program
The program analyzes player-generated content using a trained model, providing feedback and machine learning capabilities to enhance the gaming experience by improving player creations.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2026-03-24
AI Technical Summary
Players in games like 'A Girl Who Sings Love Songs at the End of the World, YU-NO' desire analysis on the content they create, but existing systems lack the capability to provide such analysis.
A program that includes a control device functioning as setting acquisition, content generation, output, and execution means, utilizing a trained model to analyze generated content, and a distribution mechanism to share this content for evaluation and machine learning.
Enables analysis of player-generated content, allowing players to understand and improve their creations through feedback and machine learning, enhancing the gaming experience.
Smart Images

Figure 2026052341000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a program.
Background Art
[0002] As an invention related to conventional programs, for example, a game described in Non-Patent Document 1 is known. In this game, by a player selecting an option, the story progresses according to the content corresponding to the option.
Prior Art Documents
Non-Patent Documents
[0003]
Non-Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] By the way, in the field of the game described in Non-Patent Document 1, players have a desire to receive analysis on the content they create.
[0005] Therefore, an object of the present invention is to provide a program that can receive analysis on the content created by a player.
Means for Solving the Problems
[0006] A first aspect is that the program functions the control device of a first computer as setting acquisition means, content generation means, output means, and execution means, the setting data is data indicating the setting of the generated content, The setting acquisition means acquires the setting data input by the first player operating the computer, The content generation means generates generated content data indicating the generated content based on the setting data, The output means can output the generated content data to the trained model. The trained model can analyze the generated content data. The execution means executes game effects after analyzing the generated content data. When the analysis results from the trained model change, the content of the game effect executed by the execution means changes. It is a program.
[0007] The second aspect is, The aforementioned trained model uses a combination of training content and the analysis results of the training content as training data. This is the program described on the first side.
[0008] The third aspect is, The analysis results of the aforementioned teacher content include a parameter indicating the number of people who evaluated the teacher content including predetermined characteristics. This is the program described on the second side.
[0009] The fourth aspect is, The analysis results of the aforementioned teacher content include comments evaluating the teacher content. This is the program described in the second or third aspect.
[0010] The fifth aspect is, The program is further configured to function as a distribution means in the control device of the first computer. The distribution means can transmit the generated content data to the first server. The first server can distribute the generated content data to one or more second computers via a network line. The second server includes the learned model, and can obtain evaluation information indicating an evaluation of the generated content indicated by the generated content data from the one or more second computers. The learned model performs machine learning using a combination of the generated content and the evaluation of the generated content as teacher data. The program according to any one of the first aspect to the fourth aspect.
[0011] The sixth aspect is The distribution means can transmit condition data indicating conditions of one or more second players that can use the content indicated by the generated content data to the first server together with the generated content data. The program according to the fifth aspect.
[0012] The seventh aspect is The game effect is the development after analysis of the game. The program according to any one of the first aspect to the sixth aspect.
[0013] The eighth aspect is The game effect is the display of the analysis result of the generated content by the learned model. The program according to any one of the first aspect to the sixth aspect.
[0014] The ninth aspect is The game effect is that a privilege is granted to the first player for the analysis result of the generated content by the learned model. The program according to any one of the first aspect to the sixth aspect.
[0015] The tenth aspect is The generated content is a video of a play. The setting data includes information about the actors in the play, information about the lighting in the play, information about the sound effects in the play, or information about the props in the play. The program according to any one of the first aspect to the ninth aspect.
[0016] The eleventh aspect is The generated content is a video of a play in which a plurality of actors perform, A proficiency parameter indicating the proficiency level of acting is set for each of the plurality of actors, When the proficiency parameter of each of the plurality of actors changes, the acting content changes. The program according to any one of the first aspect to the tenth aspect.
[0017] The twelfth aspect is The proficiency parameter changes according to the number of times the play video is played. The program according to the eleventh aspect.
Advantages of the Invention
[0018] According to the present disclosure, the content created by the player can be analyzed.
Brief Description of the Drawings
[0019] [Figure 1] FIG. 1 is an explanatory diagram of the systems 1, 1a, and 1b. [Figure 2] FIG. 2 is a block diagram of the first player terminal 10-1. [Figure 3] FIG. 3 is a block diagram of the game server 110. [Figure 4] FIGS. 4 to 9 are diagrams showing the video displayed on the display 20 of the first player terminal 10-1. [Figure 5] FIGS. 4 to 9 are diagrams showing the video displayed on the display 20 of the first player terminal 10-1. [Figure 6]Figures 4 through 9 show the images displayed on the display 20 of the first player terminal 10-1. [Figure 7] Figures 4 through 9 show the images displayed on the display 20 of the first player terminal 10-1. [Figure 8] Figures 4 through 9 show the images displayed on the display 20 of the first player terminal 10-1. [Figure 9] Figures 4 through 9 show the images displayed on the display 20 of the first player terminal 10-1. [Figure 10] Figure 10 shows a configuration table. [Figure 11] Figure 11 is a flowchart showing the operations performed by the control device 12 of the first player terminal 10-1 and the control device 112 of the game server 110 during the generated content data generation and analysis process. [Figure 12] Figure 12 is a flowchart of the subroutine for step S1 in Figure 11. [Figure 13] Figure 13 is a flowchart showing the operations performed by the control device 12 and video server 210 of the first player terminal 10-1 during the generated content data distribution process. [Figure 14] Figure 14 is an explanatory diagram of the machine learning model training process. [Figure 15] Figure 15 is a flowchart showing the operations performed by the control device 12 and video server 210 of the first player terminal 10-1 during the generated content data distribution process. [Figure 16] Figure 16 shows the image displayed on the display 20 of the first player terminal 10-1 when in instruction mode. [Figure 17] Figure 17 is an explanatory diagram of the operation of system 1b. [Modes for carrying out the invention]
[0020] (Embodiment) System 1 according to an embodiment of this disclosure will be described with reference to the drawings.
