Film and television teaching intelligent interaction method and system based on large model

Abstract

The invention provides a film and television teaching intelligent interaction method and system based on a large model, and relates to the technical field of film and television teaching. According to the method, multiple functions including teaching file management, play setting and lecture export controls are integrated into a main interface, so that efficient management and utilization of film and television teaching files in various formats are realized. According to the method, the video content can be automatically analyzed by using the large model, and the subject labels and the knowledge point labels are automatically generated, so that the subject labels and the knowledge point labels are associated with the big data knowledge base and the exercise base to form extended knowledge and exercise indexes, and organization and expansion of educational resources are facilitated. In addition, personalized teaching interaction trigger points can be set according to the indexes, and interactivity and pertinence in the teaching process are enhanced. Finally, the teacher can automatically generate and export teaching notes according to all the information, so that the teaching note making efficiency and the teaching quality are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of film and television teaching technology, and in particular to a large model-based intelligent interaction method and system for film and television teaching. Background Art

[0002] With the development of information technology and advancements in educational concepts, traditional classroom teaching models are gradually shifting towards digitalization and intelligentization. In particular, in the field of film and television education, teaching through video resources has become a common method. However, how to effectively utilize these rich video resources and combine them with interactive teaching to improve teaching effectiveness and student engagement has become an important research direction. Existing interactive models fail to fully utilize the in-depth information in video content for knowledge expansion, instant questions, or exercises, limiting students' learning interest and effectiveness. Summary of the Invention

[0003] This application provides a large-scale model-based intelligent interactive method and system for film and television teaching to solve one or more technical problems existing in the prior art and at least provide a beneficial option or create conditions.

[0004] On the one hand, the present application provides a large-scale model-based intelligent interactive method for film and television teaching, comprising the following steps: Displaying a main interface for intelligent interaction of film and television teaching based on a large model; wherein the main interface includes a teaching file management control, a playback setting control, and a teaching handout export control; In response to a trigger instruction on the teaching file management control, a teaching file management sub-interface is displayed, and film and television teaching files in various formats are uploaded. The video images, voice signals, and subtitle texts of the film and television teaching files are automatically parsed using the large model. Subject labels and knowledge point labels are automatically generated based on the parsed content of the film and television teaching files, and the labels are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index. In response to a trigger instruction on the play setting control, a play setting sub-interface is displayed, and a teaching interaction trigger point is set according to the extended knowledge index and the extended exercise index; In response to the trigger instruction of the teaching handout export control, the teaching handout export sub-interface is displayed, and the teaching handout is automatically generated and exported based on the parsed content of the film and television teaching file, the extended knowledge index, the extended exercise index and the information of the teaching interaction trigger point.

[0005] Furthermore, the teaching file management sub-interface includes a file upload control, a one-key parsing control, a label generation control, and a teaching association control; The display teaching file management sub-interface is used to upload film and television teaching files in various formats, and the large model is used to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files. Subject labels and knowledge point labels are automatically generated based on the parsed content of the film and television teaching files, and the labels are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index, including the following steps: In response to a trigger instruction for the file upload control, a multi-format file selection window is displayed to receive the video teaching file uploaded by the user; In response to a trigger instruction of the one-key parsing control, a file parsing window is displayed, the large model is called to automatically parse the video image, voice signal, and subtitle text of the film and television teaching file, and the parsed content is generated and displayed; In response to a trigger instruction for the label generation control, a label generation window is displayed, and based on the parsed content, the large model is called to generate the subject label and the knowledge point label; In response to the trigger instruction of the teaching association control, the teaching association window is displayed, and based on the subject label and the knowledge point label, it is associated with the big data knowledge base and the big data exercise library to generate the extended knowledge index and the extended exercise index.

[0006] Furthermore, the subject label and the knowledge point label are displayed in the label generation window, and the user is supported to manually adjust the subject label and the knowledge point label, and the adjusted subject label and the knowledge point label are associated with the parsed content and stored.

[0007] Furthermore, the teaching association window includes a label display area, a knowledge base matching control, an exercise bank matching control, and an index generation control; The displaying of the teaching association window, based on the subject label and the knowledge point label, associating them with the big data knowledge base and the big data exercise database to generate the extended knowledge index and the extended exercise index, includes the following steps: In the label display area, select the required film and television teaching file, and display its corresponding subject label and knowledge point label; In response to a trigger instruction of the knowledge base matching control, the associated extended knowledge is retrieved from the big data knowledge base based on the subject label and the knowledge point label, and a mapping relationship between the extended knowledge and the time axis or content location of the film and television teaching file is established as a first mapping relationship; In response to a trigger instruction for the exercise library matching control, based on the knowledge point tag, extended exercises containing the same or derived knowledge points are retrieved from the big data exercise library and associated with exercise attributes, and a mapping relationship between the extended exercises and the timeline or content location of the film and television teaching file is established as a second mapping relationship; the exercise attributes include the subject tag, the knowledge point tag, the difficulty tag, and the usage scenario tag; In response to a trigger instruction for the index generation control, the subject label, the knowledge point label, the extended knowledge, the extended exercises and their corresponding exercise attributes, the first mapping relationship and the second mapping relationship are integrated to generate the extended knowledge index and the extended exercise index.

[0008] Furthermore, the calling of the large model to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files, and generating and displaying the parsed content, includes the following steps: Sampling key frames of the video images and extracting visual features through the visual processing module of the large model; Performing endpoint detection and noise reduction processing on the speech signal, and extracting the speech transcription text through the speech processing module of the large model; Perform word segmentation and entity recognition on the subtitle text, combine it with the speech transcription text, and use the text processing module of the large model to extract the semantic features of the speech text; Through the cross-modal fusion module of the large model, the visual features and the speech text semantic features are fused to generate and display structured parsed content, including keywords associated with timestamps and their knowledge point descriptions, key picture annotations and speech summaries.

