it is easy to check whether the song you are trying to generate is similar to songs that have been generated in the past (i.e. whether the song you are trying to generate is a unique song). can be determined.

In a specific example of aspect 1 or aspect 2 (aspect 3), if it is determined that the first parameter set and the second parameter set are not similar, generating a song according to the first parameter set, If it is determined that the first parameter set and the second parameter set are similar, no music is generated using the first parameter set. In the above aspect, since a song is generated only when the first parameter set and the second parameter set are not similar, it is possible to generate a song that is dissimilar to a song that has been generated in the past.

In any one of aspects 1 to 3 (aspect 4), in the determination, an evaluation value is calculated according to the similarity of each song parameter between the first parameter set and the second parameter set, If the evaluation value is on the dissimilar side with respect to the threshold value, it is determined that the first parameter set and the second parameter set are not similar; if the evaluation value is on the similar side with respect to the threshold value, it is determined that the first parameter set and the second parameter set are not similar. In this case, it is determined that the first parameter set and the second parameter set are similar. In the above aspect, whether or not the first parameter set and the second parameter set are similar is determined according to the evaluation value according to the degree of similarity for each song parameter between the first parameter set and the second parameter set. is calculated. Therefore, it can be determined whether the first parameter set and the second parameter set are similar as a whole.

In a specific example of aspect 4 (aspect 5), in calculating the evaluation value, a weight related to the degree of similarity of each music parameter to the evaluation value is set for each music parameter. In the above aspect, it is possible to set the weight of similarity of music parameters according to the properties of the music parameters.

In a specific example of any one of aspects 1 to 5 (aspect 6), in acquiring the first parameter set, the first parameter set is generated according to input parameters representing conditions specified by the user regarding the song. be done. In the above aspect, the first parameter set is generated according to the conditions related to the music specified by the user. Therefore, the user does not have to directly specify the conditions expressed by the music parameters included in the first parameter set.

In a specific example of aspect 6 (aspect 7), in acquiring the first parameter set, the first parameter set is generated according to the input parameter and randomly set elements. In the above aspect, the first parameter set is generated according to the music condition specified by the user and randomly set elements. Therefore, various first parameter sets can be obtained.

A music generation system according to one aspect (aspect 8) of the present disclosure includes a music generation unit that generates music according to a parameter set including a plurality of music parameters each representing a condition regarding the music; and a determination unit that determines whether the two parameter sets are similar.

DESCRIPTION OF SYMBOLS 11... Control device, 12... Storage device, 13... Display device, 14... Operating device, 15... Sound emitting device, 30... Communication network, 100... Music generation system, 111... Data generation section, 112... Conversion section, 113... Judgment unit, 114...Music generation unit, 115...Sound source unit, 116...Control unit, 117...Editing unit, 133...Calculation unit, 135...Processing unit, 200...Terminal device, 300...Content generation system, C...Content, Da ...Input data, Db...Editing data, G...Operation image, M...Music data, P1...Input parameter, P2...Music parameter, PR...Sightseeing, S...Parameter set, U...User, V...Video.

Claims (8)

A music generation method that generates music according to a parameter set including a plurality of music parameters each representing a condition regarding the music, the method comprising:
A music generation method implemented by a computer that determines whether a first parameter set and a second parameter set are similar. The music generation method further includes:
obtaining the first parameter set related to the song to be generated;
The music generation method according to claim 1, wherein in the determination, it is determined whether the acquired first parameter set and the second parameter set used in the past to generate music are similar. If it is determined that the first parameter set and the second parameter set are not similar, generating a song according to the first parameter set,
3. The music generation method according to claim 1, wherein if it is determined that the first parameter set and the second parameter set are similar, a music piece using the first parameter set is not generated. In the above judgment,
calculating an evaluation value according to the degree of similarity for each song parameter between the first parameter set and the second parameter set;
If the evaluation value is on the dissimilar side with respect to the threshold value, it is determined that the first parameter set and the second parameter set are not similar;
The music generation method according to any one of claims 1 to 3, wherein if the evaluation value is on the similar side with respect to the threshold value, it is determined that the first parameter set and the second parameter set are similar. 5. The music generation method according to claim 4, wherein in calculating the evaluation value, a weight related to the degree of similarity of each of the music parameters to the evaluation value is set for each music parameter. 6. The music generation method according to claim 1, wherein in acquiring the first parameter set , the first parameter set is generated according to input parameters representing conditions specified by the user for the music. 7. The music generation method according to claim 6, wherein in obtaining the first parameter set, the first parameter set is generated according to the input parameters and randomly set elements. a music generation unit that generates music according to a parameter set including a plurality of music parameters each representing a condition regarding the music;
A music generation system comprising: a determination unit that determines whether a first parameter set and a second parameter set are similar.

Images