[0021] [System 1 Overview] First, the overall configuration of System 1 will be explained with reference to the diagram. Figure 1 is an explanatory diagram of Systems 1, 1a, and 1b.
[0022] System 1, shown in Figure 1, comprises a first player terminal 10-1 (first computer), a second player terminal 10-2, a game server 110 (second server), and a video server 210 (first server). The first player terminal 10-1, the second player terminal 10-2, the game server 110, and the video server 210 can communicate with each other via a communication network. The network can be the internet, an intranet, or the like.
[0023] Each of the first player terminal 10-1 and the second player terminal 10-2 is an information processing device used by the first player and the second player, respectively. Each of the first player terminal 10-1 and the second player terminal 10-2 is, for example, a smartphone, a tablet device, a home game console, a portable game console, or a personal computer.
[0024] The game server 110 and the video server 210 are computers. The game server 110 and the video server 210 provide responses to requests from the first player terminal 10-1 and the second player terminal 10-2. The game server 110 is the server for the first player terminal 10-1 and the second player terminal 10-2 to play the game. The video server 210 is the server for distributing generated content data D1 generated by the first player terminal 10-1 to multiple computers, including the second player terminal 10-2. In this embodiment, the generated content data D1 is video data.
[0025] [Structure of the first player terminal 10-1 and the second player terminal 10-2] Next, the structures of the first player terminal 10-1 and the second player terminal 10-2 will be described with reference to the drawings. Figure 2 is a block diagram of the first player terminal 10-1.
[0026] As shown in Figure 2, the first player terminal 10-1 includes a control device 12, a storage unit 14, a network interface 16, a graphics processing unit 18, a display 20, an audio processing unit 22, a speaker 24, and an operation unit 26.
[0027] The memory unit 14 stores programs and data. The memory unit 14 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk).
[0028] The program includes, for example, the following: • Programs for the OS (Operating System) • Programs for applications that perform information processing (e.g., web browsers, or target applications described later)
[0029] The data includes, for example, the following: • Databases referenced in information processing • Data obtained by performing information processing (i.e., the results of information processing)
[0030] The control device 12 realizes the functions of the first player terminal 10-1 by executing the program stored in the memory unit 14. The control device 12 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Array)
[0031] The control device 12 includes a setting acquisition means 120, a content generation means 122, an output means 124, an execution means 126, and a distribution means 128 as functional blocks.
[0032] The network interface 16 controls communication between the first player terminal 10-1 and external devices. The external devices are the second player terminal 10-2, the game server 110, and the video server 210.
[0033] The graphics processing unit 18 displays an image on the display 20 based on the image data generated by the control device 12. The display 20 is either a liquid crystal display or an organic EL (Electro-Luminescence) display.
[0034] The audio processing unit 22 causes the speaker 24 to output sound based on the audio data generated by the control device 12.
[0035] The control unit 26 generates operation signals based on the actions of the first player and outputs operation signals to the control device 12. The control unit 26 may be, for example, a touch panel, a gamepad, a keyboard, or a mouse.
[0036] Since the second player terminal 10-2 has the same structure as the first player terminal 10-1, its explanation will be omitted.
[0037] [Structure of Game Server 110] Next, the structure of the game server 110 will be explained with reference to the diagram. Figure 3 is a block diagram of the game server 110.
[0038] The game server 110 is capable of communicating with a first player terminal 10-1 operated by a first player and a second player terminal 10-2 operated by a second player. As shown in Figure 3, the game server 110 includes a control device 112, a storage device 114, and a network interface 116.
[0039] The storage device 114 stores programs and data. The storage device 114 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (for example, flash memory or hard disk). The program includes, for example, the following: • OS (Operating System) programs • Programs for applications that perform information processing (e.g., web browsers, or target applications described later)
[0040] The data includes, for example, the following: • Databases referenced in information processing • Data obtained by performing information processing (i.e., the results of performing information processing)
[0041] The control device 112 implements the functions of the game server 110 by executing the program stored in the storage device 114. The control device 112 is, for example, at least one of the following: ·CPU(Central Processing Unit) ·GPU(Graphic Processing Unit) ·ASIC(Application Specific Integrated Circuit) ·FPGA(Field Programmable Array)
[0042] The network interface 116 controls communication between the game server 110 and external devices. The external devices are the first player terminal 10-1, the second player terminal 10-2, and the video server 210.
