JP7493196B1 - Comparative video generation system, video generation program, and video generation method using artificial intelligence - Google Patents

Abstract

[Problem] To provide a new technology for generating videos in which the content of multiple pieces of content can be compared. [Solution] A comparative video generation system that utilizes artificial intelligence, the video generation system includes an acquisition unit and a generation unit, and stores comparison source video data including first explanatory information regarding an object explained in the video in a database, the acquisition unit acquires comparison target data including second explanatory information regarding the comparison object, and the generation unit inputs the first explanatory information and the second explanatory information into a generation artificial intelligence to generate and process comparative explanatory information, and generates a comparative video based on the comparative explanatory information and the comparison source video data. [Selected Figure] Figure 1

Description

This relates to a comparative video generation system, video generation program, and video generation method that utilizes artificial intelligence.

Technologies that allow users to edit videos by uploading video content have been known for some time. For example, Patent Literature 1 proposes an example of a technology that joins multiple video contents into a single video content.

Patent document 1 discloses identifying sources associated with video clips uploaded by users to a content hosting site or service, and if the video clip contains content such as a movie, identifying the movie or the like as the source based on a comparison of the video clip with sources in a source data store, creating a transcript (or classification data) for the video clip based on subtitle text of the video clip or speech recognition technology, identifying a set of video clips having associated content based on the classification data, stitching the set of video clips into a single integrated video according to an ordering parameter, and stitching together the video clips in such a way that the overlapping portions are included in the other video clips if there is overlapping content.

JP 2016-500218 A

However, while the invention described in Patent Document 1 can link multiple related video contents together to generate a single video content based on classification data for the uploaded video content, it cannot display the content of the video contents in a comparable manner.

In view of the above, the present invention aims to provide a new technology for generating videos in which the content of multiple pieces of content can be compared.

In order to solve the above problems, the present invention provides a comparative video generation system that utilizes artificial intelligence, the video generation system including an acquisition unit and a generation unit, and stores comparison source video data including first explanatory information about an object described in the video in a database, the acquisition unit acquires comparison target data including second explanatory information about the comparison object, the generation unit inputs the first explanatory information and the second explanatory information into a generation artificial intelligence, processes and generates comparative explanatory information, and generates a comparative video based on the comparative explanatory information and the comparison source video data.

With this configuration, a comparison video can be generated by inputting the first explanatory information of the comparison source video stored in the database and the second explanatory information of the acquired comparison target data into the generating artificial intelligence to generate comparison explanatory information. By using the generating artificial intelligence, a comparison video can be generated in which notation is appropriately integrated even if there are differences in notation between the first explanatory information and the second explanatory information.

In a preferred embodiment of the present invention, the generation unit obtains data corresponding to the comparative explanation information from an external database and generates the comparative video that displays the data.

With this configuration, the user can obtain more detailed information by retrieving related information from the database, in addition to the information obtained by inputting it into the generative artificial intelligence.

In a preferred embodiment of the present invention, the comparison video generation system utilizing artificial intelligence further includes a display processing unit and an update unit, the display processing unit publicly displays the comparison source video, and the update unit updates the comparison source video or the comparison video based on an evaluation of the comparison video from an administrator who manages the comparison source video data.

With this configuration, the comparison source video or the comparison video is updated according to the evaluation from the administrator of the comparison source video, and inappropriate videos are prevented from being generated in the comparison videos.

In a preferred embodiment of the present invention, the generation unit receives a specification of the timing of the comparison source video data, inputs the first explanatory information to be displayed at the specified timing and the acquired second explanatory information to the generation artificial intelligence, processes and generates the comparison explanatory information, and generates the comparison video in which the first explanatory information to be displayed at the specified timing and the comparison explanatory information are associated.

By configuring in this way, the comparison explanation information can be generated by specifying the timing of the comparison source video data, and the first explanation information of the specified timing can be associated with the comparison explanation information.

In a preferred embodiment of the present invention, the artificial intelligence-based comparative video generation system further includes a determination unit that determines the timing of the comparison source video at which the first explanatory information corresponding to the generated comparative explanatory information is displayed, and the generation unit generates the comparative video that displays the comparative explanatory information and the first explanatory information at that timing.

By configuring it in this way, the timing of the comparison source video at which the comparison explanation information is to be displayed can be identified without the user having to specify it.

In a preferred embodiment of the present invention, the generation unit generates the comparison video in which the comparison explanation information is superimposed on the first explanation information.

By configuring it in this way, the first explanatory information and the comparative explanatory information can be displayed overlapping on the screen, allowing the user to check the first explanatory information and the comparative explanatory information with minimal eye movement.

In a preferred embodiment of the present invention, the generation unit generates a comparison video that switchably displays the content of the first explanatory information and the content based on the comparative explanatory information.

This configuration allows the display of the contents of the first explanatory information and the contents based on the comparative explanatory information to be switched on and off. This makes it possible to accommodate both cases where you want to see only the first explanatory information and where you want to see the contents based on the comparative explanatory information.

