An artificial intelligence-based language learning simulation scenario generation method, medium, device and product
By preprocessing user data and generating virtual NPC characters using artificial intelligence models, and dynamically adjusting the learning scenario, the problem of insufficient personalization in existing technologies is solved, realizing a personalized and immersive language learning environment and improving learning efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING UNIV OF INFORMATION SCI & TECH
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies lack in-depth processing of semantic understanding, emotion recognition, and contextual association, making it difficult to generate personalized language learning environments and resulting in low learning efficiency.
By preprocessing user data, user profiles are generated, and personalized language learning scenarios are created using artificial intelligence models, including virtual NPC roles and multi-turn interactions. The scenario descriptions and roles are evaluated and updated in real time, and the learning difficulty is dynamically adjusted.
It has created a highly personalized and immersive language learning environment, which has improved learners' language proficiency and intercultural communication skills, and solved the problem of low efficiency in traditional learning methods.
Smart Images
Figure CN122154457A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer-aided education technology, and in particular to a method, medium, device, and product for generating language learning simulation scenarios based on artificial intelligence. Background Technology
[0002] Existing methods primarily rely on simple template matching or shallow natural language processing techniques, lacking in-depth processing of semantic understanding, sentiment recognition, and contextual association, making it difficult to simulate the multi-dimensional interactions in complex dialogues. Furthermore, they cannot be personalized based on learners' language proficiency, learning goals, and interests, failing to achieve individualized instruction and resulting in low learning efficiency. Summary of the Invention
[0003] The purpose of this invention is to overcome the technical problems of insufficient personalization and low efficiency in the prior art, and to provide a method, medium, device and product for generating language learning simulation scenarios based on artificial intelligence, which can generate a highly personalized and immersive language learning environment and effectively improve learners' language application ability and cross-cultural communication ability.
[0004] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution:
[0005] In a first aspect, the present invention provides a method for generating language learning simulation scenarios based on artificial intelligence, comprising:
[0006] The acquired user data is preprocessed to obtain user profiles;
[0007] Based on a pre-defined scenario theme library and the user profile, an artificial intelligence model is used to generate scenario descriptions;
[0008] Based on a pre-set library of scene NPC character models and the scene description, virtual NPC characters are generated using an artificial intelligence model;
[0009] Virtual NPC characters interact with users in multiple rounds until the scene learning task is completed;
[0010] In this process, after each round of interaction, the user's response during the interaction is evaluated using an evaluation model to obtain the evaluation result;
[0011] Update the user profile based on the evaluation results and interaction content, and adjust the scene description and virtual NPC characters according to the updated user profile before proceeding to the next round of interaction.
[0012] Optionally, the step of preprocessing the acquired user data to obtain a user profile includes:
[0013] The acquired user data is cleaned to remove noisy data and incomplete records, resulting in cleaned data.
[0014] Based on the content of the cleaned data, user profiles are obtained by categorizing and annotating metadata.
[0015] Optionally, the user profile includes the target language to be learned, learning objectives, and current language proficiency.
[0016] Optionally, the user profile may also include user personality and historical learning behavior data.
[0017] Optionally, the scene NPC character model library includes multiple scene theme basic templates. Each scene theme basic template includes the scene's background description, language, region, customs and taboos, key scene NPC characters, scene rules, learning tasks, and key topics.
[0018] Optionally, the scene NPC character model library contains a variety of virtual NPC character models, each of which includes gender, personality traits, accent characteristics, knowledge background, and dialogue style.
[0019] Optionally, the evaluation model evaluates user responses based on a comprehensive scoring formula;
[0020] The comprehensive scoring formula is as follows:
[0021] Score=[(V*Wv)+(G*Wg)+(A*Wa)+(F*Wf)]*R / 10*D− T,
[0022] Wherein, Score represents the overall score, V represents vocabulary accuracy, G represents grammatical accuracy, A represents accuracy, F represents fluency, R represents reaction speed, D represents difficulty level, T represents deduction for forbidden words, Wv represents the weight of vocabulary accuracy V, Wg represents the weight of grammatical accuracy G, Wa represents the weight of accuracy A, Wf represents the weight of fluency F, and Wv + Wg + Wa + Wf = 1.