[0043] Furthermore, the game server 110 is equipped with a pre-trained model 150. The pre-trained model 150 is a pre-trained model generated, for example, by having a machine learning model perform machine learning according to a model training program. The pre-trained model 150 is, for example, a parameterized composite function composed of multiple functions, and is defined by a combination of multiple tunable functions and parameters. The pre-trained model may be implemented as a specific example by a model such as GPT-4 (Generative Pre-Training 4) or Gemini, which is based on Transformer technology for natural language processing. The pre-trained model may also be a large-scale language model. The pre-trained model may be trained, for example, by supervised learning, or by both supervised learning and reinforcement learning methods. The pre-trained model may be a general-purpose model or a model specialized for a specific player.
[0044] Incidentally, the first player can play the game using the first player terminal 10-1. Specifically, the game server 110 stores the game program and game data and manages the game data of the first player terminal 10-1.
[0045] The first player terminal 10-1 executes the game based on the first player's operations. To this end, the first player terminal 10-1 receives (specifically downloads and installs) the game program and game data from the game server 110 via the communication network. An account containing identification information and a password is assigned to each player and is associated with the first player terminal 10-1. The account is sent from the first player terminal 10-1 to the game server 110 upon login and is used for player authentication at the game server 110.
[0046] After player authentication, communication between the game server 110 and the first player terminal 10-1 becomes possible. After logging in, the first player terminal 10-1 receives data necessary for game progression (data related to game progress) from the game server 110, and proceeds with the online game while outputting game images and sounds to the display 20 and speaker 24 based on the first player's input.
[0047] [System 1 Operation] Next, we will explain the operation of System 1 with reference to the diagrams. First, we will give an overview of the operation of System 1. System 1 performs the operations (1) to (3) described below. (1) Generated content data generation process and analysis process (2) Processing of generated content data (3) Machine learning model training process
[0048] In the generated content data generation and analysis process of (1), the first player generates generated content data D1 representing generated content C1 by operating the first player terminal 10-1. Generated content C1 is an entertainment item that the first player and the second player can enjoy visually and / or aurally using the first player terminal 10-1 and the second player terminal 10-2, respectively. Generated content C1 can be a game, an animated video, a movie video, a drama video, a theatrical video, a reading, music, an image, etc. In this embodiment, generated content C1 is a theatrical video. Furthermore, the game server 110 analyzes the generated content C1. The first player terminal 10-1 then executes a game effect according to the analysis result. The game effect is the display of a video that includes a score for evaluating the generated content C1.
[0049] In the generated content data distribution process of (2), the first player uploads the generated content data D1 to the video server 210 using the first player terminal 10-1. As a result, the generated content data D1 is distributed to the second player terminal 10-2. That is, the second player can view the generated content C1 by operating the second player terminal 10-2.
[0050] In the machine learning model training process of (3), the second player can send evaluation information D3, which indicates the evaluation of the generated content C1, to the game server 110 by operating the second player terminal 10-2. The trained model 150 of the game server 110 performs machine learning using the combination of generated content data D1 and evaluation information D3 as training data.
[0051] (1) Generated content data generation process and analysis process First, the details of the generated content data generation and analysis processes will be explained. Figures 4 through 9 show the images displayed on the display 20 of the first player terminal 10-1. Figure 10 shows the settings table. Figure 11 is a flowchart of the processes executed by the control device 12 of the first player terminal 10-1 and the control device 112 of the game server 110 during the generated content data generation and analysis processes. Figure 12 is a flowchart of the subroutine for step S1 in Figure 11.
[0052] First, the control device 12 (setting acquisition means 120) of the first player terminal 10-1 acquires the setting data D2 input by the first player operating the operation unit 26 (step S1: setting acquisition process). The details of step S1 are described below. The setting data D2 is data that indicates the settings of the generated content C1. More specifically, the setting data D2 is data for identifying the content of the generated content C1. The setting data D2 is a setting table shown in Figure 10. In this embodiment, the generated content C1 is a video of a play. Therefore, the setting data D2 is data for identifying the content of the play. Thus, as shown in Figure 10, the setting data D2 includes casting settings (information about the actors in the play), proficiency settings, actor position settings (information about the actors in the play), spotlight settings (information about the lighting in the play), sound settings (information about the sound in the play), art settings (information about the art props in the play), and dialogue settings. The proficiency settings will be explained in the first modified example.
[0053] The control device 12 of the first player terminal 10-1 displays the scenario selection video shown in Figure 4 on the display 20. The scenario selection video includes five different scenarios. The first player selects "Twilight Street Corner" from the five scenarios by operating the operation unit 26 of the first player terminal 10-1. As a result, the control device 12 of the first player terminal 10-1 acquires scenario information indicating the scenario selected by the first player (step S201). Accordingly, the control device 12 of the first player terminal 10-1 records "Twilight Street Corner," the scenario selected by the first player, in the setting table shown in Figure 10.
[0054] Here, "Twilight Street Corner" includes multiple scenes. Therefore, the first player performs the following tasks for each of the multiple scenes: "casting," "actor positioning," "spotlighting," "sound design," "art design," and "dialogue setting," as described below.
[0055] First, the control device 12 of the first player terminal 10-1 sets N to 1 (step S202). Then, the control device 12 of the first player terminal 10-1 displays the casting video shown in Figure 5 on the display 20. The casting video is a video used to determine the casting of characters in scene N. For example, in scene 1, the characters are the protagonist and the heroine. The first player then selects the actors to be used for the protagonist and heroine by operating the operation unit 26 of the first player terminal 10-1. In this embodiment, the first player selects actor A for the protagonist and actor C for the heroine. As a result, the control device 12 of the first player terminal 10-1 acquires the casting information for scene N, which shows the actors selected by the first player (step S203). Accordingly, the control device 12 of the first player terminal 10-1 records actor A and actor C, who represent the scenario selected by the first player, in the setting table shown in Figure 10.