In a preferred embodiment of the present invention, the generation unit generates the comparison video in which the comparison explanation information is arranged by dividing the screen according to the number of comparison target data obtained.

With this configuration, the comparison video screen can be divided according to the number of comparison target data obtained, and the contents and indicators of the comparison source video can be arranged. This allows the contents of the videos to be arranged in a way that is easier for the user to see, making it easier for the user to compare the comparison targets.

In a preferred embodiment of the present invention, the generation unit generates a graph based on the comparative explanatory information when the first explanatory information and the second explanatory information include a numerical data format.

By configuring in this way, if the first explanatory information and the second explanatory information in the comparison video include numerical data formats, a graph based on the comparative explanatory information is displayed, allowing the user to more easily check the comparative explanation.

The present invention provides a novel technology for generating comparison videos that allow the content of multiple pieces of content to be compared.

1 is a block diagram showing a configuration of a system according to an embodiment of the present invention. FIG. 2 is a hardware configuration diagram of a system according to an embodiment of the present invention. FIG. 2 is a functional block diagram according to an embodiment of the present invention. 13 is an example of a process flowchart for generating a comparison video according to an embodiment of the present invention. 13 is an example of a comparison video generated in one embodiment of the present invention. 13 is an example of a comparison video generated in one embodiment of the present invention. 13 is an example of a comparison video generated in one embodiment of the present invention. 13 is a diagram showing an example of a comparison video to be played back in one embodiment of the present invention. FIG. 11 is a diagram relating to generation of a comparison video in another embodiment of the present invention.

The following description will be given in more detail with reference to the attached screenshots. The drawings show preferred embodiments. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein.

For example, in this embodiment, the configuration, operation, etc. of a comparative video generation system utilizing artificial intelligence will be described, but similar effects can be achieved by the executed method, device, computer program, etc. Furthermore, the program may be stored on a recording medium. By using this recording medium, for example, the program can be installed on a computer, thereby configuring a comparative video generation device and a comparative video generation system utilizing artificial intelligence. Here, the recording medium storing the program may be a non-transitory recording medium such as a CD-ROM.

The present invention relates to a system for generating a comparison video in which the data of a comparison source video viewed by a user and the content of acquired comparison target data can be compared. In the present invention, the comparison source video data is a video explaining a comparison target, and the comparison target data is a file related to the comparison target. The comparison target in this embodiment is a thing or thing to be compared, and includes information about a company (e.g., management policy, product description, security policy, sales explanation, training materials, etc.) and information about an individual's product. Specifically, when the comparison target is a company's management policy, the management policies of two different companies are compared, and when the comparison target is a company's product description, explanations about two different products of the same company, or explanations about the same two products of different companies, etc. are compared. In addition, the file related to the comparison target includes, for example, contents to be compared with the comparison target. In addition, in the present invention, the comparison source video is a video file such as HTML5, but may be in a file format such as MP4. In addition, the comparison target data is text data and/or image data, but may be a video file such as HTML5 or MP4.

In this embodiment, an instruction to generate a comparison video is received from a user who has viewed the comparison source video, and the comparison source video data and the comparison target data are input to the generation artificial intelligence, thereby generating information (language data and/or image data) regarding the comparison result of the contents of the comparison source video data and the comparison target data. Then, the comparison video is generated based on the information regarding the comparison source video and the comparison result. More specifically, the comparison video is generated by adding the generated information regarding the comparison result to the comparison source video.

The generative artificial intelligence here is a model that learns by itself based on a huge amount of data, and generates output data based on the input of text data, image data, or audio data related to the data to be output. For example, text or images are generated as output data. Here, the text that is output may include program code such as a script, or text data written in a specific format such as CSV. The generative artificial intelligence in this embodiment receives input of source video data and target data, and generates information regarding the comparison results of their contents as output data.

In addition, in this embodiment, the generative artificial intelligence is a large-scale language model such as a Generative Pretrained Transformer (GPT), but there are no limitations on its scale or model structure, and it may be other language models that combine one or more models, such as a Convolutional Neural Network (CNN); a Recurrent Neural Network (RNN) such as a Long short-term memory (LSTM) or a Gated recurrent unit (GRU); a Transformer model, etc.

1. System configuration
Fig. 1 is a block diagram showing the configuration of a system according to an embodiment. As shown in Fig. 1, a comparative video generation system 0 using artificial intelligence includes a comparative video generation device 1, a user terminal device 3, and an administrator terminal device 4, and is configured to be capable of communicating via a communication network NW. In this embodiment, the communication network NW is an IP (Internet Protocol) network, but there is no restriction on the type of communication protocol, and there is also no restriction on the type and scale of the network.

A general-purpose server computer or personal computer can be used as the comparative video generating device 1. A smartphone, tablet terminal, personal computer, wearable device, etc. can be used as the user terminal device 3 and the administrator terminal device 4. The comparative video generating device 1 may be configured with multiple computers that are capable of sending and receiving information via the communication network NW or another network.