[0009] Furthermore, the step of calling the large model to generate the subject label and the knowledge point label based on the parsed content includes the following steps: Based on the analyzed content, outputting the subject label through the subject classification module of the large model; Through the educational knowledge graph module of the large model, the keywords in the parsed content are semantically matched with the preset educational knowledge graph to generate preliminary labels; The clustering and filtering module of the large model is used to perform semantic clustering and redundant filtering on the preliminary labels, retaining preliminary labels with a confidence level higher than 0.85 as the knowledge point labels, and supplementing the hierarchical relationship between the knowledge point labels.

[0010] Furthermore, the playback setting sub-interface includes an interactive content marking control, an interactive configuration selection control, and an index association control; The display playback setting sub-interface sets the teaching interaction trigger point according to the extended knowledge index and the extended exercise index, including the following steps: In response to a trigger instruction for the interactive content marking control, an interactive content marking window is displayed, and based on the extended knowledge index and the extended exercise index, a mapping relationship between the extended exercises and the extended knowledge and the time axis or content location of the film and television teaching file is extracted, and the teaching interaction trigger point is automatically marked on the time axis of the film and television teaching file or the corresponding content location; Wherein, the teaching interaction trigger points include the time axis mark points and content positioning suspension mark points of the film and television teaching file; In response to a trigger instruction of the interaction configuration selection control, an interaction configuration window is displayed to receive the interaction type and trigger mode of the teaching interaction trigger point selected by the user; The interaction types include knowledge point expansion, instant questioning, and exercise practice; the triggering modes include automatic triggering and manual triggering; the automatic triggering refers to automatic triggering when the video teaching file is played to the timeline mark point; the manual triggering refers to triggering when the user manually clicks the content positioning floating mark point; In response to a trigger instruction for the index association control, the extended exercises and the extended knowledge are bound to the corresponding teaching interaction trigger points according to the extended knowledge index and the extended exercise index.

[0011] Furthermore, the main interface also includes a multi-screen split-screen control; In response to the trigger instruction of the multi-screen split-screen control, the multi-screen split-screen sub-interface is displayed, and the screen and whiteboard of the specified film and television teaching file are displayed synchronously. The user can drag the dividing line to adjust the split-screen ratio, and cross-screen interaction is supported.

[0012] Furthermore, the teaching handout export sub-interface includes an export format selection control, an export content screening control, an export layout setting control, an export path selection control, and a batch export control; The display teaching handout export sub-interface automatically generates and exports teaching handouts based on the parsed content of the film and television teaching file, the extended knowledge index, the extended exercise index, and the teaching interaction trigger point information, including the following steps: In response to a trigger instruction on the export format selection control, a format selection window is displayed to receive a handout export format selected by the user; In response to a trigger instruction for the export content screening control, a content check list is displayed to receive content options selected by the user; the content check list includes an analysis content option, a subject label and knowledge point label option, an expanded knowledge option, and an expanded exercise option; In response to a trigger instruction for the export layout setting control, a layout setting window is displayed to receive layout parameters set by the user; In response to a trigger instruction for the export path selection control, displaying a path selection window and receiving a storage path selected by a user; In response to the trigger instruction of the batch export control, a batch export window is displayed, and according to the video teaching file specified by the user, based on the handout export format, the content options and the typesetting parameters, the document generation engine is called to generate the teaching handouts, and they are exported in batches according to the storage path.

[0013] On the other hand, the present application provides a large-model-based intelligent interactive system for film and television teaching, which is used to execute the aforementioned large-model-based intelligent interactive method for film and television teaching.

[0014] The beneficial effects of the present application are as follows: the present application provides an intelligent interactive method for film and television teaching based on a large model, which integrates multiple functions into one main interface, including teaching file management, playback settings, and handout export controls, thereby realizing efficient management and utilization of film and television teaching files in various formats. The method can automatically parse video content using a large model, and automatically generate subject labels and knowledge point labels, and then associate them with the big data knowledge base and exercise database to form an extended knowledge and exercise index, facilitating the organization and expansion of educational resources. In addition, it supports the setting of personalized teaching interaction trigger points based on these indexes, which enhances the interactivity and pertinence of the teaching process. Ultimately, teachers can automatically generate and export teaching handouts based on all the above information, effectively improving the efficiency of handout production and teaching quality. The present application also provides a corresponding system, and the beneficial effects of the system are similar to those of the method, so I will not elaborate on them here.

[0015] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be achieved and obtained through the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation to the technical solution of the present invention.

[0017] Figure 1 This is a flowchart of the intelligent interactive method for film and television teaching based on a large model provided by this application; Figure 2 This is a schematic diagram of the main interface of intelligent interactive film and television teaching based on a large model provided by this application; Figure 3This is a schematic diagram of the teaching file management sub-interface provided by this application; Figure 4 This is a schematic diagram of the playback settings sub-interface provided by this application; Figure 5 This is a schematic diagram of the teaching handout export sub-interface provided by this application; Figure 6 is a schematic diagram of the label generation window provided by this application; Figure 7 is a schematic diagram of the teaching association window provided by this application; Figure 8 It is a structural diagram of the large model provided by this application; Figure 9 This is a structural diagram of the large-scale model-based intelligent interactive system for film and television teaching provided by this application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0019] The present application is further described below in conjunction with the accompanying drawings and specific embodiments. The described embodiments should not be considered as limiting the present application. All other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0020] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0022] With the development of information technology, the field of education is gradually shifting towards digitalization and intelligence. In particular, video, as an intuitive teaching medium, is widely used in various courses. Through film and television materials, teachers can more vividly demonstrate complex concepts, and students can gain a richer learning experience. However, how to efficiently manage these film and television resources and integrate them with interactive teaching methods to improve teaching effectiveness and student engagement has become a major research topic.