[0056] Next, the control device 12 of the first player terminal 10-1 displays the stage setting video shown in Figure 6 on the display 20. In the stage setting video, the first player can perform tasks such as setting actor positions, spotlights, sound, art, and dialogue. The first player starts the actor position setting process by touching the selection button for actor position setting. Specifically, Figure 6 includes the stage video. The first player determines the positions of actor A and actor C by operating the operation unit 26. As a result, the control device 12 of the first player terminal 10-1 acquires actor position information for scene N, which shows the positions of actor A and actor C determined by the first player (step S204). Accordingly, the control device 12 of the first player terminal 10-1 records the positions of actor A and actor C determined by the first player in the setting table shown in Figure 10.
[0057] Next, the first player touches the spotlight setting selection button on the control unit 26, initiating the spotlight setting process. Specifically, the control device 12 of the first player terminal 10-1 displays the spotlight setting video shown in Figure 7 on the display 20. The spotlight setting video includes stage footage. The first player determines the position, size, and color of the spotlight by operating the control unit 26. As a result, the control device 12 of the first player terminal 10-1 acquires the spotlight setting information for scene N, which shows the position, size, and color of the spotlight determined by the first player (step S205). Accordingly, the control device 12 of the first player terminal 10-1 records the position, size, and color of the spotlight determined by the first player in the setting table shown in Figure 10.
[0058] Next, the first player touches the sound setting selection button on the control unit 26 to begin the sound setting process. Specifically, the control device 12 of the first player terminal 10-1 displays the sound setting video on the display 20. The sound setting video includes a list of selectable background music (BGM) for scene N. The first player determines the BGM to be used for scene N by operating the control unit 26. As a result, the control device 12 of the first player terminal 10-1 acquires the sound setting information for scene N, which indicates the BGM determined by the first player (step S206). Accordingly, the control device 12 of the first player terminal 10-1 records the BGM determined by the first player in the setting table shown in Figure 10.
[0059] Next, the first player touches the art setting selection button on the control unit 26, initiating the art setting process. Specifically, the control device 12 of the first player terminal 10-1 displays the art setting video on the display 20. The art setting video includes stage footage. The first player determines the art props to be used in scene N and their positions by operating the control unit 26. As a result, the control device 12 of the first player terminal 10-1 acquires art setting information for scene N, indicating the art props and their positions determined by the first player (step S207). Accordingly, the control device 12 of the first player terminal 10-1 records the art props and their positions determined by the first player in the setting table shown in Figure 10.
[0060] Next, the first player starts the dialogue setting process by operating the control unit 26 and touching the dialogue setting selection button. Specifically, the control device 12 of the first player terminal 10-1 displays the dialogue setting video shown in Figure 8 on the display 20. The dialogue setting video includes the lines of the protagonist and the heroine. The first player can change the lines of the protagonist and the heroine by operating the control unit 26. As a result, the control device 12 of the first player terminal 10-1 acquires the dialogue information for scene N that shows the lines decided by the first player (step S208). Accordingly, the control device 12 of the first player terminal 10-1 records the lines decided by the first player in the setting table shown in Figure 10.
[0061] Next, the control device 12 of the first player terminal 10-1 determines whether the "casting settings," "actor position settings," "spotlight settings," "sound settings," "art settings," and "dialogue settings" for all scenes have been completed (step S209). If the work is not completed, the process proceeds to step S210. On the other hand, if the work is completed, the setting tables shown in Figure 10 for all scenes are completed. That is, the setting data D2 is completed. Then the process proceeds to step S2.
[0062] If the work is not completed, the control device 12 of the first player terminal 10-1 increments N by 1 (step S210). After this, the process returns to step S203. The control device 12 of the first player terminal 10-1 then repeats steps S203 to S210 until the "casting," "actor positioning," "spotlight setting," "sound setting," "art setting," and "dialogue setting" for all scenes are completed.
[0063] When the operation is complete, the control device 12 (content generation means 122) of the first player terminal 10-1 generates generated content data D1 indicating the generated content C1 based on the setting data D2 (step S2: content generation process). The program PG is designed to generate generated content data D1 based on the setting data D2. However, the control device 12 of the first player terminal 10-1 may also cause the generation AI to generate the generated content data D1 by inputting the setting data D2 as a prompt to the generation AI.
[0064] Next, the control device 12 of the first player terminal 10-1 transmits the generated content data D1 to the game server 110 via the network interface 16 (step S3). Accordingly, the network interface 116 of the game server 110 receives the generated content data D1 and outputs the generated content data D1 to the control device 112. As a result, the control device 112 of the game server 110 acquires the generated content data D1 (step S101). Here, the game server 110 (second server) is equipped with a trained model 150. Therefore, in step S3, the control device 12 of the first player terminal 10-1 outputs the generated content data D1 to the trained model 150 (output process).