<1.1. Hardware configuration>
2 is a hardware configuration diagram of a comparison video generation system 0 utilizing artificial intelligence. As shown in FIG. 2( a), a server 10 (comparison video generation device 1) includes a processing unit 101, a storage unit 102, and a communication unit 103.

The processing unit 101 has one or more processors, such as a CPU capable of executing an instruction set, and controls the overall operation and processing of the comparison video production device 1 by executing the comparison video production program according to the present invention, an OS, and other applications.
The storage unit 102 has a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS and the comparative video generating program according to the present invention.
The communication unit 103 has a communication interface device with the communication network NW, and controls communication with the network NW to input and output information.

As shown in FIG. 2(b), the terminal device 9 (user terminal device 3 and administrator terminal device 4) includes a processing unit 91, a memory unit 92, a communication unit 93, an input unit 94, and an output unit 95.

The processing unit 91 has one or more processors, such as a CPU, capable of executing an instruction set, and controls the overall operation and processing of the terminal device 9 by executing an OS and other applications.
The storage unit 92 has a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS and the like.
The communication unit 93 has a communication interface device for connecting to a network, and controls communication with the network NW to input and output information.
The input unit 94 has an input device capable of input processing, such as a keyboard or a touch panel.
The output unit 95 has a display device capable of display processing, such as a display.

<1.2. System functional configuration>
Fig. 3 is a functional block diagram of the comparison video generating device 1. As shown in Fig. 3, the comparison video generating device 1 includes an acquisition unit 11, a generation unit 12, a registration unit 13, an update unit 14, a display processing unit 15, and a database 2. This is information processing by software (stored in a storage unit 102) specifically realized by hardware (processing unit 101).

The system configuration in this embodiment is a so-called server-client type in which the client receives the processing results performed by the comparison video generation device 1 (server) in response to a request from the user terminal device 3 (client). On the other hand, it may be a so-called standalone type in which the comparison video generation program is started in the client terminal. In this case, the user terminal device 3 may be equipped with some or all of the functional configurations (units) equipped in the comparison video generation device 1. For example, the user terminal device 3 may be equipped with an acquisition unit 11, a generation unit 12, a registration unit 13, an update unit 14, a display processing unit 15, etc.

<1.3. Database 2>
The database 2 stores the comparison source video data and the comparison video information. All of this information may be stored in the storage unit 102, or some or all of it may be stored in the storage unit 92.

<1.3.1. Source video data for comparison>
The comparison source video data is information about the comparison source video that explains the comparison target. The comparison source video data includes a comparison source video ID that uniquely identifies the comparison source video, a comparison source video name, comparison source video content information, and a video manager. Here, the comparison source video content information is information about the video and audio of the comparison source video, and includes first explanatory information related to the comparison target explained in the comparison source video, multiple combinations of timing for displaying the first explanatory information, and information specifying the display mode of the first explanatory information (coordinates, height, width, transparency, etc.). The first explanatory information in this embodiment includes content (image, audio, and/or text), content attributes (features, functions, price, etc.), and attribute type (data type).

In this embodiment, still images are used as the video, and a slideshow-style video (video in which still images are switched sequentially) in which audio based on the still images is played is used, and the slide number of the video or the playback time (the elapsed time from the time playback started) is used as the timing. The comparison source video may also be a video in a format in which continuously changing still images are switched at high speed to display the video (a general video file in an existing video format, etc.), in which case the playback time of the video is associated with the timing.

<1.3.2. Comparison video information>
The comparison video information is information about the generated comparison video. The comparison video information includes a comparison video ID that uniquely identifies the comparison video, a comparison source video ID, comparison video content information, and a creation date, and is stored in association with the user who created it. The content information of the comparison video is information obtained by adding the comparison explanation information generated by the generation unit 12 to the comparison source video content information. Here, the comparison explanation information is information about the comparison result of the first explanation information and the second explanation information, and includes data of the comparison result itself (language data (text data and audio data), image data, and graph display data) and data for displaying the comparison result in the comparison video (mapping data). The comparison result may be only data (language data and/or image data) based on the content of the second explanation information corresponding to the first explanation information, or data (language data and/or image data) based on both the content of the first explanation information and the content of the second explanation information that are related to each other.

<1.4. Acquisition unit 11>
The acquisition unit 11 acquires comparison target data from a user or a database. The acquisition unit 11 acquires comparison target data including second explanation information related to an object explained in the video. Here, the second explanation information is , content (image, audio, and/or text), content attributes (features, functions, prices, etc.), and attribute types (data types). The database may be the database 2, and the user terminal device 3 Alternatively, it may be a database in the manager terminal device 4, or an external database connected so as to be communicable via wired or wireless communication.

<1.5. Generation unit 12>
The generation unit 12 generates a comparison video by using the first explanatory information, the second explanatory information, and the generation artificial intelligence. The generation unit 12 inputs the first explanatory information and the second explanatory information to the generation artificial intelligence. Then, a comparison video is generated based on the comparison source video data and the comparison explanation information.

