Interactive narrative educational game intelligent generation method, system, device, medium and program product

Abstract

The invention belongs to the field of teaching application of information technology, and provides an interactive narrative education game intelligent generation method, system and device, a medium and a program product. The method comprises the following steps: (1) intelligent generation of interactive narration: constructing a lightweight intelligent generation model of interactive narration to generate game narration content, and constructing a multi-modal scene generation engine to generate corresponding multi-modal scene content based on the narration content; (2) user interaction feedback: a user responds to game content through voice or text, and narrative content is adjusted according to user feedback; and (3) personalized cognitive adaptation: dynamically adjusting the narrative content and the game difficulty by evaluating the element cognitive ability of the user. The dynamic narrative education game is generated through interaction with the user, the cognitive ability of the learner can be adapted, the personalized learning requirement of the learner can be met, and a new path is provided for constructing the interactive narrative education game.

Description

Technical Field

[0001] The present invention belongs to the field of teaching applications of information technology, and more specifically, relates to an interactive narrative educational game intelligent generation method, system, device, medium and program product. Background Art

[0002] Educational games can effectively stimulate students' learning motivation and cultivate high-level abilities such as problem solving, communication and collaboration, and creativity by creating learning situations and enhancing learning interactions.

[0003] Existing interactive educational games have the following technical bottlenecks: (1) The game content is fixed. The narrative generation of traditional games relies on manual script design, which is difficult to dynamically adapt to the knowledge structure of different disciplines, resulting in a rigid combination of teaching content and game plots, lack of knowledge progression logic, and inability to meet the needs of different learners; (2) The efficiency of multimodal content generation is low. Existing systems often use prefabricated material libraries to splice together, which is unable to dynamically generate scenes, dialogues, and character actions that match the narrative logic, resulting in a long development cycle and low personalization; (3) The cognitive adaptation mechanism is missing. Most educational games use a fixed difficulty curve and do not consider the dynamic differences in users' metacognitive abilities (such as self-correction and strategy adjustment abilities), which affects learning outcomes. These problems make it difficult for existing educational games to meet learners' personalized learning needs. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and to provide an interactive narrative educational game intelligent generation method, system, device, medium and program product, which can automatically generate personalized educational games that meet the learners' metacognitive abilities through interaction with the learners.

[0005] The purpose of the present invention is achieved through the following technical measures.

[0006] The present invention provides an interactive narrative educational game intelligent generation method, comprising the following steps:

[0007] (1) Intelligent generation of interactive narratives: building a lightweight intelligent generation model for interactive narratives to generate game narrative content, and building a multimodal scene generation engine to generate corresponding multimodal scene content based on the narrative content;

[0008] (2) User interaction feedback, where users respond to game content through voice or text, and the narrative content is adjusted based on user feedback;

[0009] (3) Personalized cognitive adaptation: dynamically adjusting narrative content and game difficulty by evaluating the user’s metacognitive ability.

[0010] The present invention also provides an interactive narrative educational game intelligent generation system, including an interactive narrative intelligent generation module, a user interactive feedback module, and a personalized cognitive adaptation module;

[0011] The interactive narrative intelligent generation module includes a knowledge graph mapping submodule, a lightweight interactive narrative intelligent generation model, and a multimodal scene generation engine, which generates narrative content and multimodal game scenes based on user input combined with the knowledge graph;

[0012] The user interaction feedback module includes a user input submodule and a narrative content adjustment submodule, which triggers narrative content adjustment according to user input;

[0013] The personalized cognitive adaptation module includes a metacognitive ability evaluation submodule and a difficulty adjustment strategy generation submodule, which evaluates the user's metacognitive ability according to the user's interactive content and adjusts the narrative content and difficulty based on the ability value.

[0014] The present invention also provides an electronic device, comprising a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the above-mentioned interactive narrative educational game intelligent generation method.

[0015] The present invention also provides a computer-readable storage medium, which stores a plurality of instructions, and the instructions are suitable for a processor to load to execute the steps in the above-mentioned interactive narrative educational game intelligent generation method.

[0016] The present invention also provides a computer program product, including a computer program / instruction, which implements the steps in the above-mentioned interactive narrative educational game intelligent generation method when executed by a processor.

[0017] The present invention constructs an interactive narrative educational game intelligent generation system, which can adapt to learners' cognitive abilities and meet their personalized learning needs by interacting with users to generate dynamic narrative educational games. In addition, there are currently no other public methods for intelligently generating interactive narrative educational games. The present invention provides a new path for constructing interactive narrative educational games. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the intelligent generation method of interactive narrative educational games in an embodiment of the present invention.