[0065] Next, the control device 112 of the game server 110 inputs the generated content data D1 into the trained model 150. The trained model 150 then analyzes the generated content C1 (step S102: analysis process). The trained model 150 is described below. The trained model 150 is trained using a combination of training content and the analysis results of the training content as training data. The training content is a video of a play. The analysis results of the training content are evaluations of the training content by people who have viewed it. The analysis results of the training content include parameters and comments. The parameters include a parameter indicating the number of people who have given the training content an evaluation that includes predetermined characteristics. An evaluation that includes predetermined characteristics may be a high rating of the training content, or it may be a high rating of a part of the training content, such as the story or casting. Comments are the words used to evaluate the training content.
[0066] The trained model 150 described above analyzes the generated content C1 to generate analysis result information D4, which shows the analysis results of the generated content C1. The analysis result information D4 includes scores for the story, casting, acting, and stage setting. The maximum score for each item is 100 points.
[0067] Next, the control device 112 of the game server 110 transmits the analysis result information D4 to the first player terminal 10-1 via the network interface 116 (step S103). Accordingly, the network interface 16 of the first player terminal 10-1 receives the analysis result information D4 and outputs the analysis result information D4 to the control device 12. As a result, the control device 12 of the first player terminal 10-1 acquires the analysis result information D4 (step S4 - analysis result information acquisition process).
[0068] Next, the control device 12 of the first player terminal 10-1 executes a game effect after analyzing the generated content C1 (step S5: game effect execution process). In this embodiment, the game effect is the display of the analysis results of the generated content C1 by the trained model 150. The control device 12 of the first player terminal 10-1 generates result video data based on the analysis result information D4 and displays the result video shown in Figure 9 on the display 20 based on the result video data. The result video includes scores for the story, casting, acting, stage setting, and total score. Note that if the analysis results by the trained model 150 change, the scores included in the result video will change. That is, if the analysis results by the trained model 150 change, the content of the game effect executed by the control device 12 (execution means 126) of the first player terminal 10-1 will change. Through the above process, the first player can find out the evaluation of the generated content C1 that they created. After this, this process ends.
[0069] (2) Processing of generated content data Next, we will explain the details of the generated content data distribution process. Figure 13 is a flowchart of the actions performed by the control device 12 and video server 210 of the first player terminal 10-1 during the generated content data distribution process.
[0070] First, the first player operates the control unit 26 of the first player terminal 10-1 to distribute the generated content data D1 to the video server 210. Accordingly, the control device 12 (distribution means 128) of the first player terminal 10-1 transmits the generated content data D1 to the video server 210 (first server) (step S301 - distribution process). Accordingly, the video server 210 acquires the generated content data D1 (step S401).
[0071] The video server 210 (first server) distributes the generated content data D1 to the second player terminal 10-2 via the network line (step S402). As a result, the second player can view the generated content C1 using the second player terminal 10-2. After this, this process ends.
[0072] (3) Machine learning model training process Next, we will explain the details of the machine learning model training process. Figure 14 is a flowchart of the operations performed by the control device 12 and video server 210 of the first player terminal 10-1 during the generated content data distribution process. Figure 15 is an explanatory diagram of the machine learning model training process.
[0073] First, as shown in Figure 14, the second player inputs an evaluation of the generated content C1 by operating the operation unit 26 of the second player terminal 10-2. The evaluation includes parameters and comments. The parameters represent the evaluation of the generated content C1 using the number of stars. The number of stars ranges from one to five. A higher number of stars indicates a higher evaluation of the generated content C1. The comments are the words used to evaluate the generated content C1. In this embodiment, the second player inputs a rating of 5 stars as a parameter, and also inputs the words "The acting of the lead actor A was good." as a comment. As a result, the control device 12 of the second player terminal 10-2 generates evaluation information D5 indicating the evaluation of the generated content C1 by the second player (step S501).
[0074] Next, the control device 12 of the second player terminal 10-2 transmits evaluation information D5 to the game server 110 via the network interface 16 (step S502). Accordingly, the network interface 116 of the game server 110 receives the evaluation information D5 and outputs the evaluation information D5 to the control device 12. As a result, the control device 112 of the game server 110 (second server) obtains evaluation information D5 indicating the evaluation of the generated content C1 from the second player terminal 10-2 (second computer) (step S601).
[0075] The control device 112 of the game server 110 inputs the evaluation information D5 into the trained model 150. The trained model 150 then performs machine learning using the combination of generated content C1 and the evaluation of generated content C1 as training data (step S602). After this, the process terminates.
[0076] [effect] According to System 1, the generated content C1 created by the first player can be analyzed. The control device 12 of the first player terminal 10-1 generates generated content data D1 that represents the generated content C1 based on the setting data D2 input by the first player. This allows the first player to create the generated content C1. The trained model 150 then analyzes the generated content C1. The control device 12 of the first player terminal 10-1 executes game effects such that the content of the game effects changes when the analysis results by the trained model 150 change. Therefore, the first player can know or infer the analysis results through the game effects. Thus, according to System 1, the generated content C1 created by the first player can be analyzed.
[0077] In System 1, the game effect is a display of the analysis results of the generated content C1 by the trained model 150. Therefore, the first player can easily learn the analysis results through the game effect.
[0078] In System 1, the analysis results of the training content include a parameter indicating the number of people who evaluated the training content with predetermined features. This parameter quantitatively represents the evaluation of the training content. As a result, the trained model 150 can efficiently perform machine learning based on this parameter.