The generation process in this embodiment involves sending a comparative explanation request including at least the first and second explanatory information to a generation artificial intelligence external to this system, and acquiring the generated comparative explanatory information. Alternatively, this system may have a generation artificial intelligence, and the generation artificial intelligence may generate the comparative explanatory information. Here, when the generation artificial intelligence is an LLM, a comparative explanation request including a prompt in addition to the first and second explanatory information is sent to the generation artificial intelligence.

In addition, in this embodiment, the generation unit 12 generates image data as comparative explanatory information by inputting the first explanatory information and the second explanatory information to the generation artificial intelligence, and generates a comparative video based on the generated images. On the other hand, the generation unit 12 may generate text data for displaying image data as comparative explanatory information, and generate a comparative video using image data based on the generated text data. The generation unit 12 may also obtain data corresponding to the comparative explanatory information from a database, and generate a comparative video displaying the data. Specifically, the generation unit 12 generates text data as comparative explanatory information, obtains image data corresponding to the text data from a database, and generates a comparative video displaying a combination of the text data and the image data. The database may be the database 2 of this system, or an external database that can be communicated with.

<1.6. Registration unit 13>
The registration unit 13 stores the comparison video in the database 2. The registration unit 13 stores comparison video information in association with a user who instructed the generation of the comparison video, in accordance with an evaluation of the generated comparison video received from the user.

<1.7. Update section 14>
The update unit 14 updates the comparison source video or the comparison video. The update unit 14 updates the comparison source video or the comparison video in accordance with an evaluation from an administrator associated with the comparison source video data for the generated comparison video. do.

<1.8. Display Processing Unit 15>
The display processing unit 15 performs display processing on the comparison source video and the comparison video, and causes the display processing results to be displayed on the user terminal device 3 or the administrator terminal device 4. Specific displays will be described later.

In this embodiment, the display process refers to a process in which the display processing unit 15 executes a process of generating information required for display, and transmits the generated information to the terminal device 9, thereby causing the terminal device 9 to display the generated information. On the other hand, when the display processing unit 15 is provided in the terminal device 9 (in the case of a stand-alone type), the display process may be a process in which the display processing unit 15 executes a process of generating necessary information, and transmits the generated information to the output unit 95 of the terminal device 9, thereby causing the output unit 95 to display the generated information.

A comparative video generating method using the comparative video generating system 0 utilizing artificial intelligence will be described with reference to Figures 4 to 9. Figure 4 is a flowchart showing the process in which the comparative video generating system 0 utilizing artificial intelligence receives an instruction to generate a comparative video from a user who has viewed a comparison source video, generates a comparative video by acquiring comparison destination data, and displays the generated comparative video to a third party.

<2.1. Displaying the source video for comparison>
First, in step S1 (hereinafter, "step SX" will be simply written as "SX"), the display processing unit 15 accepts a designation of a comparison source video (slide video) that the user wants to view from the user, and displays the comparison source video. In this embodiment, the display processing unit 15 displays the slide video based on the text data and/or image data of the slide.

<2.2. Timing designation acceptance>
In S2, the generating unit 12 receives an instruction to generate a comparison video from the user who has viewed the slide video, and receives a designation of the timing to display the comparison explanation information. In this embodiment, the generating unit 12 receives an instruction to generate a comparison video from the user for the displayed slide video, and receives the slide number of the displayed slide as the timing. Note that the timing may be received so that any slide number can be designated.

<2.3. Acceptance of designation of comparison mode>
In S3, the generating unit 12 receives a designation of a comparison mode indicating a display mode of the comparison explanation information. In this embodiment, the generating unit 12 receives the designation of a comparison mode including an "explanation of the difference in the comparison result", a "graph of the difference in the comparison result", and an "answer to a question item".

<2.4. Acquiring comparison data>
In S4, the acquisition unit 11 acquires the comparison target data. In this embodiment, the acquisition unit 11 acquires the comparison target data including second explanatory information different from the first explanatory information of the comparison source video from the user who viewed the slide video in S1. Note that in this embodiment, the comparison source video and the comparison target data are different data, but the comparison source video and the comparison target data may be the same data. In other words, the first explanatory information and the second explanatory information may be included in the comparison source video.

<2.5. Comparative explanation information generation process>
In S5, the generating unit 12 uses the generating artificial intelligence to generate comparative explanatory information. In this embodiment, the generating unit 12 inputs the second explanatory information and the first explanatory information included in the slide of the slide number specified in S2 to the generating artificial intelligence, so that the content attribute of the first explanatory information obtained from the text data and/or image data for displaying the slide video played in S1 is compared with the content attribute of the second explanatory information obtained from the text data and/or image data of the comparison target data acquired in S3, and obtains comparative explanatory information for explaining the difference between the first explanatory information and the second explanatory information as a comparison result. Specifically, the generating unit 12 generates mapping data indicating the correspondence between the contents of the first explanatory information and the contents of the second explanatory information, and language data and/or image data related to the explanation of the difference between the first explanatory information and the second explanatory information, as the comparative explanatory information.