[0023] The vocabulary accuracy V is obtained using the following formula:
[0024] V = Percentage of accuracy in total vocabulary * 10;
[0025] The grammatical accuracy G is obtained by the following formula:
[0026] G = Total grammar usage accuracy percentage * 10;
[0027] The accuracy A is obtained by the following formula:
[0028] A = Total accuracy percentage * 10;
[0029] The fluency F is obtained by the following formula:
[0030] F = Total percentage of fluency in responses * 10;
[0031] The reaction rate R is obtained by the following formula:
[0032] R = Number of conversations * 3 / Total number of seconds of pause in user responses;
[0033] The difficulty coefficient D is obtained using the following formula:
[0034] D = (1 + Scene Difficulty / 10) * (1 + Number of NPCs / 10),
[0035] The scene difficulty is 1, 2, 3, ..., 10.
[0036] The penalty T for prohibited words is obtained using the following formula:
[0037] T = (Preset total score * 30%) * Number of times the forbidden word appears.
[0038] In a second aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of any of the artificial intelligence-based language learning simulation scene generation methods described in the first aspect.
[0039] Thirdly, the present invention provides a computer device, comprising:
[0040] Memory, used to store computer instructions;
[0041] A processor for executing the computer instructions to implement the steps of the artificial intelligence-based language learning simulation scene generation method described in any of the first aspects.
[0042] Fourthly, the present invention provides a computer program product, including computer instructions, characterized in that, when the computer instructions are executed by a processor, they implement the steps of the artificial intelligence-based language learning simulation scene generation method described in any of the first aspects.
[0043] Compared with existing technologies, the beneficial effects achieved by this invention are as follows:
[0044] 1. An interactive virtual scene was constructed, which includes background settings, characters, task objectives, and a dynamic dialogue tree. It can generate a highly personalized and immersive language learning environment, effectively improving learners' language proficiency and intercultural communication skills.
[0045] 2. By analyzing users' voice and text input in real time through speech recognition and natural language processing technologies, the system drives the evolution of the scene and provides context-related error correction and feedback. This solves the problems of traditional learning methods being detached from reality and inefficient. It also creates a unique foreign language corner, enhancing learners' learning abilities through real-world interaction. Attached Figure Description
[0046] Figure 1 This is a flowchart of a method for generating simulated language learning scenarios based on artificial intelligence, according to an embodiment of the present invention. Detailed Implementation
[0047] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.
[0048] It should be noted that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.
[0049] Example 1
[0050] This invention discloses a method for generating simulated language learning scenarios based on artificial intelligence, with reference to... Figure 1 As shown, the specific steps include the following:
[0051] S1, perform data preprocessing on the acquired user data to obtain user profiles;
[0052] S2, Based on the preset scene theme library and the user profile, generate scene descriptions using an artificial intelligence model;
[0053] S3, Based on the preset scene NPC character model library and the scene description, generate virtual NPC characters using an artificial intelligence model;
[0054] S4 utilizes virtual NPC characters to interact with users in multiple rounds until the scene learning task is completed.
[0055] In this process, after each round of interaction, the user's response during the interaction is evaluated using an evaluation model to obtain the evaluation result;
[0056] Update the user profile based on the evaluation results and interaction content, and adjust the scene description and virtual NPC characters according to the updated user profile before proceeding to the next round of interaction.
[0057] Specifically, in step S1, data preprocessing is a crucial foundation for ensuring the quality of subsequent scene generation and interaction. By preprocessing the acquired user data to extract target feature values and standardize the data format, a user profile is obtained, including three steps: data cleaning, classification, and annotation.
[0058] First, the received natural speech and text content undergoes data cleaning to remove noise and incomplete records. The aim is to remove noise and incomplete records from the received natural speech and text content, ensuring data quality and consistency.
[0059] Secondly, the classification process divides the data into different categories based on the content and language difficulty, which facilitates the subsequent calling and management of learning scenario models;
[0060] Finally, annotation adds metadata to the data, including the user's language proficiency level, learning preferences, and the expected simulated learning scenarios. Creating user profiles and this annotated information significantly improves the accuracy and intelligence of learning scenario generation, thus providing high-quality input data for building learning scenario models.
[0061] In this embodiment, the user profile includes at least the target language, learning goal, and current language level, and usually also includes user personality and historical learning behavior data.
[0062] In this embodiment, the artificial intelligence model mainly includes a deep learning model framework and a scene generation engine, which are the core components for realizing intelligent language learning simulation scene generation in this embodiment. The artificial intelligence model integrates pre-trained artificial intelligence models through deep integration of artificial intelligence technology, realizes accurate understanding of user input, intelligent construction of dynamic scenes, and natural advancement of multi-round interactions. According to the user's real-time performance (such as comprehension accuracy, reaction speed, and error frequency), the complexity of the scene is dynamically adjusted, such as adding scene NPC characters, increasing or decreasing the scope of interactive content, and using slang expressions, to achieve progressive ability improvement.
