AI-based college book children English applet and production process

Abstract

The invention discloses an AI (artificial intelligence)-based college book child English applet and a production process thereof, relates to the technical field of education, and particularly relates to an AI-based college book child English applet and a production process thereof. Four functional modules of basic English learning (such as textbook synchronous word retrieval and pronunciation training), professional English learning (such as term library query and specification image recognition and interpretation), auxiliary learning (personalized recommendation) and social learning (learning report sharing, discussion area and group collaboration) are integrated. The manufacturing process sequentially comprises six stages of demand analysis and planning, system architecture design, core module development, technology integration and debugging, system test and optimization, and deployment online and iterative maintenance. The system adopts a cross-platform front-end and micro-service back-end layered architecture, the quality is guaranteed through multi-dimensional testing, stable deployment and continuous iteration are realized through a gray release strategy, and the multi-learning-section personalized English learning requirement is met.

Description

Technical Field

[0001] This invention relates to the field of educational technology, specifically to an AI-based English learning mini-program and its development process. Background Technology

[0002] With the acceleration of digitalization in education, the demand for English learning tools in basic and vocational education is growing. However, current English learning products on the market still have significant shortcomings in meeting the diverse and in-depth needs of users.

[0003] First, regarding the coverage of educational stages and the connection between content, most products have relatively simple functional designs, making it difficult to simultaneously meet the learning requirements of users at different educational stages. There is a lack of effective integration between the learning content of basic education stages (such as primary, junior high, and senior high school) and the professional English knowledge required for vocational education stages (such as secondary vocational schools and higher vocational colleges), resulting in fragmented learning resources and difficulty in forming a coherent knowledge system. In particular, existing tools often offer limited support for terminology learning and application training in fields such as mechanical, electrical, and computer science.

[0004] Secondly, in terms of product functionality, many learning tools only offer basic functions such as vocabulary lookup and example sentence display, lacking capabilities in providing detailed guidance for pronunciation training and analyzing complex professional texts. Users struggle to obtain effective feedback when practicing reading aloud, and when faced with materials such as professional English manuals, they lack auxiliary means for terminology recognition and sentence structure analysis, thus limiting the improvement of learning efficiency and application ability.

[0005] Furthermore, in terms of learning interaction and knowledge sharing, the social functions of existing products are usually quite simple, mostly limited to superficial interactions such as learning check-ins and grade sharing, failing to build a deep communication environment centered around the learning content itself. Users find it difficult to have effective discussions on specific professional issues, difficulties encountered during the learning process cannot be resolved in a timely manner, and high-quality learning resources and insights are difficult to accumulate and share.

[0006] Furthermore, in terms of product development methodologies, many projects lack systematic and standardized build processes, which can easily lead to problems such as unreasonable feature planning, poor system scalability, and difficulties in later maintenance. This not only affects product stability and user experience but also increases the difficulty of feature iteration and cross-domain adaptation. Summary of the Invention

[0007] The purpose of this invention is to provide an AI-based English learning mini-program and its development process. Through a systematic six-stage development process, a comprehensive platform integrating basic English learning, professional English learning, auxiliary learning, and social learning functions is built to meet the differentiated and personalized learning needs of users at multiple educational levels from primary school to higher vocational college in different scenarios, while ensuring that the system has good maintainability, scalability, and user experience.

[0008] To achieve the above objectives, this invention employs the following technical solution: an AI-based Xuehai Shutong English mini-program and its development process. The core of this process lies in its systematic and standardized phase division, ensuring full lifecycle management from concept to launch. Phase One emphasizes extensive and in-depth field research to accurately capture the real learning pain points of different user groups, from basic education to vocational education, and transforms these needs into clear and measurable functional indicators and performance standards, laying a solid foundation for subsequent development. Phase Two focuses on building a flexible, robust, and easily scalable technical framework, ensuring efficient collaboration among components through clear architectural design and interface specifications. Phase Three adopts a modular parallel development strategy, significantly improving development efficiency and reducing coupling between modules. Phases Four through Six focus on integration, verification, and delivery, ensuring the quality, stability, and long-term viability of the final product through rigorous testing, controlled releases, and continuous iteration. The entire process is interconnected, forming a highly efficient methodology that can be reused in the development of similar educational products.

