JP7254359B2 - Programs, devices, and methods of producing scripts - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Law Posting date: February 17, 2020 Posting address: https://prtimes. jp/main/html/rd/p/000000002.000054277. html

The present invention relates to a program, an apparatus, and a script production method.

Patent Literature 1 describes "to provide a text generation device, a text generation program, and a text generation method capable of accurately correcting text errors".
[Prior art documents]
[Patent Literature]
[Patent Document 1] JP 2020-34704 A

In a first aspect of the invention, a program is provided. The program may cause the computer to function as a first acquisition unit that acquires a plurality of synopsis segments. The program may cause the computer to function as a second acquisition unit that acquires sentiment scores set for each of the plurality of synopsis segments. The program may cause the computer to function as a selector that selects, for each of a plurality of synopsis segments, text from a group of texts that corresponds to the content of the synopsis segment and to the sentiment score of the synopsis segment. . The program may cause the computer to act as a generator that arranges each selected text into corresponding synopsis segments to generate a script.

The selection unit calculates a first degree of suitability of each text for the one synopsis segment based on the sentiment score obtained by analyzing each text and the sentiment score set for the one synopsis segment. 1 calculator. The selection unit may have a second calculation unit that calculates a second degree of relevance of each text to the one synopsis segment based on the degree of similarity in content between each text and the one synopsis segment. The selector may comprise an evaluator for calculating a relevance rating for the one synopsis segment based on the first relevance and the second relevance for each text. The selector may select the text corresponding to the one synopsis segment based on the match score.

The evaluation unit may calculate the fitness evaluation value by weighted addition of the first fitness and the second fitness.

The program may cause the computer to function as a segmenter that acquires the synopsis, segments it into multiple synopsis segments, and supplies the synopsis segments to the first acquirer.

The segmentation unit may generate a synopsis segment for each section obtained by dividing a work produced by realizing a script. The second acquisition unit may acquire the sentiment score set for each section.

The program may cause the computer to function as a supplier that supplies the text selected by the selector for the synopsis segment corresponding to each segment to the segmenter as a synopsis for that segment.

The program may cause the computer to function as a repeat controller that causes the first acquirer, the segmenter, the second acquirer, the selector, and the feeder to repeat the process until the amount of script exceeds the reference amount.

The program may cause the computer to function as a text generator that generates a plurality of random texts in natural language from a plurality of script data. The program may cause the computer to function as a storage unit that stores a plurality of texts generated by the text generation unit as a text group.

Each text in the text group may include a headline, dialogue and topic.

In a second aspect of the invention, an apparatus is provided. The apparatus may comprise a first obtaining unit for obtaining a plurality of synopsis segments into which the synopsis is segmented. The apparatus may comprise a second obtaining unit that obtains a sentiment score set for each of the plurality of synopsis segments. The apparatus may comprise, for each of a plurality of synopsis segments, a selector that selects text from the text group that corresponds to the content of the synopsis segment and to the sentiment score of the synopsis segment. The apparatus may comprise a generator for arranging each selected text into corresponding synopsis segments to generate a script.

In a third aspect of the invention, a method for producing a screenplay is provided. The script production method may comprise a first obtaining step of obtaining a plurality of synopsis segments into which the synopsis is segmented. The script production method may comprise a second obtaining step of obtaining a sentiment score assigned to each of the plurality of plot segments. The script production method may comprise, for each of a plurality of synopsis segments, selecting text from a set of texts corresponding to the content of the synopsis segment and corresponding to the sentiment score of the synopsis segment. The script production method may comprise a generation step of arranging each of the selected texts into corresponding synopsis segments to generate the script.

It should be noted that the above summary of the invention does not list all the necessary features of the invention. Subcombinations of these feature groups can also be inventions.

1 shows a device 1 according to the present embodiment; The operation of the device 1 is shown. An example of operation results is shown. An example computer 2200 is shown in which aspects of the present invention may be embodied in whole or in part.

Hereinafter, the present invention will be described through embodiments of the invention, but the following embodiments do not limit the invention according to the claims. Also, not all combinations of features described in the embodiments are essential for the solution of the invention.

[1. Device]
FIG. 1 shows a device 1 according to this embodiment. The apparatus 1 generates a script, and includes a text generation unit 101, a storage unit 102, a synopsis acquisition unit 103, a segmentation unit 104, a synopsis segment acquisition unit 105, a sentiment score acquisition unit 106, a selection unit 110, and a supply unit 121. , a generation unit 122 and a repetition control unit 124 .