[0063] In step S2, the scene NPC character model library includes a large number of preset scene theme basic templates, such as "ordering food in a restaurant", "checking in at the airport", "job interview", "shopping in a mall", "business meeting" etc.; each scene theme basic template includes the scene's background description, language, region, customs and taboos, key scene NPC characters, scene rules, learning tasks and key topics, etc.
[0064] In step S3, the scene NPC character model library contains a variety of virtual NPC character models. Each virtual NPC character model includes gender, personality traits (such as friendly, serious, and impatient), accent characteristics, knowledge background, and dialogue style.
[0065] In step S4, a scene NPC character initiates a guiding dialogue, introducing the scene, welcoming the user, and starting the simulated learning interaction. The NPC character guides the conversation, allowing the user to quickly immerse themselves in the simulated learning scenario. During the learning interaction, at appropriate times, scene NPC characters are inserted to explain and guide users on scene customs and taboos. When taboo words appear, scene NPC characters are inserted to provide explanations and corrections. Some scenes may randomly trigger special events, and scene NPC characters are inserted to guide users on handling unexpected situations. Multiple logical rules are used to trigger word and semantic explanations for dialogues that the user has questions about.
[0066] The evaluation model assesses user responses based on a comprehensive scoring formula:
[0067] Score=[(V*Wv)+(G*Wg)+(A*Wa)+(F*Wf)]*R / 10*D− T,
[0068] Wherein, Score represents the overall score, V represents vocabulary accuracy, G represents grammatical accuracy, A represents accuracy, F represents fluency, R represents reaction speed, D represents difficulty level, T represents deduction for forbidden words, Wv represents the weight of vocabulary accuracy V, Wg represents the weight of grammatical accuracy G, Wa represents the weight of accuracy A, Wf represents the weight of fluency F, and Wv + Wg + Wa + Wf = 1.
[0069] The vocabulary accuracy V is obtained using the following formula:
[0070] V = Percentage of accuracy in total vocabulary * 10;
[0071] The grammatical accuracy G is obtained by the following formula:
[0072] G = Total grammar usage accuracy percentage * 10;
[0073] The accuracy A is obtained by the following formula:
[0074] A = Total accuracy percentage * 10;
[0075] The fluency F is obtained by the following formula:
[0076] F = Total percentage of fluency in responses * 10;
[0077] The reaction rate R is obtained by the following formula:
[0078] R = Number of conversations * 3 / Total number of seconds of pause in user responses;
[0079] The difficulty coefficient D is obtained using the following formula:
[0080] D = (1 + Scene Difficulty / 10) * (1 + Number of NPCs / 10),
[0081] The scene difficulty is 1, 2, 3, ..., 10.
[0082] The penalty T for prohibited words is obtained using the following formula:
[0083] T = (Preset total score * 30%) * Number of times the forbidden word appears.
[0084] After the evaluation is completed, the historical learning records, the latest user responses, and the evaluation results are used as context to iteratively generate the next round of learning content for the NPC character until the scene task is completed.
[0085] This embodiment enhances the realism of the scene and the flexibility of the method by interacting with the user in real time and updating the scene NPC. Secondly, based on deep learning model data, it can dynamically adjust and respond to the learning scene state data, avoiding the technical defects of rigid response and insufficient efficiency of template-based methods, realizing personalized customization, teaching according to aptitude, and improving learning efficiency.
[0086] In summary, the AI-based language learning simulation scenario generation method proposed in this embodiment pre-defines various language learning simulation scenarios, initializes scenario data through an AI model, and further refines the scenarios by adding language, region, customs, and rules, constructing an interactive virtual scenario that includes background settings, roles, task objectives, and a dynamic dialogue tree. It uses speech recognition and natural language processing technologies to analyze the user's voice and text input in real time, driving the evolution of the scenario plot and providing context-sensitive error correction and feedback. This method can generate a highly personalized and immersive language learning environment, effectively improving learners' language proficiency and intercultural communication skills.
[0087] Example 2
[0088] Based on the same inventive concept as Embodiment 1, this embodiment of the invention discloses a method for generating language learning simulation scenarios based on artificial intelligence, specifically including the following steps:
[0089] S1 removes noise and incomplete records from the received natural speech and text content to ensure data quality and consistency; classifies the data into different categories based on content and language difficulty, adds metadata to the data, creates a user profile, and reads the user's profile information: "Native Chinese speaker, English learner at A2 level, mentions the keyword 'restaurant', and has recently learned food-related vocabulary and grammar."