[0009] Furthermore, in the needs analysis phase, the research was characterized by its comprehensiveness and relevance. The scope of the research not only spanned primary, middle, and high school education but also extended to secondary and higher vocational education, ensuring that the product could meet the diverse needs of users of different ages and with different learning goals. For primary education, the research focused on textbook synchronization, vocabulary memorization efficiency, and the fun and accuracy of pronunciation learning. For vocational education, the research emphasized exploring the practical difficulties users face in reading and understanding English technical documents and mastering professional terminology in specific professional fields (such as mechanical and electrical engineering). In addition, a detailed survey was conducted on the common needs across all education levels, such as maintaining learning motivation, knowledge sharing, and Q&A, providing direct evidence for building an immersive learning community.

[0010] Furthermore, the system architecture design adopts a clear layered and decoupled approach. The key consideration in selecting the front-end layer is cross-platform adaptability, ensuring users receive a consistent and smooth operating experience across different mini-program platforms. The back-end service layer abandons the traditional monolithic architecture, adopting a microservice design. Core functions such as user management, learning content management, and social interaction management are broken down into independently deployable and scalable service units, which greatly improves the system's maintainability and ability to handle high-concurrency access. The data storage layer is designed differently based on data characteristics. It utilizes relational databases to process structured business data, improves the read speed of frequently accessed data through caching databases, and efficiently manages unstructured resources such as user-generated images and documents using object storage services, collectively forming a stable and efficient data support system.

[0011] Furthermore, the basic English learning module emphasizes both the authority of its content and the practicality of its learning functions. The vocabulary database is strictly synchronized with mainstream domestic textbooks, ensuring that the content matches the classroom learning progress. Each word entry is meticulously compiled, including not only standard phonetic symbols but also syllable-by-syllable pronunciation guidance, supplemented by typical example sentences from textbooks to help learners master vocabulary usage in specific contexts. Functionally, the module features an interactive pronunciation learning section where learners can follow along with standard pronunciation. The system analyzes the audio and provides specific, actionable suggestions for pronunciation improvement, such as indicating the direction of mouth shape or tongue position adjustments, thereby effectively improving learners' spoken English accuracy.

[0012] Furthermore, the core value of the professional English learning module lies in its deep integration of language learning with professional scenarios. The establishment of a terminology database is fundamental, categorized according to mainstream professional fields such as mechanical manufacturing, electrical engineering, and computer technology. Each term not only provides an accurate Chinese definition but also details its applicable work scenarios and common collocations, enabling learners to "apply what they learn." The instruction manual interpretation function is an innovative tool that allows users to quickly obtain the text content of printed or electronic English technical documents by taking photos or uploading images with their mobile phones. The system then calls upon the built-in professional terminology database to automatically identify and highlight key terms in the text, providing a fluent translation of the entire passage. Simultaneously, it analyzes complex sentence structures to help learners overcome the comprehension barriers of long and difficult sentences.

[0013] Furthermore, the personalized recommendation function in the supplementary learning module aims to improve the targeting and efficiency of learning. This function doesn't randomly push content; instead, it's based on in-depth analysis of user learning behavior. The system continuously monitors data such as users' word lookup records, common error points in practice, and time spent on different learning content. Based on these behavioral patterns, the system intelligently associates and recommends relevant learning resources. For example, it recommends excerpts from instruction manuals containing a specific term to users who frequently look up that term, or recommends explanatory materials to learners struggling with a particular grammar point. This data-driven recommendation mechanism helps achieve "personalized instruction," planning a unique learning path for each learner.

[0014] Furthermore, the social learning module aims to build a positive and mutually supportive learning community. The learning report function visualizes users' learning data, generating clear and concise progress charts to motivate continuous learning and facilitate sharing of results with peers. The discussion forum is not disorganized but meticulously divided by academic level and major, ensuring users can ask and answer questions in the most relevant communities, improving the efficiency and quality of communication. Valuable questions and answers are filtered and preserved by the system, forming a searchable community knowledge base that benefits more learners. The learning group function gives users greater autonomy, allowing them to create groups around specific goals (such as exam preparation or project research), share learning materials, and initiate thematic discussions, effectively promoting collaborative learning.

[0015] Furthermore, the system testing and optimization phase is a crucial step in ensuring product quality, employing a multi-dimensional, high-standard testing strategy. Functional testing strives for comprehensive coverage, designing over a thousand test cases to verify the correctness and stability of every functionality, from word lookup to social interaction. Discovered issues are rigorously tracked and managed to ensure serious defects are thoroughly fixed. Performance testing simulates user access pressure from light to heavy, examining the system's response speed and service stability under varying concurrency loads to ensure a smooth experience for a large number of users in actual operation. Security testing focuses on user privacy and data protection, verifying the system's capabilities in data encryption storage, secure transmission, resistance to malicious network attacks, and strict access control, building a robust security defense.