[1.1. Text generation unit 101]
A text generation unit 101 generates a plurality of random texts in natural language from a plurality of script data. The text generator 101 may acquire a plurality of script data from outside the device 1 . The text generation unit 101 may generate a plurality of meaningful natural language texts by performing conventionally known short sentence generation processing using the acquired script data. As an example, the text generation unit 101 may generate a plurality of texts by dividing the screenplay data by morphological analysis and then randomly combining them. Each text may be used as a script box (a summary of key points and information for each scene (also referred to as a scene or box)), and may include a script slogan, lines, and story. The stilts may indicate the location of the scene, the time of day, and so on.

In this embodiment, as an example, the text generation unit 101 may acquire the sentiment score of each generated text. A sentiment score is an evaluation value of emotion (also referred to as sentiment), and may indicate, for example, the degree of negative/positive. The text generation unit 101 may acquire the sentiment score of the text by subjecting the text to conventionally known sentiment analysis (for example, the sentiment analysis of Cloud Natural Language API by Google (registered trademark)). The text generator 101 may supply the text to an external device that performs sentiment analysis and obtain the sentiment score of the text.

The text generation unit 101 may supply each generated text to the storage unit 102 . When the sentiment score of each text is acquired, the text generation unit 101 may supply the sentiment score to the storage unit 102 in association with the text.

[1.2. Storage unit 102]
The storage unit 102 stores a plurality of texts generated by the text generation unit 101 as a text group. For example, the storage unit 102 may store more than 1000 sentences. When the text and the sentiment score are supplied from the text generation unit 101, the storage unit 102 may store them in association with each other.

[1.3. Synopsis acquisition unit 103]
The synopsis acquisition unit 103 acquires a synopsis (also referred to as a logline, plot, or story) of the work. A work may be a work created by realizing a script generated by the apparatus 1, and may be, for example, a drama, a movie, or a stage play. The synopsis acquisition unit 103 may acquire the synopsis from the operator. The synopsis acquisition unit 103 may supply the acquired synopsis to the segmentation unit 104 .

[1.4. segment portion 104]
The segmentation unit 104 acquires the synopsis from the synopsis acquisition unit 103 and segments the synopsis into a plurality of synopsis segments. The segmentation unit 104 may generate a synopsis segment for each segment obtained by dividing the work. The segmenter 104 may generate a single synopsis segment for each segment. The segmentation unit 104 supplies each synopsis segment obtained by segmentation to the synopsis segment acquisition unit 105 .

[1.5. Synopsis segment acquisition unit 105]
The synopsis segment acquisition unit 105 is an example of a first acquisition unit and acquires a plurality of synopsis segments from the segmentation unit 104 . The synopsis segment acquisition unit 105 may supply the acquired synopsis segment to the selection unit 110 . As an example in this embodiment, the synopsis segment acquisition unit 105 may supply the corresponding synopsis segment to the selection unit 110 for each section.

[1.6. Sentiment score acquisition unit 106]
The sentiment score acquisition unit 106 is an example of a second acquisition unit, and acquires sentiment scores set for each of a plurality of synopsis segments. The sentiment score set in the synopsis segment may indicate an evaluation value of emotion that should be caused by the part of the work that realizes the synopsis segment, and may be arbitrarily set by the operator. As an example in the present embodiment, since a synopsis segment is generated for each segment, the sentiment score acquisition unit 106 obtains the sentiment score set for each segment by adding the sentiment score set for the synopsis segment corresponding to the segment. It can be obtained as a score.

The sentiment score acquisition unit 106 may acquire sentiment scores in the form of a graph (also referred to as a plot line) or a table showing the relationship between each section in the work and the sentiment score set for that section.

The sentiment score acquisition unit 106 may supply the acquired sentiment score for each synopsis segment (in this embodiment, for each section as an example) to the selection unit 110 .

[1.7. Selection unit 110]
For each of a plurality of synopsis segments, the selection unit 110 selects the sentiment score of the synopsis segment (as an example, the sentiment score of the section corresponding to the synopsis segment in this embodiment) that corresponds to the content of the synopsis segment. from the text group in the storage unit 102. Since the selected text is used by the generation unit 122 as box writing for the screenplay, it is also referred to as box writing text. Selection unit 110 includes sentiment score matching calculation unit 111 , content matching calculation unit 112 , and evaluation unit 113 .

[1.7.1. Sentiment score matching calculation unit 111]
The sentiment score matching degree calculation unit 111 is an example of a first calculation unit. Based on the sentiment score set to , the degree of conformity (referred to as the first degree of conformity) of each text to the one synopsis segment is calculated.

In this embodiment, as an example, the sentiment score matching calculation unit 111 acquires the sentiment score associated with each text in the storage unit 102 and the sentiment score set in the section corresponding to one synopsis segment. A first relevance for each text is calculated based on the sentiment score provided by the unit 106 . The sentiment score matching degree calculation unit 111 may calculate the first matching degree higher as the sentiment scores are closer.