[0090] S2, the system generates specific simulated learning scenarios based on the user's basic profile information. Using "ordering food at a restaurant" as the scenario template, it sends a request to the scenario theme library to generate a themed restaurant. This template includes basic information such as the scenario's background description, language selection, and regional cultural characteristics. Simultaneously, it requests the generation of a virtual NPC character that meets specific requirements from the character model library, such as a "patient restaurant waiter with a perfect London accent" character model. Upon completion of the requests, a complete and dynamic initial scenario is generated.
[0091] "You are in an elegant traditional restaurant in a small English town, and a waiter is smiling as he approaches you."
[0092] S3. Based on the initial scene description of the featured restaurant, select a virtual NPC character model of "patient restaurant waiter with a pure London accent" from the preset scene NPC character model library, and use artificial intelligence model to generate scene NPC character: "Waiter Pierre, about 40 years old, enthusiastic and professional".
[0093] S4, the core task objectives are: 1. Successfully order food; 2. Engage in at least 3 rounds of lighthearted conversation (such as discussing the weather, travel, and local impressions).
[0094] A guided dialogue is initiated by a scene NPC. A waiter welcomes you in fluent London English, guides you to your seat, presents the menu, and introduces the restaurant's specialties. You select a medium-rare steak and a salad. You successfully complete your order.
[0095] While waiting for the food, the evaluation results indicate that the interaction difficulty can be increased:
[0096] Added a new NPC character to the right-side scene to initiate casual conversation.
[0097] During casual conversation, NPCs will tell you about local customs and taboos, such as the nearest city to the town being London and the prohibition of smoking in public places.
[0098] During casual conversation, if a taboo word is unintentionally uttered, the NPC character sitting next to you may react to your behavior, thereby further enriching the realism and challenge of the scene.
[0099] Based on the evaluation results, it was determined that the interaction difficulty could be increased further:
[0100] Continue adding new NPCs to the chat.
[0101] Once the meal is served, the scenario enters its final stage. The system provides an overall evaluation of the user's performance throughout the scenario, including the accuracy of language use, fluency of dialogue, and task completion rate.
[0102] The system updates and improves user profiles, recording changes in users' language proficiency, frequently occurring error types, and learning preferences across different scenarios. This updated profile information will provide a more accurate reference for generating subsequent learning scenarios, thereby enabling continuous improvement of personalized learning.
[0103] Overall assessment: "The performance in this conversation was good. The task of ordering food was successfully completed, and communication with friends around the participants was also successful."
[0104] End of scene.
[0105] Example 3:
[0106] Based on the same inventive concept as Embodiment 1, this embodiment of the invention discloses a method for generating language learning simulation scenarios based on artificial intelligence, specifically including the following steps:
[0107] Based on user profiles, such as the user's professional background, target position, and expected interview scenario type, the system selects a standard difficulty interview scenario template from the scenario theme library and generates a neutral interviewer role from the role model library. After initialization, the user enters the interview training scenario, where the interviewer role greets the user professionally and begins the interview process.
[0108] In an interview scenario, the interviewer first prompts the user to introduce themselves, which is a crucial step in assessing the user's language skills. The system analyzes the user's input in real time using speech recognition and natural language processing technologies, and dynamically adjusts the difficulty and linguistic complexity of the interviewer's questions based on the user's performance.
[0109] For example, the user completed the self-introduction fluently, using the correct tenses and professional vocabulary.
[0110] The user completed the self-introduction fluently, using the correct tenses and professional vocabulary.
[0111] Analyze user input to determine if their current performance is above their set A2 level. Dynamically adjust subsequent dialogue:
[0112] NPC behavior: Instruct the interviewer role to ask more challenging questions, such as "Please describe an experience you had in resolving a conflict within a team."
[0113] Language complexity: Introduce more advanced grammatical structures (such as conditional and subjunctive moods) and more specialized industry terminology in the interviewer's questions and subsequent system feedback.
[0114] Conversely, if users frequently make grammatical errors or express themselves incoherently in their answers, the AI model will automatically reduce the difficulty, for example, by having the interviewer ask simpler questions or provide more hints.
[0115] After the interview, a detailed feedback report is generated based on the user's performance during the interview. The report includes an evaluation of the accuracy of language use, the logic of answers, and an overall assessment of performance.
[0116] This dynamic adjustment mechanism not only enhances the interactivity of the scenario but also ensures that users can gradually improve their language skills through challenges tailored to their individual levels. In this way, the system can provide users with optimal learning challenges that are "just within reach," thereby maximizing learning efficiency.