[0016] Furthermore, a cautious and controlled canary release strategy is adopted during the deployment and launch phase to minimize the risks associated with launching a new version. The release process is divided into three distinct phases: first, new features are rolled out to a small group of representative users, during which the system's operational status and user feedback are closely monitored; after confirming no major anomalies, the user coverage is gradually expanded to about one-third for broader verification; finally, only after ensuring system stability and reliability is the version fully rolled out to all users. A clear risk response mechanism is in place throughout this process. If an abnormally high system error rate is detected or a certain number of negative user feedback are received, a rollback process will be immediately initiated to quickly restore the previous stable version, thereby ensuring a normal user experience for the majority of users.

[0017] Furthermore, this English learning mini-program, as an organic whole, is centered on the synergy and data integration of its three main functional units. The basic English learning unit lays a solid foundation in language for learners, the professional English learning unit focuses on enhancing professional skills, and the social learning unit creates an atmosphere and motivation for continuous learning. These three units do not exist in isolation but are closely connected through shared user data and learning behavior records. For example, new words looked up in the professional unit can be included in the basic unit for review, and questions discussed in the social unit can deepen understanding of professional content. This design truly achieves a seamless transition from general English to professional English, meeting a learner's full-cycle, multi-dimensional learning needs from the initial stage to professional skills enhancement.

[0018] This invention provides an AI-based English learning mini-program and its production process, which has the following beneficial effects: 1. Achieving a smooth transition and professional expansion of the English learning system across all educational stages: This invention is the first to construct a complete English learning system covering primary school education to higher vocational education, effectively solving the problem of gaps in learning content between different educational stages. Especially for the vocational education stage, it innovatively integrates English learning content from mainstream majors such as mechanical, electrical, and computer science, providing functions for looking up professional terminology and interpreting English instruction manuals in real-world scenarios. This allows students' English learning to be closely integrated with their professional skills development, achieving a natural transition and deepening from general language proficiency to professional application skills, filling the gap in specialized English learning tools in the vocational education field.

[0019] A highly standardized product development process has been established, significantly improving development efficiency and product quality: The six-stage production process established in this invention—"requirements analysis - architecture design - module development - integration and debugging - testing and optimization - deployment and iteration"—has clearly defined stage tasks, cycle definitions, and quantifiable acceptance standards. This standardized process ensures the controllability and predictability of the product development process, effectively avoiding common problems such as functional redundancy and performance instability. Practice has shown that adopting this process can significantly shorten the development cycle of adding new professional functional modules, while the stability and error rate control after system launch are better than the industry average, providing an efficient and reliable process template for subsequent functional iterations and the development of similar educational products.

[0020] Deeply integrating social interaction mechanisms, this invention effectively stimulates learning enthusiasm and knowledge sharing: It organically integrates learning and social functions, forming a virtuous cycle of "learning-sharing-discussion-reflection." The system automatically generates visual learning reports for users to share and establishes discussion areas categorized by grade level and major, facilitating targeted communication. Furthermore, the learning group function supports resource sharing and online discussions. This design not only enhances the fun and interactivity of learning but also promotes the explicit expression of tacit knowledge and the accumulation of collective wisdom, thus significantly improving users' learning time and problem-solving efficiency.

[0021] Through modular and microservice architecture design, the system's excellent scalability and maintainability are ensured: The system architecture of this invention adopts the front-end and back-end separation and microservice design concept, breaking down core functions into independent service units. This architecture reduces the coupling between various modules of the system. When new learning content needs to be added (such as adding a new finance major) or a certain function needs to be updated, specific services can be developed, tested, and deployed independently without large-scale modifications to the entire system, greatly enhancing the system's flexibility and scalability. At the same time, this also facilitates daily operation and maintenance management and fault isolation, ensuring the long-term stable operation of the system.