The sentiment score matching calculation unit 111 supplies the evaluation unit 113 with the first matching calculated for each text for each synopsis segment (in this embodiment, for each section as an example).

[1.7.2. Content matching degree calculation unit 112]
The content adaptability calculation unit 112 is an example of a second calculation unit, and calculates the adaptability of each text to the one synopsis segment (second adaptability degree) is calculated.

For example, the content matching degree calculation unit 112 may determine the degree of similarity from words included in the text. In this embodiment, as an example, the content matching degree calculation unit 112 evaluates the degree of importance of words in the text by the TF/IDF method, and calculates a higher degree of similarity as the degrees of importance are closer. The content adaptability calculator 112 may use the similarity calculated between the text and the synopsis segment as the second adaptability of the text.

The content suitability calculation unit 112 supplies the evaluation unit 113 with the second suitability calculated for each text for each synopsis segment (in this embodiment, for each section as an example).

[1.7.3. Evaluation unit 113]
The evaluation unit 113 calculates a matching evaluation value for one synopsis segment based on the first matching level and the second matching level for each text. For example, the evaluation unit 113 may calculate the fitness evaluation value by weighted addition of the first fitness and the second fitness. A weighting factor for each of the first goodness of fit and the second goodness of fit may be set arbitrarily. The evaluation unit 113 may select any text corresponding to one synopsis segment as the boxed text based on the matching evaluation value. As an example, the evaluation unit 113 may select the text with the highest matching evaluation value as the boxed text. The evaluation unit 113 may supply the selected boxed text to the supply unit 121 .

[1.8. supply unit 121]
The supply unit 121 supplies the boxed text selected by the selection unit 110 to the repetition control unit 124 and the synopsis acquisition unit 103 and/or generation unit 122 . When the boxed text is supplied from the supply unit 121 to the synopsis acquisition unit 103, the boxed text selected for the synopsis segment corresponding to each section is supplied as the synopsis for the section, and is supplied to the synopsis acquisition unit 103. may be supplied to the segment 104 from.

[1.9. generation unit 122]
The generation unit 122 arranges the boxed texts supplied from the supply unit 121 in the order of the synopsis segments corresponding to the boxed texts (in this embodiment, as an example, the order of the sections corresponding to the synopsis segments) to generate a script. The generation unit 122 may output the generated script to the outside of the device 1 .

[1.11. Repeat control unit 124]
The repetition control unit 124 causes the synopsis segment acquisition unit 105, the segmentation unit 104, the sentiment score acquisition unit 106, the selection unit 110, and the supply unit 121 to repeatedly perform processing. The repetition control unit 124 may supply the synopsis segment acquisition unit 105, the segmentation unit 104, the sentiment score acquisition unit 106, the selection unit 110, and the supply unit 121 with a signal CTRL that controls whether or not to execute processing.

For example, when causing the segmentation unit 104 to perform new processing by repetition, the repetition control unit 124 may generate a synopsis segment for each segment obtained by recursively dividing the work. To divide a work recursively means that the first division divides the work into multiple sections, the second division divides each of the obtained sections into multiple smaller sections, and then divides the work in the same way. It can be done.

Further, when causing the supply unit 121 to perform new processing by repetition, the repetition control unit 124 causes the synopsis acquisition unit 103 to supply the boxed text selected by the selection unit 110, and causes the supply unit 121 to perform the new processing. If not, the selected boxed text may be supplied to the generator 122 . As a result, when box-written texts are selected for sections recursively divided by repetition, the script is generated by arranging the box-written texts corresponding to each of the smallest sections in the order of the sections.

The repetition control unit 124 may cause the processing to be repeated until the amount of script generated by the generation unit 122 exceeds the reference amount. The repetition control unit 124 determines whether or not the amount of the script exceeds the reference amount based on the sum of the amount of boxed text supplied from the supply unit 121 (for example, the number of characters, the number of words, the number of sentences, and the number of sentences). can be determined. The reference amount may be set arbitrarily.

According to the device 1 described above, for each of a plurality of synopsis segments, text corresponding to the contents of the synopsis segment and corresponding to the sentiment score of the synopsis segment is selected as boxed text from the text group and sequentially selected. They are arranged and a script is generated. Therefore, by setting a sentiment score for each synopsis segment, it is possible to obtain a script that follows the content of the synopsis segment and that causes emotions according to the sentiment score.

Further, based on the sentiment score of each text and the sentiment score of one synopsis segment, a first degree of conformity of each text to the one synopsis segment is calculated, and the content of each text and one synopsis segment is calculated. Based on the degree of similarity, a second degree of relevance of each text to the one synopsis segment is calculated, and based on these degrees of relevance, a relevance evaluation value for the one synopsis segment is calculated to select boxed text. Therefore, it is possible to obtain a script that reliably matches the content of the synopsis segment and the sentiment score.