[0079] In System 1, the analysis results of the training content include comments that evaluate the training content. The comments contain a wide variety of evaluations of the training content. As a result, the trained model 150 can perform machine learning based on training content with many variations, and can output analysis results for generated content C1, which also has many variations.
[0080] In System 1, the trained model 150 performs machine learning using the combination of generated content C1 and the evaluation of generated content C1 as training data. This ensures that the trained model 150 continues to perform machine learning even while System 1 is in operation.
[0081] (First variation) Next, the system 1a relating to the first modified example will be described with reference to the drawings. Figure 16 shows the image displayed on the display 20 of the first player terminal 10-1 when in instruction mode.
[0082] System 1a differs from System 1 in that it allows for improvement of the generated content C1 by executing the instruction mode. The following describes System 1a, focusing on this difference.
[0083] The generated content C1 is a video of a play in which actors A and C perform. As shown in Figure 10, actors A and C have proficiency settings (proficiency parameters) that indicate their level of acting proficiency. After the flowchart shown in Figure 11 is executed, the control device 12 of the first player terminal 10-1 can execute the instruction mode. As shown in Figure 16, in instruction mode, the first player can provide acting instruction to actors A and C by operating the operation unit 26 of the first player terminal 10-1. Specifically, the first player can provide instruction on the volume of their voice, the positions of actors A and C, and the facial expressions of actors A and C by tapping the instruction video icons shown in Figure 16. When the first player provides appropriate instruction, the control device 12 of the first player terminal 10-1 increases the proficiency settings shown in Figure 10.
[0084] When the proficiency settings (proficiency parameters) of actors A and C change, the content of their acting changes. In this embodiment, as the proficiency setting increases, the actors' acting improves. Improved acting means that actors make fewer mistakes and are able to improvise more effectively. The other configurations and operations of system 1a are the same as those of system 1, so their explanation is omitted. Furthermore, system 1a can achieve the same effects as system 1.
[0085] In System 1a, as shown in Figure 10, actors A and C have proficiency settings (proficiency parameters) that indicate their level of acting proficiency. The first player can play the instruction mode to increase their proficiency settings. Therefore, the first player actively plays the instruction mode to improve the quality of the generated content C1.
[0086] (Second variation) Next, we will explain system 1b, which relates to the second modified example, with reference to the drawings. Figure 17 is an explanatory diagram of the operation of system 1b.
[0087] System 1b differs from System 1 in that the game effect is the development of the game after the analysis of generated content C1. More specifically, generated content C1 is a movie inserted into a story-driven game such as a role-playing game. That is, the first player can generate a movie as generated content C1. The game's development after the movie changes depending on the analysis result of generated content C1. For example, if the total score of the analysis result of generated content C1 is 300 points or more, the story leads to a development in which the protagonist (actor A) survives. If the total score of the analysis result of generated content C1 is less than 300 points, the story leads to a development in which the protagonist (actor A) dies. Note that the other configurations and operations of System 1b are the same as System 1, so the explanation is omitted. Also, System 1b can produce the same effects as System 1.
[0088] In System 1b, the game effect is the game's progression after the analysis of the generated content C1. In other words, the analysis results of the generated content C1 influence the development of the game's story. This increases the degree of freedom in the development of the game's story.
[0089] (Other embodiments) The various control means and processing procedures described in the above embodiments are examples and are not intended to limit the scope of the present invention, its applications, or its uses. The various control means and processing procedures can be modified as appropriate without altering the essence of the present invention.
[0090] The game effect may also be that the first player receives a reward for the analysis results of the generated content C1 by the trained model 150. The reward may be, for example, in-game currency, or data for characters or items that can be used in the game. The reward may also be that the first player can use settings in the settings data D2 that are normally unavailable to the first player. Settings that are normally unavailable to the first player may include, for example, the first player being able to select a rare actor X, or the first player being able to select a rare art prop. The reward may also be the right to perform a gacha, which grants data for actors, art props, or video data such as weather and backgrounds. The gacha is a system in which the first player uses in-game currency or money to randomly acquire items. In the gacha, the probability of the first player acquiring an item is set for each item. The probability of the user acquiring a rare item is lower. This encourages the first player to play the game in order to perform the gacha.
[0091] In addition, in systems 1, 1a, and 1b, restrictions may be placed on the second player who can use the generated content C1. The restriction may be, for example, age. That is, a second player under the age of 18 may not be able to view the generated content C1. In this case, the control device 12 (distribution means 128) of the first player terminal 10-1 may transmit condition data indicating the conditions for one or more second players who can use the generated content C1, along with the generated content data D1, to the video server 210 (first server). The condition is, as described above, that a second player under the age of 18 cannot view the generated content C1. Another condition may be that a second player under the age of 18 cannot rate the generated content C1. When a second player cannot rate the generated content C1, it means that the second player cannot access the rate page for the generated content C1, or even if they can access the rate page, they cannot input a rate. Furthermore, the age information of the second player is entered by the second player as registration information when they start playing the game. This registration information is, This information is stored by game server 110.