[0022] A multi-dimensional, rigorous testing and continuous optimization mechanism has been established to ensure a superior end-user experience: Before the product's official launch, comprehensive testing covering functionality, performance, user experience, and security was implemented. Through verification with over a thousand test cases, high-concurrency stress testing, real-world scenario experiences with hundreds of target users, and professional security penetration testing, potential issues can be identified and fixed promptly. After launch, a long-term user feedback collection and quarterly iteration mechanism has been established to ensure the product continuously responds to changing user needs and is optimized accordingly. This rigorous quality assurance process is a crucial foundation for the product's high user satisfaction and low error rate. Attached Figure Description

[0023] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0024] Figure 1 This is an overview diagram of the overall manufacturing process of the system of the present invention; Figure 2 This is a detailed flowchart of Phase One and Phase Two of the present invention; Figure 3 This is a flowchart illustrating the development process of the three core modules at each stage of this invention. Figure 4 This is a flowchart illustrating the workflow and test deployment for phases four through six of this invention. Figure 5 This is a diagram showing the system composition and data association of the present invention. Detailed Implementation

[0025] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses consistent with some aspects of this disclosure as detailed in the appended claims.

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0027] Example 1: Specific Implementation of the Requirements Analysis and Planning Phase In the requirements analysis and planning phase, this embodiment details how to quantify user needs through a systematic research method and output a standardized requirements specification. First, a stratified survey was conducted for users across multiple educational levels: For the K-12 level (primary to high school), the survey covered six provinces nationwide (including North China, East China, and South China), collecting at least 500 valid samples through questionnaires (no fewer than 150 for each level), combined with interviews with teachers from 30 public and private schools, focusing on confirming needs such as textbook-synchronized word retrieval and phonics correction (e.g., word syllable breakdown and comparison with recordings). The survey found that 85% of users expected real-time pronunciation correction, and the vocabulary database needed to be entered unit-by-unit according to mainstream textbook versions such as People's Education Press and Foreign Language Teaching and Research Press. For vocational high school students (mechanical, electrical, and computer science majors), eight vocational schools were selected, collecting at least 200 professional English manuals (e.g., machine tool operation manuals, software development documents). Questionnaires and interviews clarified the needs for professional terminology retrieval, manual paragraph translation, and grammar analysis, with 92% of users requiring support for image upload and recognition. The social needs across all educational levels were validated through online focus groups (15 people per group). 78% of users hoped that the Q&A in the discussion forum could be automatically compiled into a knowledge base. Based on the survey results, the MoSCoW principle was used to define functional priorities: P0 level (must be implemented) includes basic word search and a professional terminology database; P1 level (should be implemented) includes pronunciation correction and a discussion forum. Performance indicators were quantified as follows: pronunciation recognition accuracy no less than 98%, professional translation accuracy no less than 92%, system page loading time no more than 3 seconds, interface response time no more than 1 second, and no crashes when handling 1000 concurrent users. The final output is a requirements specification document conforming to the GB / T 9385-2008 standard, serving as the basis for subsequent development. This embodiment, through rigorous research and quantification standards, ensured that the product accurately matched user needs, laying the foundation for subsequent architecture design and module development.

[0028] Example 2: Specific Implementation in the System Architecture Design Phase In the system architecture design phase, this embodiment focuses on describing the selection of layered technologies and the definition of interface specifications to ensure the system's compatibility, scalability, and stability. The architecture design cycle was 15 working days, adopting a layered approach: the front-end layer uses Vue.js 3.0 combined with the UniApp framework to achieve multi-platform adaptation such as WeChat Mini Programs and Alipay Mini Programs; the UI component library uses uView UI, with differentiated visual styles designed for different educational stages (e.g., cartoon elements for primary school modules and a simple industrial style for professional modules). The back-end layer is based on Spring Boot 2.7 and Spring Cloud Alibaba microservice architecture, divided into 5 independent services: user service (responsible for account management and access control), learning service (including basic English and professional English sub-modules), social service (managing discussion forums and study groups), data service (handling storage and backup), and an independent auxiliary engine service (for integrating translation and recommendation functions). The data storage layer uses a MySQL 8.0 relational database to store user information, vocabulary, and terminology databases, and is configured with a master-slave replication architecture to improve query efficiency; Redis 6.2 is used as a cache database to handle high-frequency requests; file storage uses Alibaba Cloud OSS, supporting image compression to save bandwidth. The architecture documentation includes a system topology diagram (drawn using Visio 2023), clearly illustrating the network interactions between the front-end, back-end services, and data storage. It also defines specifications for over 20 core API interfaces, such as a word search interface (request parameters include grade level and keywords; response parameters include words, phonetic symbols, and example sentences; response time ≤ 1 second) and a manual interpretation interface (request parameters include image URLs and professional fields; response time ≤ 3 seconds). This embodiment, through standardized architecture design, achieves decoupling and efficient collaboration between modules, providing technical support for subsequent parallel development.