<2.6. Creating comparison videos>
In S6, the generation unit 12 generates a comparison video based on the comparison explanatory information and the first explanatory information of the comparison source video data. In this embodiment, the generation unit 12 generates a comparison video in which the first explanatory information to be displayed on the slide having the slide number specified in S2 and the language data and/or image data of the comparison explanatory information generated in S5 are associated with each other based on the mapping data generated in S5.

<2.6.1. Example of comparison video: Comparative explanation using text>
FIG. 5 shows an example of a series of screens displayed until a comparison video is generated when "explanation of differences in comparison results" is specified as a comparison mode. On a slide (FIG. 5(a)) with a certain slide number displayed on the user terminal device 3 of a user who is viewing a slide video, an input of "compare" is accepted as an instruction to generate a comparison video from the user, and a slide number is accepted. Then, a screen (FIG. 5(b)) for accepting the specification of a comparison mode is displayed superimposed on the slide video. Then, when a specification of "explanation of differences in comparison results" (in the illustrated example, "compare and ask for an explanation of the differences") is accepted as a comparison mode from the user, a screen (FIG. 5(c)) for acquiring comparison target data is displayed. Then, by accepting an upload of comparison target data from the user, a comparison video (FIG. 5(d)) is generated and displayed on the user terminal device 3.

In this embodiment, when "description of the difference in the comparison result" is specified as the comparison mode, text data (character strings such as functional features, price, etc.) explaining the difference between the first explanatory information and the second explanatory information generated by the generation unit 12 as comparison explanatory information is displayed superimposed on the first explanatory information displayed on the slide of the specified slide number, and a comparison video is generated in which audio data corresponding to the text data is played. In the illustrated example, a description (text data) of the content of the second explanatory information corresponding to the content of the first explanatory information is generated as text data explaining the difference between the first explanatory information and the second explanatory information, and the description of the second explanatory information is displayed superimposed on the first explanatory information of the slide video.

In this embodiment, the content of the first explanatory information and the content of the second explanatory information are arranged so as to be adjacent to each other, but the text strings of the first explanatory information and the second explanatory information may be arranged so as to overlap in part. In addition to this, a comparison video may be generated in which the content of the first explanatory information and the content of the second explanatory information are displayed on the same screen. Specifically, a comparison video may be generated that includes a slide in which the content attributes (functions, features, price, etc.) of the first explanatory information are arranged adjacent to each of the content attributes of the second explanatory information. In another specific example, a comparison video may be generated in which a slide displaying the content attributes of the first explanatory information and the content attributes of the second explanatory information is divided vertically or horizontally, and the content attributes of the first explanatory information are arranged on one side and the content attributes of the second explanatory information are arranged on the other side.

<2.6.2. Example of a comparison video: Comparative explanation using graphs>
Fig. 6 shows an example of a series of screens displayed until a comparison video is generated when "Graph of the difference in the comparison result" is specified as the comparison mode. Fig. 6(a) and Fig. 6(c) are omitted because they are similar to the operation explanation in Fig. 5(a) and Fig. 5(c). When the user specifies "Graph of the difference in the comparison result" as the comparison mode on the screen (Fig. 6(b)) for accepting the specification of the comparison mode (in the illustrated example, "Perform a comparison and display the difference in a graph") and the comparison target data is acquired, a comparison video (Fig. 6(d)) is generated and displayed on the user terminal device 3.

In this embodiment, when "Graph of the difference in the comparison result" is specified as the comparison mode, a graph based on the comparison explanatory information is generated. When the first explanatory information and the second explanatory information include a numerical data format, the generation unit 12 generates graph display data as the comparison explanatory information, and generates a graph based on the graph display data. Specifically, the generation unit 12 generates a graph arranged for each content attribute of the first explanatory information and the second explanatory information (in the illustrated example, sales, number of contracts, operating profit, and ordinary profit) based on the graph display data. Then, the generation unit 12 generates a comparison video including a slide displaying the graph as a new slide. Note that the generation unit 12 may generate image data of the graph as the comparison explanatory information.

<2.6.3. Example of a comparison video: Answer to a question>
Fig. 7 shows an example of a series of screens displayed until a comparison video is generated when "answer to question item" is specified as the comparison mode. Fig. 7(a) and Fig. 7(c) are omitted because they are similar to the operation explanation in Fig. 5(a) and Fig. 5(c). When the user specifies "answer to question item" (in the illustrated example, "listen to answer to question item") as the comparison mode on a screen for accepting the specification of a comparison mode (Fig. 7(b)), and comparison target data including the question is acquired, a comparison video (Fig. 7(d)) is generated and displayed on the user terminal device 3.