[0092] Furthermore, the proficiency setting (proficiency parameter) may change according to the number of times the theatrical video is played. Specifically, when the video server 210 distributes the generated content C1, the generated content C1 is played by the second player terminal 10-2. The video server 210 then transmits playback count information indicating the number of times the generated content C1 has been played to the first player terminal 10-1. The control device 12 of the first player terminal 10-1 can update the proficiency setting in the setting table shown in Figure 10 based on the playback count information. As a result, the first player terminal 10-1 can generate generated content data D1 with setting data D2 including the updated proficiency setting. In addition, during the playback of the theatrical video, the story of the play may change if actors A, C, etc. make mistakes or improvise, and the development of the game after the play (game movie) may also change. Furthermore, during the playback of the theatrical video, if actors A, C, etc., make mistakes or improvise, the reaction of the viewers of the theatrical video may change the development of the game after the play (game movie).
[0093] Furthermore, the control device 112 of the game server 110 may output a verbal evaluation (comment) as the result of analyzing the generated content data D1, rather than outputting a score. The verbal evaluation may be an evaluation of a specific item in the play, such as an evaluation of actor A's performance or an evaluation of the sound settings. Alternatively, the verbal evaluation may evaluate the play as a whole, such as whether the play was a comedy or a tragedy.
[0094] In system 1a, the game's progression after the analysis of generated content C1 may change depending on the content of the play (analysis result). More specifically, if the lighting is yellow, the background is sunny, actor A or actor C is a comedic actor, and actor A or actor C is smiling and acting happily, the trained model 150 will determine that the play is enjoyable. In this case, the story after the play will unfold in an enjoyable way. Also, if the lighting is red, the background is a thunderstorm, and actor A or actor C is acting angrily with a stern expression, the trained model 150 will determine that the play is a revenge drama. In this case, the story after the play will unfold in a revenge drama. Furthermore, if actor A is illuminated by a dark spotlight such as gray, and actor A is acting in an anxious manner, the trained model 150 will determine that the play is about a protagonist who is troubled. In this case, the story after the play will unfold in a way that amplifies the protagonist's anxiety. If actor A is illuminated by a dark spotlight, such as gray, but appears to be enjoying the performance, the trained model 150 will determine that the play is highly unexpected. In this case, the story after the play will take an unexpected turn.
[0095] The training data may, for example, be movie ratings distributed on websites or social media. In this case, the training content is the movie itself, and the analysis results of the training content are movie ratings. Alternatively, the training data may also be text reviews expressing impressions of the movie.
[0096] In systems 1, 1a, and 1b, if the second player terminal 10-2 sends evaluation information D3 to the game server 110, the second player may receive a reward. The reward may be, for example, in-game currency, or characters or items that can be used in the game. The reward may also be managed by, for example, blockchain.
[0097] Furthermore, the generated content C1 may be data that has been proven to be unique and irreplaceable by NFT (Nong Fungible Token).
[0098] In addition, in systems 1, 1a, and 1b, the analysis results may include scores for each emotion: joy, anger, sadness, and happiness.
[0099] In systems 1, 1a, and 1b, the number of second player terminals 10-2 only needs to be 1 or more.
[0100] In system 1b, the score may or may not be displayed on the first player terminal 10-1.
[0101] The generated content C1 may also be a collection of setting data D2 created by the first player. Setting data D2 includes information about theatrical performances, such as the positions of actors A and C, their lines, voices, and acting styles. Therefore, systems 1, 1a, and 1b can generate setting data D2 that quantifies all theatrical information, such as pitch, the speaking speed of actors A and C, the coordinates of their positions, and the movement data of the avatars that perform. Thus, systems 1, 1a, and 1b can generate videos using setting data D2.
[0102] In System 1a, the first player may provide guidance on improvisation. Specifically, the first player can provide guidance to actor A regarding the volume of their voice, the content of the improvised lines, and so on.
[0103] In System 1a, the first player may provide instructions regarding the intonation of actors A and C's voices, their directions and positions, and the movements of their hands and feet.
[0104] Furthermore, if the first player creates multiple plays in the same scenario, the proficiency levels of actors A and C in system 1a may be carried over without being reset. Also, proficiency may include multiple types of parameters. In this case, certain parameters may be carried over without being reset. Additionally, there may be a master-apprentice relationship between the two actors. In multiple plays in the same scenario, the proficiency levels of the two actors in a master-apprentice relationship may be carried over.
[0105] Furthermore, the level of proficiency may change as the theatrical video is played. The increase in proficiency when the theatrical video is played may differ from the increase in proficiency when the proficiency is played in instruction mode. For example, the increase in proficiency when the theatrical video is played may be greater than the increase in proficiency when the proficiency is played in instruction mode.
[0106] In addition, in system 1b, a flag may be generated based on the evaluation obtained from the analysis results of the generated content C1. If this flag is generated, the game's progression may branch if a specific event occurs during the game's progression after the movie.
[0107] Furthermore, in System 1b, the attributes of game elements such as characters and weapons may change based on the evaluation obtained from the analysis of generated content C1. For example, if the presentation of generated content C1 is evaluated as exciting, the weapon may change to the fire attribute, and if the presentation of generated content C1 is evaluated as cold, the weapon may change to the ice attribute. The attributes of weapons, etc., are related to the compatibility between game elements. Specifically, the water attribute is strong against the fire attribute. In addition, the parameters of characters may be strengthened or enemy characters may be weakened based on the evaluation obtained from the analysis of generated content C1.