In addition, since the fitness evaluation value is calculated by weighting addition of the first fitness and the second fitness, it is possible to adjust the weight between the content of the synopsis segment and the sentiment score.

Also, since the synopsis of the work is segmented into a plurality of synopsis segments, by inputting the synopsis, synopsis segments can be generated and the script can be obtained.

In addition, since a synopsis segment is generated for each segment into which the work is divided, by inputting the synopsis of the work, a synopsis segment is generated for each segment, and the text in line with the content of the synopsis segment of the relevant segment is displayed as boxed text. is selected as In addition, since the sentiment score set for each section is acquired, by setting the sentiment score for each section of the work, text that evokes emotions in each section according to the sentiment score of the section is selected as boxed text. Therefore, by inputting the synopsis of the work and setting the sentiment score for each section, a script that follows the content of the corresponding synopsis segment in each section and causes emotions according to the corresponding sentiment score is obtained. be able to.

In addition, since the boxed text selected for the synopsis segment corresponding to each section is further supplied to the segmentation unit 104 as a synopsis for that section, the synopsis for each section is further segmented into synopsis segments for smaller sections, Boxed text is selected for each synopsis segment after the segment. Therefore, by supplying a synopsis of the work, it is possible to obtain a script including boxed text for each segment of the work. Therefore, it is possible to easily acquire a script for a long work.

In addition, since the process is repeated until the amount of script exceeds the reference amount, it is possible to obtain an arbitrary amount of script.

Also, since a plurality of random texts for box writing are generated from a plurality of script data and stored as a text group in the storage unit 102, various texts for box writing can be included in the text group. In addition, since text in natural language is generated from script data, it is possible to save the trouble of preparing high-quality script data.

In addition, since each text includes a headline, lines, and a story line, it is possible to reliably generate a script by arranging the texts.

[2. motion]
FIG. 2 shows the operation of device 1 . The device 1 generates a screenplay by executing the processes of steps S11 to S27. In this embodiment, as an example, it is assumed that a plurality of texts and their sentiment scores are associated and stored in the storage unit 102 at the start of operation.

In step S11, the synopsis acquisition unit 103 and the sentiment score acquisition unit 106 respectively acquire the synopsis and plotline of the work from the operator or the like.

In step S13, the segmentation unit 104 acquires a synopsis segment for each section obtained by dividing the work. When the repetition control unit 124 repeats the process of step S13 via step S23, which will be described later, the segmentation unit 104 may obtain a synopsis segment for each segment into which the work is recursively segmented. In the present embodiment, as an example, the work may be recursively divided into four equal sections for each process of step S13. In the process of step S13 from the second time onward, a synopsis segment may be obtained for each section obtained by further dividing each section obtained by the previous process of step S13 into four equal parts. Note that the segmentation unit 104 does not necessarily have to perform calculations for dividing the work as long as the number of segments after division can be obtained, and if synopsis segments are generated for each of a plurality of segments obtained when the work is segmented. good.

The segmentation unit 104 may generate a synopsis segment by segmenting the synopsis. As an example, the segmentation unit 104 may segment the synopsis by the number of characters, or segment it by semantic breaks detected by morphological analysis. When segmenting the synopsis by the number of characters, the segmentation unit 104 may equally divide the synopsis into equal numbers of characters, or may segment the synopsis by the number of characters of a preset ratio.

In step S15, the selection unit 110 selects one of the four sections as the selection section. The selection unit 110 may select the earliest (leading) section from among the unselected sections.

In step S17, the sentiment score acquisition unit 106 acquires the sentiment score of the selected section. In the present embodiment, as an example, since the sentiment score acquisition unit 106 acquires the plot line in which the sentiment score is set for each section in step S1, the sentiment score of the selected section is acquired from the plot line. you can

In step S19, the selection unit 110 selects from the text group of the storage unit 102 text corresponding to the content of the synopsis segment of the selected section and corresponding to the sentiment score of the selected section as boxed text. For example, the sentiment score matching calculation unit 110 of the selecting unit 110 calculates the first matching of each text, and the content matching calculating unit 112 of the selecting unit 110 calculates the second matching of each text. Then, the evaluation unit 113 of the selection unit 110 calculates a matching evaluation value for each text based on the first matching degree and the second matching degree of the text, and selects one of the texts with the highest matching evaluation value in a box. Select as text.

In step S21, the repetition control unit 124 determines whether or not all sections associated with synopsis segments have been selected in step S13, that is, whether or not boxed text has been selected for all sections. The repetition control unit 124 counts down the number of all sections associated with the synopsis segment in step S13 for each process in step S19, and when the count value reaches 0, it is determined that all sections have been selected. You can judge.