[0019] Figure 2 It is an architecture diagram of the intelligent generation system of interactive narrative educational games in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and implementation cases. It should be understood that the specific implementation cases described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0021] This embodiment provides an interactive narrative educational game intelligent generation method, such as Figure 1 As shown, the following steps are included:

[0022] (1) Intelligent generation of interactive narratives: construct a lightweight intelligent generation model for interactive narratives to generate game narrative content, and construct a multimodal scene generation engine to generate corresponding multimodal scene content based on the narrative content.

[0023] (1-1) Narrative logic generation based on knowledge graph and large model distillation generates narrative content that meets the teaching objectives according to the user-specified learning stage, subject, knowledge unit, game style (including fantasy, science fiction, oriental, casual, two-dimensional, etc.), character gender and other parameters, combined with the corresponding knowledge graph. For example, if the user specifies junior high school, mathematics, plane geometry, and science fiction adventure, plane geometry-related entities and relationships are extracted from the knowledge graph to generate narrative content such as "The interstellar expedition team repairs the space station energy system by solving geometric puzzles."

[0024] (1-1-1) Knowledge graph mapping: vectorize the learning stage, subject, and knowledge unit parameters input by the user and establish a mapping relationship with the knowledge graph entity.

[0025] (1-1-2) Perform knowledge distillation on the pre-trained large model, build a lightweight narrative generation model, and dynamically generate narrative content based on user input combined with the mapped knowledge graph.

[0026] (1-1-2-1) Use a large language model such as Qwen-70B to act as a student, randomly output different stages of study, subjects, knowledge units, game styles, and genders, and use a large reasoning model such as DeepSeek-r1 to generate overall narrative training data based on the given stage of study, subject, knowledge unit, game style, and gender, combined with related knowledge triples.

[0027] (1-1-2-2) Randomly select game scenario questions from different parts of the overall narrative training data and provide them to the student language model. Then, the output results of the student language model are fed back to the reasoning model to generate interactive narrative training data.

[0028] (1-1-2-3) Using the generated overall narrative and interactive narrative training data, the SFT method is used for supervised fine-tuning of small models (less than 8B) such as Qwen-3B to obtain a lightweight narrative generation model that can generate narrative content in real time.

[0029] (1-2) Multimodal scene generation. The construction of a multimodal scene generation engine includes: a text-image generation model, a speech synthesis model, and a speech-driven character action model. The text-image generation model is used to generate background images that match the narrative context, the speech synthesis model is used to generate character dialogues, and the speech-driven character action model is used to generate expressions and actions.

[0030] (1-2-1) Text-to-image generation models that use a diffusion model architecture, such as the Hunyuan-DiT model, use a large language model to generate scene description prompts based on the current narrative content and generate background images that match the narrative plot, such as "The interstellar expedition team repairs the polygonal console in the weightless environment of the space station's energy system by solving geometric puzzles."

[0031] (1-2-2) Use a speech synthesis model such as ChatTTS, combined with the character’s gender, to generate dialogue audio from the dialogue content in the narrative, such as “Please calculate the sum of the interior angles of quadrilateral ABCD to start the engine.”

[0032] (1-2-3) Use a voice-driven character expression and action generation model such as the TANGO model to drive the character's expression and action based on the generated dialogue audio.

[0033] (2) User interaction feedback: users respond to game content through voice or text, and the narrative content is adjusted based on user feedback.

[0034] (2-1) User input: The user inputs a response to the current narrative content through a text box or voice buttons.

[0035] (2-2) Narrative content adjustment: In the first N rounds of interaction with the user (N is generally 2-3), the user's response content is directly fed back to the lightweight narrative generation model in step (1-1-2). After N rounds of interaction, the narrative content is adjusted through the personalized cognitive adaptation module.

[0036] (3) Personalized cognitive adaptation: dynamically adjusting narrative content and game difficulty by evaluating the user’s metacognitive ability.

[0037] (3-1) Metacognitive ability assessment: the user's ability value is calculated through the metacognitive ability assessment submodule.

[0038] (3-1-1) Collect five-dimensional time series data including error rate, number of retries, time spent on answering questions, seeking help, and learning path completion, and train a metacognitive ability assessment model to evaluate users’ knowledge mastery.

[0039] (3-1-2) A sliding time window is used to calculate the user's metacognitive ability value within the window after each interaction.

[0040] (3-2) Difficulty adjustment strategy generation: Dynamically generate difficulty adjustment strategy based on capability value.

[0041] (3-2-1) Construct a multi-dimensional difficulty adjustment matrix and establish a mapping relationship between metacognitive ability value and adjustment parameters such as question complexity, prompt number limit, and time constraint.