[0023] Existing systems allow users to upload, store, and play teaching videos. However, most of these systems lack support for multiple formats and are inconvenient in the classification and retrieval of video resources. At the same time, the lack of effective tag recognition makes it difficult to find and use video resources. Some platforms have begun to try to use artificial intelligence technology to analyze video content, but most are limited to basic voice recognition or simple text extraction, making it difficult to deeply understand the complex information in the video. Although some online learning platforms provide interactive functions such as Q&A and discussion areas, there is little support for learning paths and interactive points customized according to specific video content. In addition, under traditional methods, teachers need to manually organize the content of teaching videos and combine them with relevant knowledge systems to produce handouts. This process is time-consuming and inefficient.

[0024] Based on the technical problem, this application proposes an intelligent interactive method and system for film and television teaching based on a large model. The method first realizes the efficient management and operation of film and television teaching files in various formats through a main interface that integrates teaching file management controls, playback setting controls, and teaching handout export controls. The large model is used to automatically parse the content of the uploaded film and television teaching files, including video images, voice and text, and automatically generate subject labels and knowledge point labels, which are associated with the big data knowledge base and exercise database to generate extended knowledge indexes and extended exercise indexes. In addition, personalized teaching interaction trigger points are set according to these index information to support various forms of intelligent interactive activities such as knowledge point expansion, instant questions and exercise exercises. Finally, teaching handouts containing rich content are automatically generated and exported based on all relevant information, which greatly improves the organizational efficiency of teaching resources, personalized learning experience and overall improvement of teaching effectiveness. This method not only simplifies the management process of film and television education resources, but also effectively promotes the development of interactive teaching.

[0025] First, the intelligent interactive method for film and television teaching based on a large model provided by an embodiment of the present application will be described in detail with reference to the accompanying drawings.

[0026] Reference Figures 1 to 5 The implementation process of the intelligent interactive method for film and television teaching based on a large model provided in the embodiment of the present application includes but is not limited to the following steps.

[0027] Step S110, displaying the intelligent interactive main interface 100 of film and television teaching based on the large model.

[0028] Among them, reference Figure 2 The main interface 100 includes a teaching file management control 101, a play setting control 102 and a teaching handout export control 103.

[0029] In step S110, an intuitive and unified operation entrance is provided to the user through the main interface 100. By building an integrated main interface 100, teachers can quickly access the main functional modules of the system, including uploading and managing teaching videos, setting playback interaction logic, and generating teaching handouts. The design of the main interface 100 not only improves the convenience of user operation, but also provides a good interactive foundation for subsequent intelligent processes. The teaching file management control 101 is responsible for guiding the user to enter the resource management link, the playback setting control 102 is used to configure the teaching interaction logic, and the teaching handout export control 103 supports the output of teaching results. This structured interface design significantly improves the usability and functionality of the system.

[0030] Step S120, in response to the trigger instruction of the teaching file management control 101, displays the teaching file management sub-interface 200, uploads film and television teaching files in various formats, and uses the big model to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files, automatically generates subject labels and knowledge point labels based on the parsed content of the film and television teaching files, and associates them with the big data knowledge base and the big data exercise library to generate an extended knowledge index and an extended exercise index.

[0031] In step S120, in-depth analysis and semantic understanding of the teaching video content are carried out. By introducing a large model (such as a multimodal AI model), the system can simultaneously process multi-source information such as images, voice and subtitle text in the video, breaking through the limitations of traditional single-modal analysis. On this basis, the system can automatically identify the subject categories involved in the video (such as Chinese, mathematics, English, etc.) and extract the key knowledge points therein, thereby generating subject labels and knowledge point labels. These labels are not only a structured expression of the teaching content, but also provide data support for subsequent knowledge expansion and interactive settings. Furthermore, the system associates these knowledge points with the background big data knowledge base and exercise bank to generate an extended knowledge index and an extended exercise index, providing a rich content source and personalized recommendation basis for subsequent teaching interactions. This step realizes the transformation from "video" to "knowledge" and is the basis for realizing intelligent teaching.

[0032] Step S130 , in response to the trigger instruction of the play setting control 102 , the play setting sub-interface 300 is displayed, and the teaching interaction trigger point is set according to the extended knowledge index and the extended exercise index.

[0033] In step S130, after completing the analysis of the teaching video and establishing the knowledge index, focus on the dynamic interaction design in the teaching process. The system allows teachers or learners to set specific time points as teaching interaction trigger points during video playback, such as inserting relevant questions, knowledge point extension prompts or exercise pop-ups when explaining a key knowledge point. The setting of these trigger points is based on the extended knowledge index and exercise index generated in the previous step, ensuring a high degree of matching and accurate push of interactive content and teaching videos. In this way, students can participate in Q&A, consolidate knowledge, and expand their thinking while watching the video, greatly enhancing the immersion and efficiency of learning. In addition, this interactive mechanism also supports the construction of personalized learning paths. Students at different levels can choose different interactive content according to their own needs, truly realizing the teaching concept of teaching students in accordance with their aptitude.

[0034] Step S140, in response to the trigger instruction of the teaching handout export control 103, displays the teaching handout export sub-interface 400, and automatically generates and exports the teaching handout according to the parsed content, extended knowledge index, extended exercise index and teaching interaction trigger point information of the film and television teaching file.

[0035] In step S140, by integrating multi-dimensional data such as video content, knowledge point labels, extended knowledge index, exercise index, and interactive trigger points, the system can automatically generate a teaching handout with a clear structure, complete content, and rich illustrations. The handout not only contains a summary of the teaching content of the original video, but also includes a summary of the corresponding knowledge points, links to extended materials, and preset interactive questions or exercises, which are convenient for teachers to use directly or share with students as after-class review materials. In addition, the system supports the export of handouts in multiple formats (such as PDF, Word, etc.), which improves the flexibility and compatibility of use. This function greatly improves the work efficiency of teachers and ensures the consistency and integrity of teaching content. It is an important bridge connecting teaching preparation and teaching implementation.