In this embodiment, when "answers to questions" is specified as the comparison mode, comparative explanatory information including answers to the content (questions) of the second explanatory information is generated. Specifically, the generation unit 12 inputs the second explanatory information including the questions and the first explanatory information to the generating artificial intelligence, processes and generates comparative explanatory information that associates the questions of the second explanatory information with answers (language data and/or graph display data) to the questions based on the content of the first explanatory information, and generates a comparison video that displays new slides based on the comparative explanatory information.

The above is an example of a display screen of the generated comparison video, but in this embodiment, the content including the first explanatory information and the content based on the comparative explanatory information associated with the first explanatory information are displayed in a switchable manner. Specifically, the generation unit 12 generates a comparison video in which the content based on the comparative explanatory information displayed superimposed on the content of the first explanatory information can be switched on and off.

In addition, the generation unit 12 generates a comparison video in which slides can be switched by transitioning between slides of the content of the first explanatory information and slides of the content based on the comparative explanatory information associated with the content of the first explanatory information.

The generation unit 12 may also generate a comparison video that displays one avatar that explains the comparison explanatory information, or two or more avatars that interactively explain the comparison video. When the comparison source video includes an avatar, the generation unit 12 may generate the comparison video based on the avatar displayed in the comparison source video. Specifically, the generation unit 12 acquires master information of the avatar linked to the comparison source video, or an element of the avatar displayed in the comparison source video, and generates a comparison video that displays an avatar based on the master information or element. At this time, when the generation unit 12 displays a slide of the content of the first explanatory information and a slide of the content based on the comparison explanatory information associated with the content in a transitional manner, the generation unit 12 may display an avatar in the content based on the comparison explanatory information based on the avatar displayed in the comparison source video. In addition, when the comparison video is divided and displayed as described later, one avatar may be displayed in one content, or the number of avatars may be changed depending on the number of divisions of the slide. In addition, when generating a comparison video that displays an avatar, the generation unit 12 may generate a script for generating a synthetic voice emitted by the avatar as the comparison explanatory information, as the comparison explanatory information. Furthermore, when two or more avatars are displayed, the script and comparison explanation information of the comparison source video may be associated with each avatar, and a comparison video may be generated in which an avatar that speaks based on the associated script is displayed. Note that when the comparison source video does not include an avatar, the generation unit 12 may generate a comparison video in which an avatar is displayed.

<2.7. Editing Reception>
Then, in S7, the comparison video is edited by the user. Specifically, the text data of the comparison explanation information is displayed in an editable manner, and an editing unit (not shown) accepts input from the user regarding editing of the text data. The generation unit 12 then updates the comparison video based on the edited text data.

<2.8. Registration of comparison videos>
In S8, the registration unit 13 stores the comparison video in the database 2 depending on whether or not a GOOD rating has been received for the comparison video from a user who has viewed the comparison video. If a GOOD rating has been received for the comparison video (YES in S8), the registration unit 13 stores the comparison video in S9. On the other hand, if a GOOD rating has not been received for the comparison video (NO in S8), the process ends.

<2.9. Public display of comparison videos>
In S10, the update unit 14 updates the comparison source video depending on whether or not a GOOD rating has been received from an administrator linked to the comparison source video on which the comparison video viewed by the user is based. If a GOOD rating has been received for the comparison video from the administrator (YES in S10), the update unit 14 updates the comparison source video in S11 and makes the comparison video generated in S6 displayable to a third party. On the other hand, if a GOOD rating has not been received for the comparison video from the administrator (NO in S10), the process ends.

As described above, the processes of S1 to S11 are executed to generate a comparison video, thereby updating the comparison source video and making it possible to view a comparison video based on the comparison source video. FIG. 8 is an example of a display when the generated comparison video is played. In the illustrated example, an interactive button (in the illustrated example, "display graph") for displaying a new slide based on the comparison explanatory information corresponding to the first explanatory information displayed on the slide of slide number B of the comparison video being played is added. Then, by pressing the interactive button, the comparison video is displayed so as to be able to transition from the slide displaying the contents (contents) of the first explanatory information to the slide displaying the contents based on the comparison explanatory information. In this embodiment, an interactive button for displaying a slide displaying the first explanatory information corresponding to the comparison explanatory information is displayed on the slide displaying the contents based on the comparison explanatory information, and the slide displaying the contents of the first explanatory information is displayed by accepting the interactive button (in the illustrated example, "end viewing") from the user. Note that when the playback time of the slide displaying the first explanatory information corresponding to the comparison explanatory information ends, the slide displaying the comparison explanatory information is no longer displayed, and the next slide of the comparison video may be displayed. Furthermore, if the second explanatory information is a slide video, a slide displaying the second explanatory information corresponding to the first explanatory information is displayed as comparative explanatory information in a switchable manner from the slide of the first explanatory information, and the video of that slide is played. Then, when the video of the slide displaying the second explanatory information ends, the next slide of the comparative video is displayed.