[0029] Example 3: Specific Implementation of the Development of the Basic English Learning Module In the development phase of the basic English learning module, this example uses the K-12 stage to detail the construction of a textbook-synchronized vocabulary database and the implementation of phonics functionality. The development cycle was 45 working days. First, a vocabulary database synchronized with mainstream textbook versions was entered: over 1200 words from the People's Education Press (PEP) textbook for grades 1-6, over 1800 words from the Foreign Language Teaching and Research Press (FLTRP) textbook for grades 7-9, and over 3500 words from the new curriculum standard textbook for grades 1-3. Each word includes International Phonetic Alphabet (IPA) symbols (British / American versions), phonics syllable breakdown (e.g., "apple" broken down into "ap-ple"), textbook example sentences (with source annotations such as "PEP Grade 6, Unit 3"), and scene images (e.g., "apple" paired with a fruit image). The vocabulary data was reviewed by three English teachers to ensure an error rate not exceeding 0.5%. The phonics function development includes speech synthesis and pronunciation correction: speech synthesis integrates a third-party speech engine, supporting switching between British and American pronunciation for words and example sentences, with a pronunciation clarity of no less than 95%; the pronunciation correction function allows users to record audio, and the system compares it with standard audio by extracting pronunciation features, outputting visual correction suggestions (such as prompts for tongue position adjustment), with a correction accuracy of no less than 98%. The search function uses the Elasticsearch 8.0 search engine, supporting multi-dimensional searches by textbook unit, part of speech, difficulty level, and scenario, with a response time controlled within 1 second, and automatically recommending similar words when no results are found. This embodiment improves the accuracy and user experience of basic English learning through structured data entry and function optimization, meeting the personalized needs of K-12 students.

[0030] Example 4: Specific Implementation of the Development of the Professional English Learning Module In the development phase of the professional English learning module, this embodiment uses mechanical engineering as an example to illustrate the construction of the terminology database and the implementation of the instruction manual interpretation function. The development cycle runs parallel to other modules. First, a mechanical engineering terminology database is built: referencing the *Standard for English Terminology in Mechanical Engineering* and real enterprise instruction manuals, 1200 terms such as "lathe" and "cutting tool" are entered. Each term is annotated with its Chinese definition, application scenario (e.g., "lathe operation"), collocation phrase (e.g., "operate thelathe"), and professional image. The terminology database is reviewed by two mechanical engineering teachers and one English teacher to ensure zero errors. The instruction manual interpretation function is implemented in steps: First, an OCR text recognition component is integrated, supporting users to upload JPG / PNG format instruction manual images, with a recognition accuracy of no less than 99%; second, the mechanical engineering terminology database is called to highlight the terms in the recognized text; third, a translation engine outputs the translation, and long and complex sentences (e.g., "Before starting the machine, check the lubricating oil level") are grammatically analyzed, and sentence components and tenses are marked. During the testing phase, five untrained instruction manuals (such as milling machine operation manuals) were selected to verify that the accuracy rate of terminology translation reached 94% and the correctness rate of analyzing long and difficult sentences reached 90%. This embodiment effectively improves the professional English application ability of secondary and higher vocational students through professional terminology management and functional design.

[0031] Example 5: Specific Implementation of Social Learning Module Development This embodiment details the development process of the learning report, discussion forum, and learning group functions during the social learning module development phase. The development cycle was 45 working days. The learning report function automatically generates daily / weekly / monthly reports, including vocabulary mastery rate, weakness analysis, and learning time. It supports generating image-based reports for sharing to WeChat or QQ, with a sharing success rate of no less than 99%. The discussion forum function is divided by grade level and major (e.g., "Vocational College - Mechanical Engineering Discussion Forum"). Users can post questions (e.g., "What is the full name of PLC?"), and the system automatically extracts high-quality questions and answers through keyword extraction and adds them to the knowledge base for easy retrieval. The learning group function allows users to create or join groups (e.g., "Grade 11 English Intensive Group"). Groups support resource sharing (single files not exceeding 200MB) and online discussions (supporting real-time messaging). During the testing phase, 300 target users were invited to conduct on-site testing, optimizing the interface (e.g., enlarging the terminology search button to 24px) and simplifying the process. This embodiment, through the closed-loop design of social functions, promotes mutual learning among users, improving learning motivation and knowledge accumulation efficiency.