When it is determined in step S21 that all sections have not been selected (step S21; No), the repetition control unit 124 proceeds to step S15. As a result, the processing of steps S15 to S19 is repeated to select all the sections and select the boxed text. When it is determined in step S21 that all sections have been selected (step S21; Yes), the repetition control unit 124 shifts the process to step S23.

In step S23, the repetition control unit 124 determines whether or not the amount of script generated by the generation unit 122 exceeds a reference amount. As an example in this embodiment, the repetition control unit 124 determines whether or not the total amount of boxed text supplied from the supply unit 121 exceeds a reference amount. When it is determined in step S23 that the script amount exceeds the reference amount (step S23; Yes), the repeat control unit 124 shifts the process to step S27. If it is determined in step S23 that the script amount does not exceed the reference amount (step S23; No), the repeat control unit 124 shifts the process to step S25. As a result, as will be described later, the process of step S13 is performed again after the process of step S25, and synopsis segments are generated for each section into which the work is recursively further divided.

Here, when the process of step S23 is performed, the boxed text has already been selected for each section obtained by recursively dividing the work by the number of times the process of step S13 has been performed. Therefore, the repetition control unit 124 may determine whether or not the total amount of boxed text exceeds the reference amount based on the number of times the process of step S13 has been performed.

The number of times the process of step S13 has been performed so far is equal to the number of times the process of step S23 has been performed so far plus one. Therefore, the repetition control unit 124 may use a number obtained by adding 1 to the number of times the process of step S23 has been performed by itself instead of the number of times the process of step S13 has been performed.

In step S<b>25 , the synopsis acquisition unit 103 acquires the boxed text selected for each section from the supply unit 121 as a synopsis of the section, and supplies it to the segmentation unit 104 . As a result, the process proceeds to step S13 described above.

However, when the process of step S13 is performed after the process of step S25, the segmentation unit 104 generates a synopsis segment by segmenting the synopsis of each of the plurality of segments before the segmentation into four segments, and divides the segment into four equal parts. can be associated with each section. For example, when synopses of four segments (also referred to as first to fourth segments) have been supplied to the segmentation unit 104 by the process of step S25, the segmentation unit 104 divides the synopsis of the first segment into four segments. A synopsis segment may be segmented to generate a synopsis segment and associated with each quadrant of the first segment, the synopsis of the second segment may be segmented to generate a synopsis segment, and the second segment may be divided into four Each equally divided section may be associated, and the same processing may be performed for the third and fourth sections.

Then, in step S27, the generating unit 122 generates a screenplay by arranging the boxed texts selected for each section in the order of the sections.

[3. Operation example]
FIG. 3 shows an example of operation results. "Synopsis" in the figure indicates the synopsis of the work. "Large box" and "small box" correspond to sections within the work, and their contents indicate boxed text. The numbers on the right end indicate the sentiment score set for each section.

As shown in this figure, according to the apparatus 1 according to the present embodiment, by inputting the synopsis shown in the figure and the sentiment score for each section, the content of the synopsis segment corresponding to each section is displayed. , and a script that evokes emotion according to the corresponding sentiment score can be obtained.

[4. Modification]
In this modified example, the content of the headline (for example, the location of the scene, the time period, etc.) is not included in the box-written text, but is used as an additional attribute of the box-written text.

In this case, the text generation unit 101 generates at least a text indicating the storyline (also referred to as text of boxed text) from a plurality of script data, and attribute data indicating the location and time of the scene, etc., and the storage unit 102 can be stored in Further, the text generation unit 101 may acquire the sentiment score of each text of the generated boxed text, and store it in the storage unit 102 in association with the corresponding boxed text. Similarly, the text generation unit 101 may acquire the sentiment score of each generated attribute data and store it in the storage unit 102 in association with the corresponding attribute data.

Further, for each of the plurality of synopsis segments, the selection unit 110 selects, from the text group in the storage unit 102, the boxed body text corresponding to the contents of the synopsis segment and corresponding to the sentiment score of the synopsis segment. You may choose. Similarly, for each of a plurality of synopsis segments, the selection unit 110 stores attribute data corresponding to the content of the synopsis segment and corresponding to the sentiment score of the synopsis segment. You can choose from the data.

As an example, when selecting attribute data corresponding to one synopsis segment, the selection unit 110 selects the sentiment score of each attribute data by the sentiment score matching calculation unit 111 and the sentiment score set for the one synopsis segment. A third relevance of each attribute data to the one synopsis segment may be calculated based on the comment score. Further, the selection unit 110 calculates a fourth degree of conformity of each attribute data to the one synopsis segment based on the degree of similarity between the contents of each attribute data and the one synopsis segment by the content adaptability calculation unit 112. You can Then, the selecting unit 110 calculates the matching evaluation value for one synopsis segment based on the third matching degree and the fourth matching degree for each attribute data by the evaluating unit 113, and selects the attribute data corresponding to the one synopsis segment. You may choose.