In addition, even if an instruction to generate a further comparison video is received from a user who has viewed the comparison video, S2 to S11 are executed, the comparison video viewed by the user is updated, and a comparison video based on the comparison video becomes viewable. Specifically, the acquisition unit 11 further acquires comparison destination data from the user who has viewed the comparison video. Then, the generation unit 12 inputs the comparison source video data, at least one second explanatory information of one or more comparison destination data acquired before the acquired comparison destination data, and the second explanatory information of the further acquired comparison destination data to the generation artificial intelligence, and generates the comparison explanatory information. Then, the update unit 14 updates the comparison video generated based on the comparison explanatory information and the comparison video, according to the evaluation from the user who instructed to generate the comparison video and the administrator of the comparison source video on which the comparison video was based. That is, when a certain user uploads new comparison destination data for a comparison video viewed by the user, a comparison video is generated that further displays the comparison explanatory information based on the comparison destination data uploaded by the user in addition to the one or more comparison explanatory information displayed in the comparison video.

In a preferred embodiment of the present invention, a comparison video is generated in which the screen is divided according to the number of comparison target data acquired and the comparison explanatory information is arranged. Specifically, for example, when three comparison target data are acquired, the slide of the comparison video is divided into four parts vertically and horizontally, and a comparison video is generated in which the content related to the first explanatory information contained in the comparison explanatory information and the content related to the three pieces of second explanatory information are arranged and displayed in each divided block so as not to overlap. In an even more preferred embodiment, when a predetermined number or more of comparison target data are acquired, the acquired comparison target data may be prevented from being used in the generation process of the comparison explanatory information in order of oldest to newest.

In a preferred embodiment, when further comparison data is received for a comparison video that has already been released, when the acquisition unit 11 acquires second explanatory information from the database, the acquired information is different each time. Specifically, the acquisition unit 11 acquires second explanatory information including better or similar content compared to the first explanatory information and second explanatory information of the comparison video that has already been released, from the database by inputting the second explanatory information and the first explanatory information uploaded by the user and stored in the database to the generation artificial intelligence. Then, the generation unit 12 generates a comparison video by inputting the acquired second explanatory information and the first explanatory information to the generation artificial intelligence. For example, FIG. 9 is a diagram showing that a comparison video is generated by acquiring different information. In the left diagram of the illustrated example, second explanatory information including product B as content is acquired from the database, and a comparison video is generated and released. Meanwhile, in the right diagram of the illustrated example, second explanatory information that includes product C as content is obtained from the database as second explanatory information that includes content that is better or more similar to the first explanatory information that includes product A as content and the second explanatory information that includes product B as content, and a comparison video is generated.

Here, the process of acquiring second explanatory information that includes better content compared to the content of the first explanatory information is a process of acquiring second explanatory information that includes better content, in which a numerical value is assigned to each content attribute, and the numerical value obtained by substituting each numerical value into a multiple regression model having a weighting coefficient for each content attribute is greater than the numerical value of the first explanatory information and other second explanatory information. Also, the process of acquiring second explanatory information that includes similar content compared to the content of the first explanatory information is a process of acquiring second explanatory information that includes more similar content, in which the difference between the numerical value obtained by substituting each numerical value assigned to each content attribute of the second explanatory information into a multiple regression model having a weighting coefficient for each content attribute and the numerical value obtained from the first explanatory information is smaller than the difference between the other information.

In addition, in this embodiment, instead of accepting a designation of a comparison mode, an instruction to generate a summary and/or details of the comparison source video can be accepted to generate a comparison video that displays a summary and/or details of the comparison source video. Specifically, by accepting a designation of "summary or detailed analysis of the comparison source video" (in the example of FIG. 5(b) , "listen to a summary explanation of the question items" or "listen to a detailed explanation of the question items"), the generation unit 12 inputs the text and/or image of the first explanatory information to be displayed on the slide with the designated slide number to the generating artificial intelligence. The generation unit 12 then generates a comparison video that displays a slide that displays the obtained summary and/or details of the first explanatory information in a transitionable manner from the slide with the designated slide number.

In addition, in this embodiment, by receiving an instruction to fill in the blanks on a checklist, if the contents corresponding to the checklist items received as the comparison target data are included in the contents of the comparison source video, a comparison video can be generated in which the checklist is filled in and the contents corresponding to the filled-in items are displayed. Specifically, by receiving a designation for "filling in the blanks on the checklist" (in the example of FIG. 5(b) , "fill in the blanks and listen to an explanation of the results"), the generation unit 12 inputs the first explanatory information and the second explanatory information including the check items to the generation artificial intelligence, and generates text data corresponding to the checklist items based on the content of the first explanatory information as comparison explanatory information. The generation unit 12 then generates a comparison video in which the check items of the second explanatory information and the generated text data are used as new slides.

In this embodiment, a timing specification is accepted from a user, and a comparison video is generated in which the first explanatory information to be displayed at the specified timing corresponds to the generated comparative explanatory information. Meanwhile, a determination unit (not shown) may obtain a timing from the generating artificial intelligence and determine the timing as the timing for displaying the comparative explanatory information. Specifically, the determination unit obtains from the generating artificial intelligence the timing of the comparison source video at which the first explanatory information corresponding to the generated comparative explanatory information is displayed, and determines the timing as the timing for displaying the comparative explanatory information.