[0036] In some embodiments of the present application, reference is made to Figure 3 The teaching file management sub-interface 200 includes a file upload control 201, a one-click parsing control 202, a label generation control 203, and a teaching association control 204. In step S120, in the teaching file management sub-interface 200, film and television teaching files of various formats are uploaded, and the video images, voice signals, and subtitle texts of the film and television teaching files are automatically parsed using a large model. Subject labels and knowledge point labels are automatically generated based on the parsed content of the film and television teaching files, and are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index. The implementation process includes but is not limited to the following steps.

[0037] Step S210 , in response to a trigger instruction for the file upload control 201 , a multi-format file selection window is displayed to receive the video teaching file uploaded by the user.

[0038] In step S210, a flexible and convenient video upload entry is provided for the user. In response to the user's operation of the file upload control 201, the system pops up a multi-format file selection window, which supports the import of multiple mainstream video formats including MP4, AVI, MKV, MOV, etc., ensuring that teachers can easily upload various teaching video resources without the need for prior format conversion. This step not only improves the compatibility and practicality of the system, but also lays the foundation for subsequent automatic parsing and content processing. In addition, this step also has a certain file verification function, which can identify the integrity and parsability of video files, avoid subsequent process interruptions due to file damage or unsupported formats, thereby improving user experience and system stability.

[0039] Step S220, in response to the trigger instruction of the one-key analysis control 202, the file analysis window is displayed, the large model is called to automatically analyze the video images, voice signals, and subtitle texts of the film and television teaching files, and the analysis content is generated and displayed.

[0040] In step S220, the video content is structured and the original video is converted into understandable content. In response to the user clicking on the one-click parsing control 202, the system enters the file parsing window and calls a content extraction engine based on a large model (such as a multimodal AI model) to perform a comprehensive analysis of the uploaded video. The large model can simultaneously process multimodal data such as image information (such as characters, scenes, and charts), voice content (such as explanation words and background sounds), and subtitle text in the video to extract key teaching information. The system then structures this information and presents it to the user in a visual manner in the parsing window, such as key frames on the timeline, text summaries of speech-to-text, and image recognition results. This process not only provides a data basis for subsequent label generation, but also enhances teachers' understanding and control of video content, enabling them to design teaching more accurately.

[0041] Step S230 , in response to the triggering instruction of the label generation control 203 , the label generation window 500 is displayed, and based on the parsed content, the large model is called to generate subject labels and knowledge point labels.

[0042] In step S230, after completing the automatic parsing of the video content, the extracted information is further semantically analyzed and classified to generate a label system with teaching significance. By triggering the label generation control 203, the system calls the large model to perform in-depth semantic understanding of the parsed content, automatically identifies the subject areas involved in the video, such as physics, Chinese, biology, etc., as well as specific knowledge points, such as "universal gravitation", "Pythagorean theorem", "oxidation-reduction reaction", "ancient Chinese appreciation", etc., and generates corresponding subject labels and knowledge point labels. These labels not only help teachers quickly understand the theme and teaching value of the video content, but also provide data support for subsequent resource classification, retrieval and teaching interaction. The label generation window 500 also supports users to manually edit or supplement labels, taking into account the flexibility of automation and manual intervention. This step realizes the transition from content understanding to knowledge organization, and is an important part of building an intelligent teaching system.

[0043] Step S240, in response to the trigger instruction of the teaching association control 204, the teaching association window 600 is displayed, and based on the subject label and the knowledge point label, it is associated with the big data knowledge base and the big data exercise library to generate an extended knowledge index and an extended exercise index.

[0044] In step S240, the system is upgraded from content recognition to knowledge expansion, which is the core support for realizing intelligent teaching interaction. In response to the user's operation of the teaching association control 204, the system enters the teaching association window 600. Based on the generated subject labels and knowledge point labels, it intelligently matches with the background big data knowledge base (such as the teaching knowledge point library, subject knowledge map) and big data exercise library (such as past exam questions and simulation question banks), automatically retrieves the knowledge expansion points and exercises related to the current video content, and generates corresponding extended knowledge indexes and extended exercise indexes. These indexes are not only the data source for setting subsequent teaching interaction trigger points, but also provide teachers with rich teaching resource recommendations to help them build a more complete and systematic teaching content system. In addition, this step also supports personalized matching logic, which can recommend knowledge points and exercises of different difficulty levels based on factors such as students' learning level, interest preferences, etc., thereby realizing the teaching concept of teaching students in accordance with their aptitude.

[0045] In summary, the four controls and corresponding steps in the teaching file management sub-interface 200 constitute a complete process from video upload to knowledge development. Through large-scale model-driven multimodal analysis, semantic tag generation, and knowledge association mechanisms, the system achieves in-depth mining and structured processing of teaching video content, providing solid data support for subsequent teaching interaction settings, handout generation, and other functions. This process not only significantly improves the management efficiency and usability of teaching resources, but also provides technical support for the implementation of intelligent and personalized teaching.

[0046] In some embodiments of the present application, in step S230, subject labels and knowledge point labels are displayed in the label generation window 500, supporting manual adjustment by the user, and the adjusted subject labels and knowledge point labels are associated with the parsed content and stored.

[0047] This step introduces manual intervention on top of automatic label generation, aiming to improve the accuracy, applicability, and teaching practicality of system-generated labels. After parsing the content of the film and television teaching file, the system automatically generates preliminary subject and knowledge point labels based on the understanding capabilities of the large model, such as "Function Graph," "Usage of Classical Chinese Function Words," and "Principles of Photosynthesis." These labels are presented to the user in a clear and editable manner in the label generation window 500.

[0048] The significance of this design lies in the fact that, while the large model possesses powerful semantic understanding capabilities, it may still experience recognition bias when faced with certain highly specialized, ambiguous, or interdisciplinary teaching content. Therefore, by providing a visual tag editing interface, teachers can manually adjust, supplement, or delete system-generated tags based on their own teaching experience and course requirements. This can include correcting incorrect tags, adding missing knowledge points, merging similar tags, and so on, ensuring that the final tag system is more closely aligned with actual teaching scenarios.