Furthermore, the generation unit 122 may generate a script by arranging pairs of boxed body text and attribute data selected for each synopsis segment by the selection unit 110 in the order of the synopsis segment.

[5. Other Modifications]
It should be noted that in the above-described embodiment and modification, the device 1 is provided with the text generation unit 101, the storage unit 102, the synopsis acquisition unit 103, the segmentation unit 104, the supply unit 121, and the repetition control unit 124. , it is also possible not to provide any of these. For example, when the device 1 does not include the text generation unit 101 , the text group may be acquired from the outside and stored in the storage unit 102 . Also, when the device 1 does not include the storage unit 102, it may be externally connected to a storage device that stores text groups. Further, when the apparatus 1 does not include the synopsis acquisition unit 103 or the segmentation unit 104, the synopsis segment acquisition unit 105 may acquire a synopsis segment from outside the apparatus 1 to generate a script. Also, when the device 1 does not include the supply unit 121 or the repetition control unit 124, the boxed text selected for the initial synopsis segment may be arranged as is to generate a script.

Further, although the repetition control unit 124 has been described as determining whether or not the amount of the script exceeds the reference amount based on the total amount of boxed text supplied from the supply unit 121, the determination can be made by other methods. you can go For example, the repetition control unit 124 may make a determination based on the number of times any one of the segmentation unit 104, the synopsis segment acquisition unit 105, the supply unit 121, the selection unit 110, etc. performs processing. The determination may be made based on the amount of script being generated.

Further, although the repetition control unit 124 repeats the process until the amount of script exceeds the reference amount, it may be repeated a fixed number of times. Alternatively, the repetition control unit 124 may repeat until the granularity of the section is less than the reference granularity. In this case, the repetition control unit 124 may repeat the process until boxed text is selected for the synopsis segment corresponding to each section smaller than the reference granularity obtained by recursively dividing the work. The granularity of the section may indicate the time length of the section, or may be the reciprocal of the number of sections (that is, the ratio of one section to the entire work). When the granularity of the section indicates the time length of the section, the device 1 may be provided with a time length acquisition unit that acquires the time length of the work from the operator or the like and supplies it to the repeat control unit 124 .

Also, although the text generation unit 101 has been described as generating text from script data and acquiring the sentiment score of the text, the text generation unit 101 does not have to acquire the sentiment score. In this case, when the sentiment score matching calculation unit 111 of the selecting unit 110 calculates the first matching of the text, the sentiment score of the text may be acquired.

Also, although it has been described that the sentiment score is set for each section into which the work is divided, it may be set for each point in the work. As an example, sentiment scores may be set for multiple time points within the interval. In this case, the sentiment score acquisition unit 106 may acquire a representative value of a plurality of sentiment scores set within the section as the sentiment score of the section. The representative value may be the average value or median value of a plurality of sentiment scores within the interval, or may be the sentiment score set at the beginning, end, or center of the interval. Also, the representative value may be a value derived from an arbitrary function. The representative value derivation function may be modularized.

In addition, although the content matching calculation unit 112 has been described as calculating the second matching from the content similarity between the text and the synopsis segment, in addition/instead of this, the content similarity between the text and the entire synopsis can be calculated. A second goodness of fit may be calculated from the degree.

In addition, it was explained that when boxed text is selected for a recursively divided section, the script is generated by arranging the boxed text corresponding to each of the smallest sections in the order of the section. A screenplay may be generated by arranging boxed text corresponding to each section in order of section regardless of length.

Additionally, various embodiments of the invention may be described with reference to flowchart illustrations and block diagrams, where blocks represent (1) steps in a process in which operations are performed or (2) roles that perform operations. may represent a section of equipment that has Certain steps and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer readable instructions stored on a computer readable medium, and/or processor provided with computer readable instructions stored on a computer readable medium. you can Dedicated circuitry may include digital and/or analog hardware circuitry, and may include integrated circuits (ICs) and/or discrete circuitry. Programmable circuits include logic AND, logic OR, logic XOR, logic NAND, logic NOR, and other logic operations, memory elements such as flip-flops, registers, field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), etc. and the like.

Computer-readable media may include any tangible device capable of storing instructions to be executed by a suitable device, such that computer-readable media having instructions stored thereon may be designated in flowcharts or block diagrams. It will comprise an article of manufacture containing instructions that can be executed to create means for performing the operations described above. Examples of computer-readable media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, and the like. More specific examples of computer readable media include floppy disks, diskettes, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), Electrically Erasable Programmable Read Only Memory (EEPROM), Static Random Access Memory (SRAM), Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD), Blu-ray (RTM) Disc, Memory Stick, Integration Circuit cards and the like may be included.