In this embodiment, a slide video is described as the comparison source video, but if the comparison source video is a video file such as MP4, in S2, the timing (frame) of the comparison source video is specified, and in S5, the first explanatory information and the second explanatory information are identified for the video file using a technology such as OCR (Optical Character Recognition) to generate comparative explanatory information, and in S6, a comparison video is generated in which the first explanatory information displayed in the identified frame and the comparative explanatory information generated in S5 correspond to each other.

In this embodiment, the acquisition unit 11 acquires a file related to the comparison target as the comparison target data, but may also acquire the URL (Uniform Resource Locator) of the site related to the comparison target. In this case, the generation unit 12 inputs the content of the site and the first explanation information to the generation artificial intelligence and processes to generate the comparison explanation information. The acquisition unit 11 may also accept the specification of another comparison source video as the comparison target data and acquire the other comparison source video.

0: Comparative video generation system using artificial intelligence 1: Comparative video generation device 3: User terminal device 4: Administrator terminal device 10: Server 101: Control unit 102: Storage unit 103: Communication unit 9: Terminal device 91: Control unit 92: Storage unit 93: Communication unit 94: Input unit 95: Output unit 11: Acquisition unit 12: Generation unit 13: Registration unit 14: Update unit 15: Display processing unit NW: Communication network

Claims (11)

A comparative video generation system using artificial intelligence,
The comparative video generation system utilizing artificial intelligence includes an acquisition unit and a generation unit,
storing a comparison source video data including first explanatory information related to an object explained in the video in a database;
The acquisition unit acquires comparison target data including second explanatory information related to a comparison target,
A comparative video generation system utilizing artificial intelligence, wherein the generation unit inputs the first explanatory information and the second explanatory information into a generation artificial intelligence to generate and process comparative explanatory information, and generates a comparative video based on the comparative explanatory information and the comparison source video data. The generation unit acquires data corresponding to the comparative explanation information from an external database,
The comparative video generation system using artificial intelligence according to claim 1 , which generates the comparative video displaying the data. The comparative video generation system using artificial intelligence further includes a display processing unit and an update unit,
The display processing unit publicly displays the comparison source video,
The comparison video generation system using artificial intelligence as described in claim 1 , wherein the update unit updates the comparison source video or the comparison video based on an evaluation of the comparison video from an administrator who manages the comparison source video data. the generating unit receives a designation of a timing of the comparison source video data, inputs the first explanatory information to be displayed at the designated timing and the acquired second explanatory information to the generating artificial intelligence, and generates the comparison explanatory information;
The comparative video generation system using artificial intelligence according to claim 1 , which generates the comparative video in which the first explanatory information to be displayed at the specified timing and the comparative explanatory information are associated with each other. The comparative video generation system using artificial intelligence further includes a specifying unit,
The identification unit identifies a timing of the comparison source video to display the first explanatory information corresponding to the generated comparison explanatory information,
The comparative video generating system utilizing artificial intelligence according to claim 1 , wherein the generating unit generates the comparative video in which the comparative explanatory information and the first explanatory information are displayed at the timing. The comparative video generating system utilizing artificial intelligence according to claim 3 or claim 4, wherein the generating unit generates the comparative video by superimposing the comparative explanatory information on the first explanatory information. The comparative video generation system utilizing artificial intelligence according to claim 3 or claim 4, wherein the generation unit generates the comparative video in a switchable manner by displaying the content of the first explanatory information and the content based on the comparative explanatory information. The comparative video generation system using artificial intelligence according to claim 3 , wherein the generation unit generates the comparative video in which the comparison explanation information is arranged by dividing a screen according to the number of the comparison target data acquired. The comparative video generation system using artificial intelligence according to claim 1 , wherein the generation unit generates a graph based on the comparative explanatory information when the first explanatory information and the second explanatory information include a numerical data format. A comparative video generation program using artificial intelligence,
The comparative video generation program using artificial intelligence causes a computer to function as an acquisition unit, a generation unit, and a database,
The database stores comparison source video data including first explanation information related to an object explained in the video;
The acquisition unit acquires comparison target data including second explanatory information related to a comparison target,
A comparison video generation program utilizing artificial intelligence, wherein the generation unit inputs the first explanatory information and the second explanatory information into a generation artificial intelligence, processes and generates comparison explanatory information, and generates a comparison video based on the comparison explanatory information and the comparison source video data. A comparative video generation method using artificial intelligence,
The computer
A process of storing comparison source video data including first explanation information related to an object explained in the video in a database;
A process of acquiring comparison target data including second explanatory information related to the comparison target;
A process of inputting the first explanatory information and the second explanatory information into a generating artificial intelligence to generate comparative explanatory information, and generating a comparative video based on the comparative explanatory information and the comparison source video data;
A comparative video generation method that utilizes artificial intelligence to perform the above.