[0049] In addition, the system also has the ability to associate and store user-modified labels with the original video analysis content. This means that all manually optimized label information will be recorded by the system and structured in association with the corresponding video clips, knowledge point summaries and other content. This not only provides an accurate data foundation for subsequent knowledge expansion, interactive settings, handout generation and other functions, but also provides support for building a personalized teaching resource library. At the same time, this human-computer collaborative label generation mechanism also helps to train and optimize the performance of large models in specific teaching scenarios, gradually improving the intelligence level of the system.

[0050] Therefore, this step achieves the unity of accuracy, flexibility and controllability in the teaching label generation process by combining the automated processing capabilities of artificial intelligence and the subjective judgment ability of users. It is an important link in promoting the system's evolution from content recognition to knowledge understanding, and lays a solid foundation for the efficient operation of subsequent teaching functions.

[0051] In some embodiments of the present application, reference is made to Figure 6Taking the teaching video of "DNA Double Helix Structure" as an example, in response to the teacher's triggering instruction on the label generation control 203, the system displays a label generation window 500; based on the parsed content of the teaching video of "DNA Double Helix Structure", including the dynamic demonstration screen of the helix structure, the transcribed text of the voice explanation of "base complementary pairing", and the keywords such as "Watson-Crick Discovery Process" in the subtitles, the large model is called to automatically generate the subject label "Biology" and knowledge point labels, such as "DNA Double Helix Structure", "Base Complementary Pairing Principle", "Deoxynucleotide Composition", and "Watson and Crick Experiment"; the left side of the label generation window 500 displays the above labels in list form, and the right side previews the corresponding video clip thumbnail in real time; the teacher clicks the edit button next to the "Base Complementary Pairing Principle" label and refines it to "Base Complementary Pairing Principle (AT, CG)" in the pop-up input box; after the adjustment is completed, the system automatically associates the modified label with the parsed content corresponding to the video and stores it in the database, and updates the label mapping relationship in the multimodal teaching association index.

[0052] In some embodiments of the present application, reference is made to Figure 7 The teaching association window 600 includes a label display area 601, a knowledge base matching control 602, an exercise database matching control 603, and an index generation control 604. In step S240, in the teaching association window 600, based on the generated subject labels and knowledge point labels, they are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index. The implementation process includes but is not limited to the following steps.

[0053] Step S310: In the label display area 601, select the required video teaching file and display its corresponding subject label and knowledge point label.

[0054] In step S310, an intuitive operating interface and basic data support are provided for the entire teaching resource association process. By displaying the uploaded and parsed film and television teaching files and their automatically generated subject labels and knowledge point labels in the label display area 601, users can clearly understand the teaching content summary of each video. This not only makes it convenient for teachers to quickly locate the required teaching materials, but also lays the foundation for subsequent knowledge base matching and exercise bank matching operations. This step ensures that all pending teaching resource information is transparent and visible, allowing users to perform the next step in a targeted manner according to actual teaching needs.

[0055] Step S320, in response to the trigger instruction of the knowledge base matching control 602, retrieve the related extended knowledge from the big data knowledge base based on the subject label and the knowledge point label, and establish a mapping relationship between the extended knowledge and the timeline or content positioning of the film and television teaching file as the first mapping relationship.

[0056] In step S320, the system uses subject labels and knowledge point labels as retrieval basis to automatically search for extended knowledge closely related to the current teaching video content from the big data knowledge base. This extended knowledge may include but is not limited to relevant theoretical backgrounds, application examples, cutting-edge research progress, etc., which greatly enriches the content depth and breadth of the original teaching materials. At the same time, the system will also establish an accurate mapping relationship (i.e., a first mapping relationship) between the found extended knowledge and the timeline or specific content points of the original video to ensure that it can be accurately inserted or linked to the corresponding supplementary materials during playback. The benefit of doing this is that it not only enhances the relevance and coherence of the teaching content, but also improves students' ability to acquire comprehensive knowledge and promotes in-depth learning.

[0057] Step S330, in response to the trigger instruction of the question bank matching control 603, based on the knowledge point tags, extended exercises containing the same or derived knowledge points are retrieved from the big data question bank and associated with the exercise attributes, and a mapping relationship between the extended exercises and the time axis or content positioning of the film and television teaching file is established as the second mapping relationship.

[0058] Among them, exercise attributes include subject labels, knowledge point labels, difficulty labels and usage scenario labels.

[0059] In step S330, interactive elements, namely highly targeted exercises, are added to the teaching video through an intelligent matching mechanism. When the user triggers the exercise bank matching control 603, the system searches the big data exercise bank based on the knowledge point tags, looking for exercises that are directly related to or derived from the current teaching video content. In addition to the exercises themselves, the system also records the detailed attributes of each exercise, such as the subject to which it belongs, the specific knowledge point, the difficulty level, and the applicable teaching context. The system then links these selected exercises to their corresponding positions in the video (timeline or content point) to form dynamic interactive nodes. This approach not only helps students instantly test their learning outcomes, but also provides personalized exercise suggestions based on the different levels of students, further enhancing the effectiveness and fun of teaching.

[0060] Step S340 , in response to the trigger instruction of the index generation control 604 , integrate the subject labels, knowledge point labels, extended knowledge, extended exercises and their corresponding exercise attributes, the first mapping relationship and the second mapping relationship to generate an extended knowledge index and an extended exercise index.

[0061] In step S340, all relevant information collected in the previous steps is systematically organized and finally a structured index file is generated. This includes but is not limited to subject labels, knowledge point labels, extended knowledge obtained from the big data knowledge base, and extended exercises selected from the big data exercise library and their attributes. By creating an extended knowledge index and an extended exercise index, the system can store these valuable teaching resources in an orderly and easily accessible manner, so that both teachers and students can quickly find the required information. In addition, these index files also provide strong support for subsequent teaching activities, such as customizing personalized learning paths, designing interactive teaching plans, etc., thereby achieving the maximum utilization of educational resources and the goal of personalized education.