The computer readable instructions may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state setting data, or object oriented programming such as Smalltalk, JAVA, C++, etc. language, and any combination of one or more programming languages, including conventional procedural programming languages, such as the "C" programming language or similar programming languages. good.

Computer readable instructions may be transferred to a processor or programmable circuitry of a general purpose computer, special purpose computer, or other programmable data processing apparatus, either locally or over a wide area network (WAN), such as a local area network (LAN), the Internet, or the like. ) and may be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, and the like.

FIG. 4 illustrates an example computer 2200 upon which aspects of the invention may be implemented in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as one or more sections of an operation or apparatus associated with an apparatus according to embodiments of the invention, or may Sections may be executed and/or computer 2200 may be caused to execute processes or steps of such processes according to embodiments of the present invention. Such programs may be executed by CPU 2212 to cause computer 2200 to perform certain operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

Computer 2200 according to this embodiment includes CPU 2212 , RAM 2214 , graphics controller 2216 , and display device 2218 , which are interconnected by host controller 2210 . Computer 2200 also includes input/output units such as communication interface 2222, hard disk drive 2224, DVD-ROM drive 2226, and IC card drive, which are connected to host controller 2210 via input/output controller 2220. there is The computer also includes legacy input/output units such as ROM 2230 and keyboard 2242 , which are connected to input/output controller 2220 through input/output chip 2240 .

CPU 2212 operates according to programs stored in ROM 2230 and RAM 2214, thereby controlling each unit. Graphics controller 2216 retrieves image data generated by CPU 2212 into itself, such as a frame buffer provided in RAM 2214 , and causes the image data to be displayed on display device 2218 .

Communication interface 2222 communicates with other electronic devices over a network. Hard disk drive 2224 stores programs and data used by CPU 2212 within computer 2200 . DVD-ROM drive 2226 reads programs or data from DVD-ROM 2201 and provides programs or data to hard disk drive 2224 via RAM 2214 . The IC card drive reads programs and data from IC cards and/or writes programs and data to IC cards.

ROM 2230 stores therein programs that are dependent on the hardware of computer 2200, such as a boot program that is executed by computer 2200 upon activation. Input/output chip 2240 may also connect various input/output units to input/output controller 2220 via parallel ports, serial ports, keyboard ports, mouse ports, and the like.

A program is provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The program is read from a computer-readable medium, installed in hard disk drive 2224 , RAM 2214 , or ROM 2230 , which are also examples of computer-readable medium, and executed by CPU 2212 . The information processing described within these programs is read by computer 2200 to provide coordination between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing the manipulation or processing of information in accordance with the use of computer 2200 .

For example, when communication is performed between the computer 2200 and an external device, the CPU 2212 executes a communication program loaded in the RAM 2214 and sends communication processing to the communication interface 2222 based on the processing described in the communication program. you can command. The communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in a recording medium such as the RAM 2214, the hard disk drive 2224, the DVD-ROM 2201, or an IC card under the control of the CPU 2212, and transmits the read transmission data. Data is transmitted to the network, or received data received from the network is written to a receive buffer processing area or the like provided on the recording medium.

In addition, the CPU 2212 causes the RAM 2214 to read all or necessary portions of files or databases stored in external recording media such as a hard disk drive 2224, a DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. Various types of processing may be performed on the data in RAM 2214 . CPU 2212 then writes back the processed data to the external recording medium.

Various types of information, such as various types of programs, data, tables, and databases, may be stored on recording media and subjected to information processing. CPU 2212 performs various types of operations on data read from RAM 2214, information processing, conditional decision making, conditional branching, unconditional branching, and information retrieval, as specified throughout this disclosure and by instruction sequences of programs. Various types of processing may be performed, including /replace, etc., and the results written back to RAM 2214 . In addition, the CPU 2212 may search for information in a file in a recording medium, a database, or the like. For example, if a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 2212 determines that the attribute value of the first attribute is specified. search the plurality of entries for an entry that matches the condition, read the attribute value of the second attribute stored in the entry, and thereby associate it with the first attribute that satisfies the predetermined condition. an attribute value of the second attribute obtained.

The programs or software modules described above may be stored in a computer readable medium on or near computer 2200 . Also, a recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium, thereby providing the program to the computer 2200 via the network. do.

Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It is obvious to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the description of the scope of claims that forms with such modifications or improvements can also be included in the technical scope of the present invention.