[0042] (3-2-2) Generate prompt words according to the adjustment parameters corresponding to the ability values ​​in the difficulty adjustment matrix, and guide the lightweight narrative generation model in step (1-1-1) to adjust the narrative content.

[0043] This embodiment also provides an interactive narrative educational game intelligent generation system, such as Figure 2 As shown, it includes interactive narrative intelligent generation module, user interaction feedback module, and personalized cognitive adaptation module;

[0044] The interactive narrative intelligent generation module includes a knowledge graph mapping submodule, a lightweight interactive narrative intelligent generation model, and a multimodal scene generation engine, which generates narrative content and multimodal game scenes based on user input combined with the knowledge graph;

[0045] The user interaction feedback module includes a user input submodule and a narrative content adjustment submodule, which triggers narrative content adjustment according to user input;

[0046] The personalized cognitive adaptation module includes a metacognitive ability evaluation submodule and a difficulty adjustment strategy generation submodule, which evaluates the user's metacognitive ability according to the user's interactive content and adjusts the narrative content and difficulty based on the ability value.

[0047] This embodiment also provides an electronic device, which can be a terminal, a server, etc. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a notebook computer, a personal computer, etc. The server can be a single server or a server cluster composed of multiple servers, etc.

[0048] In this embodiment, the electronic device of this embodiment is a server as an example for detailed description. For example, the server may include one or more processors of processing cores, one or more computer-readable storage media memories, a power supply, an input module, a communication module and other components. Among them:

[0049] The processor is the control center of the server. It uses various interfaces and lines to connect various parts of the entire server. It executes various functions of the server and processes data by running or executing software programs and / or modules stored in the memory and calling data stored in the memory. In some embodiments, the processor may include one or more processing cores; in some embodiments, the processor may integrate an application processor and a modem processor, wherein:

[0050] The application processor mainly processes the operating system, user interface and application programs, and the modem processor mainly processes wireless communications. It is understandable that the modem processor may not be integrated into the processor.

[0051] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the server, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0052] The server also includes a power supply for supplying power to various components. In some embodiments, the power supply can be connected to the processor logic through a power management system, so that the power management system can manage charging, discharging, and power consumption management. The power supply can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0053] The server may also include an input module, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0054] The server may also include a communication module, and in some embodiments, the communication module may include a wireless module, through which the server may perform short-range wireless transmission, thereby providing users with wireless broadband Internet access. For example, the communication module may be used to help users send and receive emails, browse web pages, and access streaming media.

[0055] The server may also include a display unit, etc., which will not be described in detail here. Specifically in this embodiment, the processor in the server will load the executable files corresponding to the processes of one or more applications into the memory according to the following instructions, and the processor will run the application stored in the memory to achieve various functions, as follows:

[0056] Interactive narrative intelligent generation: build a lightweight interactive narrative intelligent generation model to generate game narrative content, and build a multimodal scene generation engine to generate corresponding multimodal scene content based on narrative content;

[0057] User interaction feedback, where users respond to game content through voice or text, and the narrative content is adjusted based on user feedback;

[0058] Personalized cognitive adaptation dynamically adjusts narrative content and game difficulty by evaluating user metacognitive abilities.

[0059] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0060] To this end, this embodiment also provides a computer-readable storage medium, which stores a plurality of instructions, which can be loaded by a processor to execute the steps in any of the interactive narrative educational game intelligent generation methods provided in the embodiments of the present application. For example, the instructions can execute the following steps:

[0061] Interactive narrative intelligent generation: build a lightweight interactive narrative intelligent generation model to generate game narrative content, and build a multimodal scene generation engine to generate corresponding multimodal scene content based on narrative content;

[0062] User interaction feedback, where users respond to game content through voice or text, and the narrative content is adjusted based on user feedback;

[0063] Personalized cognitive adaptation dynamically adjusts narrative content and game difficulty by evaluating user metacognitive abilities.

[0064] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0065] According to one aspect of the present application, a computer program product or a computer program is also provided, the computer program product or the computer program including a computer program / instruction, the computer program / instruction being stored in a computer-readable storage medium. A processor of a computer device reads the computer program / instruction from the computer-readable storage medium, and the processor executes the computer program / instruction, so that the computer device executes the method provided in various optional implementations provided in the above embodiments.

[0066] The contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0067] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent generation method for interactive narrative educational games, characterized in that The method comprises the following steps: (1) Intelligent generation of interactive narratives: building a lightweight intelligent generation model for interactive narratives to generate game narrative content, and building a multimodal scene generation engine to generate corresponding multimodal scene content based on the narrative content; (2) User interaction feedback, where users respond to game content through voice or text, and the narrative content is adjusted based on user feedback; (3) Personalized cognitive adaptation: dynamically adjusting narrative content and game difficulty by evaluating the user’s metacognitive ability.