[0032] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An AI-based English learning mini-program and its production process, characterized by: The process consists of the following six stages: Phase 1: Needs Analysis and Planning. Through user surveys and teacher interviews across multiple educational levels and regions, learning needs are quantified, and a requirements specification document containing functional priorities and performance indicators is output. Phase Two: System Architecture Design, selecting technologies based on cross-platform front-end frameworks and microservice back-end architecture, drawing system topology diagrams and defining core interface specifications; Phase Three: Core Module Development, with parallel development of four functional modules: basic English learning, professional English learning, auxiliary learning, and social learning; Phase Four: Technical Integration and Debugging, integrating the core functional modules with the auxiliary engine, and conducting module-by-module debugging and joint debugging; Phase 5: System Testing and Optimization. Through multi-dimensional testing of functionality, performance, user experience, and security, the system will be repaired for defects and the user experience will be optimized. Phase Six: Deployment, Launch, and Iterative Maintenance. A phased canary release strategy will be adopted to deploy the mini-program to various platforms, and a long-term product iteration and architecture upgrade mechanism will be established.

2. The AI-based English learning program and production process of claim 1, wherein: In Phase One, the user needs survey covers multiple educational stages from primary school to higher vocational education. Specifically, it includes: the need for textbook-synchronized vocabulary retrieval and pronunciation learning for basic education stages; the need for interpretation of professional English manuals and retrieval of professional terminology for vocational education stages; and the need for sharing learning reports, discussing problems, and collaborative social interaction in groups for all educational stages.

3. The AI-based English learning program and production process of claim 1, wherein: In Phase Two, the system architecture adopts a layered design, including: The front-end layer uses a framework that supports cross-platform adaptation to implement the user interface; The backend service layer adopts a microservice architecture, which is divided into independent services such as user management, learning content management, social interaction management, and data management; The data storage layer combines relational databases and cache databases, and utilizes object storage services to manage file resources.

4. The AI-based English learning program and production process of claim 1, wherein: In Phase Three, the construction of the basic English learning module includes: inputting a vocabulary database synchronized with mainstream textbook versions, with each word containing phonetic symbols, syllable breakdowns, and example sentences; and developing pronunciation learning functions that support follow-up reading and feedback.

5. The AI-based English learning program and production process of claim 1, wherein: In Phase Three, the construction of the professional English learning module includes: establishing a terminology database categorized by majors such as mechanical, electrical, and computer science, with each term annotated with its industry definition and application scenarios; and developing a manual interpretation function, which extracts text through image recognition, calls the professional terminology database for terminology annotation, and outputs translations and sentence structure analysis.

6. The AI-based English learning program and production process of claim 1, wherein: In Phase Three, the learning assistance module includes a personalized recommendation function, which recommends related words, learning materials, or instruction manual excerpts to users based on their learning behavior data.

7. The AI-based Xuehai Shutong English mini-program and its production process according to claim 1, characterized in that: In Phase Three, the social learning module includes: the ability to automatically generate learning reports and support sharing; discussion forums divided by academic level and major, which can accumulate high-quality Q&A into a knowledge base; and learning groups that allow users to create or join and share materials and conduct online discussions.

8. The AI-based English learning program and production process of claim 1, wherein: The system testing in Phase 5 includes: Functional testing involved designing over a thousand test cases to cover all core functions and tracking and fixing defects. Performance testing simulates different levels of concurrent user access to ensure that system response time and resource utilization meet preset standards; Security testing verifies the effectiveness of data encryption, secure transmission, interface protection, and access control.

9. The AI-based English learning program and production process of claim 1, wherein: The deployment and launch in Phase 6 adopts a canary release strategy, gradually expanding the user access scope in three phases, and has a rollback mechanism. When the system error rate or user complaints exceed the threshold, it will immediately revert to the previous stable version.

10. The AI-based English learning program and production process according to any one of claims 1 to 9, characterized in that: The system includes: The basic English learning unit is designed to provide vocabulary learning and pronunciation training that is synchronized with the textbook. The professional English learning unit provides access to professional terminology and interpretation of professional manuals; Social learning units are designed to support learning sharing, content discussion, and group collaboration; The system is adapted to the learning needs of multiple educational stages from primary school to higher vocational education, and the data and services of each unit are interconnected.

Images