The execution order of each process such as actions, procedures, steps, and stages in the devices, systems, programs, and methods shown in the claims, the specification, and the drawings is particularly "before", "before etc., and it should be noted that they can be implemented in any order unless the output of the previous process is used in the subsequent process. Regarding the operation flow in the claims, the specification, and the drawings, even if the description is made using "first," "next," etc. for the sake of convenience, it means that it is essential to carry out in this order. not a thing

1 device
101 text generator
102 storage unit
103 Synopsis Acquisition Unit
104 segments
105 Synopsis Segment Acquisition Unit
106 Sentiment Score Acquisition Unit
110 selection unit
111 Sentiment score matching calculation unit
112 content relevance calculator
113 Evaluation Department
121 supply unit
122 Generator
124 repeat control unit
2200 computer
2201 DVD-ROM
2210 host controller
2212 CPUs
2214 RAM
2216 graphics controller
2218 display device
2220 Input/Output Controller
2222 communication interface
2224 hard disk drive
2226 DVD-ROM drive
2230 ROMs
2240 input/output chip
2242 keyboard

Claims (10)

the computer,
a first acquiring unit for acquiring a plurality of synopsis segments obtained by segmenting a synopsis into a preset number of segments for a screenplay to be created ;
a second acquisition unit that acquires the sentiment score set for each of the sections as the sentiment score of the synopsis segment corresponding to the section ;
a selection unit for selecting, from a group of texts, text corresponding to the content of the synopsis segment and corresponding to the sentiment score of the synopsis segment for each of the plurality of synopsis segments;
A program that functions as a generating unit that generates the script by arranging each text selected for each synopsis segment in the order of the sections corresponding to the synopsis segment. The selection unit
a first calculation unit that calculates a first degree of suitability of each text for one synopsis segment based on the sentiment score obtained by analyzing each text and the sentiment score set for the one synopsis segment; ,
a second calculation unit that calculates a second degree of conformity of each text to the one synopsis segment based on the degree of similarity between the content of each text and the one synopsis segment;
an evaluation unit that calculates a relevance evaluation value for the one synopsis segment based on the first relevance and the second relevance for each text;
has
2. The program of claim 1, wherein text corresponding to said one synopsis segment is selected based on said relevance score. 3. The program according to claim 2, wherein said evaluation unit performs weighted addition of said first degree of suitability and said second degree of suitability to calculate said suitability evaluation value. said computer,
Further functions as a segmentation unit that acquires a synopsis, segments it into the plurality of synopsis segments, and supplies the synopsis to the first acquisition unit.
The segmented part is
dividing the obtained synopsis by the number of characters in the number of the sections, or segmenting the synopsis at semantic breaks detected by morphological analysis to generate the synopsis segments from the synopsis;
The second acquisition unit according to any one of claims 1 to 3, wherein the sentiment score set for each synopsis segment is the sentiment score set for the section corresponding to the synopsis segment. program. said computer,
The text selected by the selection unit for the synopsis segment corresponding to each section is supplied to the segmentation section as the synopsis for the section, and the segmentation section, the first acquisition section, the second acquisition section, and At least one of causing the selection unit to select text for each synopsis segment obtained by segmenting the synopsis, and supplying each text selected by the selection unit to the generation unit to generate the script. 5. The program according to claim 4 , further functioning as a supply unit for performing said computer,
6. The method further functions as a repeat control unit that causes the first acquisition unit, the segmentation unit, the second acquisition unit, the selection unit, and the supply unit to repeatedly perform processing until the amount of the script exceeds a reference amount. program described in . said computer,
a text generation unit that generates a plurality of random texts in natural language from a plurality of script data;
7. The program according to any one of claims 1 to 6 , further functioning as a storage unit that stores a plurality of texts generated by said text generation unit as said text group. 8. The program according to any one of claims 1 to 7 , wherein each text in said group of texts includes a headline, dialogue and topic. a first acquiring unit for acquiring a plurality of synopsis segments obtained by segmenting a synopsis into a preset number of segments for a screenplay to be created ;
a second acquisition unit that acquires the sentiment score set for each of the sections as the sentiment score of the synopsis segment corresponding to the section ;
a selection unit for selecting, from a group of texts, text corresponding to the content of the synopsis segment and corresponding to the sentiment score of the synopsis segment for each of the plurality of synopsis segments;
a generating unit that generates the script by arranging each text selected for each synopsis segment in the order of the section corresponding to the synopsis segment ;
A device comprising a first obtaining step in which a computer obtains a plurality of synopsis segments that segment the synopsis into a preset number of intervals for the screenplay to be created ;
a second obtaining step in which the computer obtains the sentiment score set for each of the intervals as the sentiment score of the synopsis segment corresponding to the interval ;
a selection step in which , for each of the plurality of synopsis segments, a computer selects text from a group of texts corresponding to the content of the synopsis segment and corresponding to the sentiment score of the synopsis segment;
a generating step in which a computer arranges each text selected for each synopsis segment in order of the interval corresponding to that synopsis segment to generate the script;
A method for producing a screenplay, comprising:

Images