[0062] In some embodiments of the present application, taking the teaching video of "DNA Double Helix Structure" as an example, in the teaching-related window 600, the left label display area 601 displays the "Biology" subject label and the knowledge labels "DNA Double Helix Structure", "Base Complementary Pairing Principle", "Deoxynucleotide Composition", and "Watson and Crick Experiment"; After the teacher clicks the knowledge base matching control 602, the system retrieves and generates cards from the big data knowledge base based on the tags, displaying them in the extended knowledge area 605, including cards for extended knowledge such as "The Construction Process of the Watson-Crick DNA Model," "DNA X-ray Diffraction Pattern (Franklin Data)," and "Deoxyribonucleotide Molecular Structure." The system also automatically maps these cards to the video's timeline. Clicking on the corresponding card displays the details of the extended knowledge.

[0063] The teacher then triggers the question bank matching control 603. The system selects and generates three matching questions from the big data question bank based on the label "Base Complementary Pairing Principle" and displays them in the extended question area 606 - multiple-choice questions (examining the AT / CG pairing rules), fill-in-the-blank questions (marking the complementary chain corresponding to the DNA single-strand sequence), and experimental analysis questions. It also automatically associates attributes such as "Difficulty: Medium" and "Usage Scenario: After-class Consolidation" and automatically establishes a timeline mapping relationship with the video.

[0064] After clicking the index generation control 604, the system integrates the above-mentioned tags, extended knowledge, extended exercises (including answer analysis paths) and their mapping relationship with the video timeline, generates an extended knowledge index and an extended exercise index, and stores them in the database to support subsequent teaching interaction triggering.

[0065] In some embodiments of the present application, reference is made to Figure 8 The large model includes a visual processing module 810, a speech processing module 820, a text processing module 830, a cross-modal fusion module 840, a subject classification module 850, an educational knowledge graph module 860 and a clustering filtering module 870.

[0066] Visual processing module 810 is responsible for sampling key frames from video images in film and television teaching files and extracting visual features from them. By identifying and analyzing key images in the video, such as characters, scenes, or charts, this module provides basic data support for subsequent semantic understanding and label generation. This process not only helps understand the core themes of the video content, but also helps teachers more intuitively grasp the key points of the teaching materials.

[0067] The speech processing module 820 performs endpoint detection and noise reduction on the speech signals in the film and television teaching files, and converts the processed speech into text. This is crucial for accurately extracting the explanation information in the video. It enables the system to understand and use the speech content to enrich the description of teaching resources, and also provides a basis for subsequent knowledge point expansion and exercise matching.

[0068] The text processing module 830 is responsible for word segmentation and entity recognition of the subtitle text, and further extracts the semantic features of the speech text by combining it with the speech transcription. Through comprehensive analysis of the subtitles and speech text, this module can capture detailed information conveyed in the video, including specific concepts and terminology. This information is the basis for generating structured parsed content.

[0069] Cross-modal fusion module 840 is used to fuse visual and semantic features extracted from video and audio text to generate structured parsed content, including timestamp-linked keywords and their corresponding knowledge point descriptions, key scene annotations, and audio summaries. This integration of multimodal information facilitates a comprehensive understanding of video content and improves the efficiency and quality of educational resource organization.

[0070] The subject classification module 850 is used to output subject labels based on the parsed content. By analyzing the semantic information of the video content, it can determine the main subject areas involved in the video, such as physics, chemistry, history, etc., providing important identification for the classification and retrieval of educational resources.

[0071] The Educational Knowledge Graph module 860 is used to semantically match keywords in the parsed content with a pre-set educational knowledge graph to generate preliminary tags. The educational knowledge graph is a structured knowledge representation that reflects the relationships between different knowledge points. By matching with the knowledge graph, this module can help the system identify specific knowledge points involved in the video, laying the foundation for further interactive teaching design.

[0072] Clustering and filtering module 870 is responsible for semantic clustering and redundant filtering of preliminary tags, retaining preliminary tags with a confidence level higher than 0.85 as the final knowledge point tags, and supplementing the hierarchical relationships between knowledge point tags. This process ensures the accuracy and logic of the tagging system, avoids interference from duplicate and irrelevant tags, and improves the effectiveness and pertinence of teaching resource management.

[0073] In some embodiments of the present application, in step S220, the large model is called to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files, and the implementation process of generating and displaying the parsed content includes but is not limited to the following steps.

[0074] Step S410: perform key frame sampling on the video image and extract visual features through the visual processing module 810810 of the large model.

[0075] In step S410, key frames are sampled from the video, and visual features are extracted using the large-scale model's visual processing module 810. This process aims to identify the most representative image frames from the video instructional files and extract visual information such as scenes, characters, and objects. By analyzing these key frames, the system can understand the basic structure and themes of the video content, providing important visual data support for subsequent knowledge point extraction and interactive teaching design.

[0076] In step S420, endpoint detection and noise reduction processing are performed on the speech signal, and the speech transcription text is extracted through the speech processing module 820 of the large model.

[0077] In step S420, endpoint detection and noise reduction are performed on the speech signal, and the speech transcription is extracted by the large model's speech processing module 820 to generate text. This step first pre-processes the audio portion of the video to remove background noise and accurately segment the speech segments, which are then converted into text. This not only improves the accuracy of speech recognition but also converts the lecture content into text data that can be further analyzed and utilized, facilitating the generation of subject and knowledge point labels.

[0078] Step S430, perform word segmentation and entity recognition on the subtitle text, combine it with the speech transcription text, and use the text processing module 830 of the large model to extract the semantic features of the speech text.