[0108] Furthermore, actors and characters may acquire nicknames through theatrical performances. These nicknames may be generated by a machine learning model based on the characteristics of the performance.
[0109] In addition to actors and directors, various other characters may be involved in the production, such as art directors, acting coaches, music coaches, choreographers, and instrumentalists. These characters may be NPCs (Non-Player Characters) or PCs (Player Characters). NPCs may be assigned ratings, and PCs may be assigned ratings. Furthermore, NPCs may be able to grow by using the ratings they receive as learning data.
[0110] Furthermore, the game's progression before the analysis of generated content C1 may change based on the analysis results of the generated content.
[0111] The generated content C1 may be a musical. In this case, the evaluation of the songs in the musical will be emphasized. The singing ability of the actors and the quality of the lyrics may also be evaluated. The quality of the lyrics may be, for example, the strength of their relevance to the musical's story.
[0112] The generated content C1 may be a mystery film. In this case, the unexpectedness of the culprit may be included in the evaluation.
[0113] Furthermore, in the evaluation of generated content C1, plot twists that are too illogical, such as endings in a dream, may receive a low rating.
[0114] Additionally, a compatibility parameter indicating how well actors get along with each other may be set for each actor. Furthermore, compatibility between actors may be higher, for example, if they belong to the same talent agency.
[0115] Based on the generated content C1 in which the actor has previously appeared, the actor's strengths and weaknesses in different roles may be determined. Furthermore, based on the generated content C1 in which the actor has previously appeared, it may be determined whether the casting of the actor is unexpected or not. This casting may then be considered in evaluating the actor's performance.
[0116] The game could also be a simulation game about managing a theater company. Specifically, in the game, the first player would hold a series of plays and earn in-game currency based on the ratings of the plays. The first player would then use the earned in-game currency for stage expenses, actor salaries, and operating costs for the academy or theater company. In this case, the game could use ratings from other users only, or it could be automatically evaluated using only AI ratings. If using ratings from other users, the performance times for the plays could be reserved and decided in advance.
[0117] The effects and advantages of this embodiment will also be achieved when these other embodiments are adopted. Furthermore, it is possible to combine this embodiment with other embodiments, and other embodiments with each other, as appropriate. [Explanation of Symbols]
[0118] 1,1a,1b: System 10-1: First Player Terminal 10-2: Second Player Terminal 12: Control device 14: Storage part 16: Network Interface 18: Graphics Processing Unit 20: Display 22: Audio Processing Unit 24: Speaker 26:Operation unit 110: Game Server 112: Control device 114: Storage device 116: Network Interface 120: Method for obtaining settings 122: Content generation methods 124: Output means 126: Means of Execution 128: Distribution method 150: Pre-trained model 210: Video Server C1: Generated content D1: Generated content data D2: Configuration Data D3: Evaluation Information D4: Analysis results information D5: Evaluation Information PG: Program
Claims
1. The program causes the control unit of the first computer to function as a setting acquisition means, a content generation means, an output means, and an execution means. Configuration data is data that indicates the settings for the generated content. The setting acquisition means acquires the setting data input by the first player operating the computer, The content generation means generates generated content data indicating the generated content based on the setting data, The output means can output the generated content data to the trained model. The trained model can analyze the generated content, The execution means executes game effects after analyzing the generated content. When the analysis results from the trained model change, the content of the game effect executed by the execution means changes. program.
2. The aforementioned trained model uses a combination of training content and the analysis results of the training content as training data. The program according to claim 1.
3. The analysis results of the aforementioned teacher content include a parameter indicating the number of people who evaluated the teacher content including predetermined characteristics. The program according to claim 2.
4. The analysis results of the aforementioned teacher content include comments evaluating the teacher content. The program according to claim 2 or claim 3.
5. The program is further configured to function as a distribution means in the control device of the first computer. The distribution means can transmit the generated content data to the first server. The first server can distribute the generated content data to one or more second computers via a network line. The second server is equipped with the trained model and can acquire evaluation information indicating the evaluation of the generated content from the one or more second computers. The trained model performs machine learning using the combination of the generated content and the evaluation of the generated content as training data. The program according to any one of claims 1 to 3.
6. The distribution means can transmit conditional data indicating the conditions for one or more second players who can use the content indicated by the generated content data to the first server together with the generated content data. The program according to claim 5.
7. The aforementioned game effect is the game's development after the analysis of the generated content. The program according to any one of claims 1 to 3.
8. The aforementioned game effect is a display of the analysis results of the generated content by the trained model. The program according to any one of claims 1 to 3.
9. The aforementioned game effect is that a reward is given to the first player based on the analysis results of the generated content by the trained model. The program according to any one of claims 1 to 3.
10. The generated content is a video of a theatrical performance. The aforementioned configuration data includes information about the actors in the play, information about the lighting in the play, information about the sound in the play, or information about the stage props in the play. The program according to any one of claims 1 to 3.
11. The generated content is a video of a play in which actors perform. The aforementioned actors are assigned proficiency parameters that indicate their level of acting proficiency. When the aforementioned actor's proficiency parameter changes, a change occurs in the content of their performance. The program according to any one of claims 1 to 3.
12. The aforementioned proficiency parameter changes according to the number of times the video of the play is viewed. The program according to claim 11.