[0117] Example 4:
[0118] This embodiment provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the steps of the artificial intelligence-based language learning simulation scene generation method as described in any of the embodiments.
[0119] Example 5:
[0120] This embodiment provides a computer device, including:
[0121] Memory, used to store computer instructions;
[0122] A processor is configured to execute the computer instructions to implement the steps of the artificial intelligence-based language learning simulation scene generation method as described in any one of Embodiment 1.
[0123] Example 6:
[0124] This embodiment provides a computer program product, including computer instructions, characterized in that, when the computer instructions are executed by a processor, they implement the steps of the artificial intelligence-based language learning simulation scene generation method as described in any one of Embodiment 1.
[0125] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0126] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0127] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0128] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0129] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.
Claims
1. A method for generating language learning simulation scenarios based on artificial intelligence, characterized in that, include: The acquired user data is preprocessed to obtain user profiles; Based on a pre-defined scenario theme library and the user profile, an artificial intelligence model is used to generate scenario descriptions; Based on a pre-set library of scene NPC character models and the scene description, virtual NPC characters are generated using an artificial intelligence model; Virtual NPC characters interact with users in multiple rounds until the scene learning task is completed; In this process, after each round of interaction, the user's response during the interaction is evaluated using an evaluation model to obtain the evaluation result; Update the user profile based on the evaluation results and interaction content, and adjust the scene description and virtual NPC characters according to the updated user profile before proceeding to the next round of interaction.
2. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 1, characterized in that, The step of preprocessing the acquired user data to obtain a user profile includes: The acquired user data is cleaned to remove noisy data and incomplete records, resulting in cleaned data. Based on the content of the cleaned data, user profiles are obtained by categorizing and annotating metadata.
3. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 1, characterized in that, The user profile includes the target language to be learned, learning objectives, and current language proficiency.
4. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 3, characterized in that, The user profile also includes user personality and historical learning behavior data.
5. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 1, characterized in that, The scene NPC character model library includes multiple scene theme basic templates. Each scene theme basic template includes the scene's background description, language, region, customs and taboos, key scene NPC characters, scene rules, learning tasks, and key topics.
6. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 1, characterized in that, The scene NPC character model library contains a variety of virtual NPC character models, each of which includes gender, personality traits, accent characteristics, knowledge background, and dialogue style.
7. The method for generating simulated language learning scenarios based on artificial intelligence according to claim 1, characterized in that, The evaluation model assesses user responses based on a comprehensive scoring formula; The comprehensive scoring formula is as follows: Score=[(V*Wv)+(G*Wg)+(A*Wa)+(F*Wf)]*R / 10*D− T, Wherein, Score represents the overall score, V represents vocabulary accuracy, G represents grammatical accuracy, A represents accuracy, F represents fluency, R represents reaction speed, D represents difficulty level, T represents deduction for forbidden words, Wv represents the weight of vocabulary accuracy V, Wg represents the weight of grammatical accuracy G, Wa represents the weight of accuracy A, Wf represents the weight of fluency F, and Wv + Wg + Wa + Wf = 1. The vocabulary accuracy V is obtained using the following formula: V = Percentage of accuracy in total vocabulary * 10; The grammatical accuracy G is obtained by the following formula: G = Total grammar usage accuracy percentage * 10; The accuracy A is obtained by the following formula: A = Total accuracy percentage * 10; The fluency F is obtained by the following formula: F = Total percentage of fluency in responses * 10; The reaction rate R is obtained by the following formula: R = Number of conversations * 3 / Total number of seconds of pause in user responses; The difficulty coefficient D is obtained using the following formula: D = (1 + Scene Difficulty / 10) * (1 + Number of NPCs / 10), The scene difficulty is 1, 2, 3, ..., 10. The penalty T for prohibited words is obtained using the following formula: T = (Preset total score * 30%) * Number of times the forbidden word appears.
8. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instruction is executed by the processor, it implements the steps of the artificial intelligence-based language learning simulation scene generation method as described in any one of claims 1-7.
9. A computer device, characterized in that, include: Memory, used to store computer instructions; A processor for executing the computer instructions to implement the steps of the artificial intelligence-based language learning simulation scene generation method as described in any one of claims 1-7.
10. A computer program product comprising computer instructions, characterized in that, When executed by a processor, the computer instructions implement the steps of the method for generating language learning simulation scenarios based on artificial intelligence, as described in any one of claims 1-7.