[0079] In step S430, the subtitles undergo word segmentation and entity recognition, and combined with the speech transcription, the large model's text processing module 830 extracts semantic features from the speech text. This step first analyzes the subtitles in the video, identifying key terms and concepts through natural language processing techniques. This information is then combined with the previously extracted speech transcription to capture the specific knowledge conveyed in the video. This process helps the system more comprehensively understand the educational information in the video, laying the foundation for generating detailed analysis content.

[0080] In step S440, the cross-modal fusion module 840 of the large model is used to fuse the visual features and the semantic features of the speech text to generate and display structured parsed content.

[0081] The parsed content includes keywords associated with timestamps and their knowledge point descriptions, key screen annotations, and voice summaries.

[0082] In step S440, the cross-modal fusion module 840 of the large model is used to fuse visual features and speech text semantic features to generate and display structured parsed content. At this stage, the system integrates the visual and text information obtained in the first two steps to create a structured document containing keywords associated with timestamps and their knowledge point descriptions, key picture annotations, and voice summaries. This integration of multimodal information not only makes the video content easier to understand and use, but also provides teachers with rich materials for personalized teaching design and resource management. The final parsed content can serve as an important basis for organizing educational resources, building an extended knowledge index, and producing personalized handouts.

[0083] In some embodiments of the present application, in step S230, the implementation process of calling the large model to generate subject labels and knowledge point labels based on the parsed content includes but is not limited to the following steps.

[0084] Step S510: Based on the parsed content, the subject classification module 850 of the large model outputs the subject label.

[0085] In step S510, based on the parsed content, the subject classification module 850 of the macro model outputs subject labels. The core function of this step is to categorize the video content into disciplines at a macro level. Based on the parsed content, the subject classification module 850 can identify the primary subject area (e.g., mathematics, physics, Chinese, etc.) to which the video content belongs, thereby providing a basic classification basis for subsequent knowledge point extraction, label system construction, and teaching resource organization. This step helps the system quickly identify the teaching purpose of the video, improving the efficiency of resource retrieval and recommendation.

[0086] In step S520, the educational knowledge graph module 860 of the large model is used to semantically match the keywords in the parsed content with the preset educational knowledge graph to generate preliminary labels.

[0087] In step S520, the educational knowledge graph module 860 of the large model is used to semantically match the keywords in the parsed content with the preset educational knowledge graph to generate preliminary labels. This step aims to establish a connection between the video content and the existing structured educational knowledge system. The educational knowledge graph module 860 uses the semantic expression ability of nodes and relationships in the knowledge graph to semantically understand and match the keywords extracted from the video, and identify the knowledge points corresponding to the teaching syllabus or curriculum standards. This step not only achieves accurate docking of video content with standard knowledge points, but also provides semantic support for subsequent knowledge point indexing, expansion recommendations, and teaching interactions.

[0088] In step S530 , the clustering and filtering module 870 of the large model is used to perform semantic clustering and redundant filtering on the preliminary tags, retaining preliminary tags with a confidence level higher than 0.85 as knowledge point tags, and supplementing the hierarchical relationship between the knowledge point tags.

[0089] In step S530, the clustering and filtering module 870 of the large model is used to perform semantic clustering and redundant filtering on the preliminary labels, retaining preliminary labels with a confidence level higher than 0.85 as knowledge point labels, and supplementing the hierarchical relationship between knowledge point labels. This step is an important part of the optimization and structuring of the label system. The clustering and filtering module 870 performs semantic similarity analysis on the preliminary generated labels, merges semantically similar labels, removes duplicate, low-relevance or low-confidence label items, thereby ensuring that the final output knowledge point labels are highly accurate and concise. At the same time, based on the structure of the knowledge graph, the module will supplement the hierarchical relationship of the knowledge point labels in the knowledge system (such as parent knowledge points, child knowledge points, parallel knowledge points, etc.), and provide structured support for subsequent teaching content organization and personalized learning path recommendation.

[0090] In some embodiments of the present application, reference is made to Figure 4 The play setting sub-interface 300 includes an interactive content marking control 301, an interactive configuration selection control 302, and an index association control 303. In step S130, in the play setting sub-interface 300, the implementation process of setting the teaching interaction trigger point according to the extended knowledge index and the extended exercise index includes but is not limited to the following steps.

[0091] Step S610, in response to the trigger instruction of the interactive content marking control 301, the interactive content marking window is displayed, and based on the extended knowledge index and the extended exercise index, the mapping relationship between the extended exercises and extended knowledge and the time axis or content positioning of the film and television teaching file is extracted, and the teaching interaction trigger point is automatically marked on the time axis of the film and television teaching file or the corresponding content positioning.

[0092] Among them, the teaching interaction trigger points include the timeline mark points of the film and television teaching files and the content positioning floating mark points.

[0093] In step S610, an intuitive and convenient way is provided for teachers to seamlessly integrate extended knowledge and extended exercises into teaching videos. By clicking the interactive content markup control 301, the system will pop up an interactive content markup window, displaying information based on the previously generated extended knowledge index and extended exercise index. This information is used to determine which knowledge points or exercises should appear as interactive elements during video playback, and corresponding markup points will be automatically added to the video's timeline or specific content locations. These markup points are divided into two forms: timeline markup points (i.e., triggered when the video plays to a specific moment) and content positioning floating markup points (allowing users to manually select triggers while watching the video). This design not only enhances the interactivity of teaching videos, but also ensures that learners can obtain additional learning resources or test their understanding at key knowledge points.

[0094] Step S620: In response to the trigger instruction of the interaction configuration selection control 302, the interaction configuration window is displayed to receive the interaction type and trigger mode of the teaching interaction trigger point selected by the user.

[0095] Interaction types include knowledge point expansion, instant questions, and exercises. Trigger modes include automatic and manual triggering. Automatic triggering occurs when the video tutorial file plays to a timeline marker. Manual triggering occurs when the user manually clicks on a content-based floating marker.