2. The method for intelligently generating interactive narrative educational games according to claim 1, characterized in that The interactive narrative intelligent generation in step (1) is specifically as follows: (1-1) Narrative logic generation based on knowledge graph and large model distillation, based on the learning stage, subject, knowledge unit, game style, character gender parameters specified by the user, combined with the corresponding knowledge graph to generate narrative content that meets the teaching objectives; (1-1-1) Knowledge graph mapping: vectorize the learning stage, subject, and knowledge unit parameters input by the user and establish a mapping relationship with the knowledge graph entity; (1-1-2) Perform knowledge distillation on the pre-trained large model, build a lightweight narrative generation model, and dynamically generate narrative content based on user input combined with the mapped knowledge graph; (1-1-2-1) Use the large language model to play the role of students, randomly output different stages of study, subjects, knowledge units, game styles and genders, and use the reasoning large model to generate overall narrative training data based on the given stages of study, subjects, knowledge units, game styles and genders, combined with the associated knowledge triples; (1-1-2-2) Randomly select game scenario questions from different parts of the overall narrative training data and provide them to the student language model. Then, the output results of the student language model are fed back to the reasoning model to generate interactive narrative training data. (1-1-2-3) Using the generated overall narrative and interactive narrative training data, the SFT method is used to fine-tune the small model through supervision, thus obtaining a lightweight narrative generation model that can generate narrative content in real time; (1-2) Multimodal scene generation: building a multimodal scene generation engine including: text-image generation model, speech synthesis model, and speech-driven character action model. The text-image generation model is used to generate background images that match the narrative context, the speech synthesis model is used to generate character dialogues, and the speech-driven character action model is used to generate expressions and actions. (1-2-1) A text-image generation model based on a diffusion model architecture is used to generate scene description prompts based on the current narrative content using a large language model, and to generate background images that match the narrative plot; (1-2-2) Using a speech synthesis model and taking into account the gender of the characters, the dialogue content in the narrative content is generated into dialogue audio; (1-2-3) A voice-driven character expression and action generation model is used to drive the character's expression and action based on the generated dialogue audio.

3. The method for intelligently generating interactive narrative educational games according to claim 1, characterized in that The user interaction feedback in step (2) is specifically: (2-1) User input: the user inputs a response to the current narrative content through a text box or voice button; (2-2) Narrative content adjustment: In the first N rounds of interaction with the user, where N is 2-3, the content of the user's reply is directly fed back to the lightweight narrative generation model in step (1). After N rounds of interaction, the narrative content is adjusted through the personalized cognitive adaptation module.

4. The method for intelligently generating interactive narrative educational games according to claim 1, characterized in that The personalized cognitive adaptation in step (3) is specifically as follows: (3-1) Metacognitive ability assessment, calculating the user's ability value through the metacognitive ability assessment submodule; (3-1-1) Collect five-dimensional time series data including error rate, number of retries, time spent on answering questions, seeking help, and learning path completion, and train a metacognitive ability assessment model for evaluating user knowledge mastery; (3-1-2) Using a sliding time window, the metacognitive ability value of the user in the window is calculated after each interaction; (3-2) Difficulty adjustment strategy generation, dynamically generating difficulty adjustment strategies based on capability values; (3-2-1) Construct a multidimensional difficulty adjustment matrix and establish the mapping relationship between metacognitive ability value and question complexity, prompt number limit, and time constraint adjustment parameters; (3-2-2) Generate prompt words according to the adjustment parameters corresponding to the ability values ​​in the difficulty adjustment matrix, and guide the lightweight narrative generation model in step (1) to adjust the narrative content.

5. An intelligent generation system for interactive narrative educational games, characterized in that: The system includes an interactive narrative intelligent generation module, a user interaction feedback module, and a personalized cognitive adaptation module; The interactive narrative intelligent generation module includes a knowledge graph mapping submodule, a lightweight interactive narrative intelligent generation model, and a multimodal scene generation engine, which generates narrative content and multimodal game scenes based on user input combined with the knowledge graph; The user interaction feedback module includes a user input submodule and a narrative content adjustment submodule, which triggers narrative content adjustment according to user input; The personalized cognitive adaptation module includes a metacognitive ability evaluation submodule and a difficulty adjustment strategy generation submodule, which evaluates the user's metacognitive ability according to the user's interactive content and adjusts the narrative content and difficulty based on the ability value.

6. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the intelligent generation method of an interactive narrative educational game as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the method for intelligently generating an interactive narrative educational game as described in any one of claims 1 to 4.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps in the method for intelligently generating an interactive narrative educational game as described in any one of claims 1 to 4 are implemented.

Images