[0096] In step S620, teachers are allowed to flexibly configure the interaction method according to specific teaching objectives and student needs. By triggering the interaction configuration selection control 302, the system will open the interaction configuration window, providing options for various types of interaction, such as knowledge point expansion (providing additional relevant background knowledge), instant questions (asking questions directly to students to test their immediate understanding ability) and exercises (arranging relevant exercises for students to answer). In addition, teachers can also choose the trigger mode - automatic triggering (automatically starting the interaction when the video playback reaches the preset timeline mark point) or manual triggering (the learner actively clicks on the content to locate the floating mark point to trigger the interaction). Such flexibility allows the teaching process to be adjusted according to actual conditions, supporting both autonomous learning and guided teaching, thereby improving teaching efficiency and effectiveness.

[0097] Step S630 , in response to the trigger instruction of the index association control 303 , the extended exercises and the extended knowledge are bound to the corresponding teaching interaction trigger points according to the extended knowledge index and the extended exercise index.

[0098] In step S630, all prepared interactive elements are integrated together to form a complete interactive teaching program. By activating the index association control 303, the system accurately binds the selected extended exercises and extended knowledge to the previously set teaching interaction trigger points based on the information provided by the extended knowledge index and the extended exercise index. This means that when the video is played to a specific mark point, the system can accurately display the corresponding knowledge point extension materials, ask instant questions or provide exercises, thereby achieving seamless connection between teaching content and interactive elements. This process not only simplifies teaching preparation work, but also ensures the effectiveness and pertinence of the interactive links, which helps to improve students' participation and learning outcomes. In this way, the system realizes the transformation from simple knowledge transfer to interactive and exploratory learning mode.

[0099] In some embodiments of the present application, the main interface 100 further includes a multi-screen split-screen control 104. In response to a trigger instruction to the multi-screen split-screen control 104, a multi-screen split-screen sub-interface is displayed, which synchronously displays the screen of the specified video teaching file and the whiteboard. The user can drag the dividing line to adjust the split-screen ratio, supporting cross-screen interaction.

[0100] This step greatly enhances the flexibility and interactivity of the teaching process by introducing a multi-screen split-screen function. When the user triggers the multi-screen split-screen control 104, the system will display a multi-screen split-screen sub-interface, on which the screen of the specified video teaching file and a virtual whiteboard can be viewed simultaneously. This design allows teachers to use the whiteboard to provide supplementary explanations, draw charts, or write formulas while explaining the video content, thereby providing a richer teaching experience. In addition, users can freely adjust the ratio of the two screen areas by dragging the dividing line to adapt to different teaching needs and personal preferences. More importantly, this function supports cross-screen interaction, which means that teachers can directly connect the marks made on the whiteboard with the video content, or let students participate in the operation of the whiteboard to achieve two-way interaction. This not only promotes the effective communication of information, but also stimulates students' enthusiasm for participation, making the learning process more lively and interesting. In this way, the multi-screen split-screen control 104 provides a powerful tool for modern education, which not only improves teaching efficiency but also enriches teaching methods.

[0101] In some embodiments of the present application, reference is made to Figure 5 The teaching handout export sub-interface 400 includes an export format selection control 401, an export content filter control 402, an export layout setting control 403, an export path selection control 404, and a batch export control 405. In step S140, in the teaching handout export sub-interface 400, the process of automatically generating and exporting teaching handouts based on the parsed content, extended knowledge index, extended exercise index, and teaching interaction trigger point information of the film and television teaching file includes but is not limited to the following steps.

[0102] Step S710 , in response to a trigger instruction on the export format selection control 401 , a format selection window is displayed to receive the lecture export format selected by the user.

[0103] In step S710, a flexible selection mechanism is provided for users, allowing them to choose the most suitable format for exporting teaching materials based on their actual needs. By clicking the export format selection control 401, the system will pop up a format selection window, listing various available document formats (such as PDF, Word, etc.) for the user to choose from. Different formats are suitable for different scenarios and purposes. For example, PDF format is suitable for ensuring the consistency and security of document layout, while Word format facilitates further editing and modification. This design not only enhances the versatility and adaptability of the system, but also enables teachers to prepare and distribute teaching materials more efficiently.

[0104] Step S720 , in response to the triggering instruction of the export content filter control 402 , a content check list is displayed to receive the content options checked by the user.

[0105] Among them, the content check list includes analytical content options, subject label and knowledge point label options, extended knowledge options and extended exercise options.

[0106] In step S720, the user is able to precisely control the specific content included in the final generated teaching handouts. By activating the export content filter control 402, the system will display a content checklist that lists all available content types, such as the parsed content of the original video, automatically generated subject labels and knowledge point labels, extended knowledge obtained from the big data knowledge base, and related exercises selected from the question bank. Teachers can freely check the content items that need to be included in the handouts based on the focus of the course and personal preferences, thereby customizing a comprehensive and targeted teaching material. This step not only improves the flexibility of handout production, but also ensures the high relevance and practicality of the teaching content.

[0107] Step S730: In response to the trigger instruction of the export layout setting control 403, the layout setting window is displayed to receive the layout parameters set by the user.

[0108] In step S730, in order to meet the personalized needs of different users, this step provides a detailed layout setting function. When the user clicks the export layout setting control 403, the system will open a layout setting window, allowing the user to adjust various layout parameters, such as font size, line spacing, page margins, title style, etc. These settings are crucial to ensuring the professional appearance and readability of the handouts, especially when dealing with complex or highly technical content. By providing such fine control, the system helps teachers create visually appealing and easy-to-understand teaching handouts, thereby improving students' learning experience and efficiency.

[0109] Step S740 : In response to the trigger instruction of the export path selection control 404 , a path selection window is displayed to receive the storage path selected by the user.

[0110] In step S740, after completing the handout content selection and layout settings, the generated file is saved. By triggering the export path selection control 404, the system will pop up a path selection window, guiding the user to specify a suitable storage location. This step may seem simple, but it is indispensable for managing and organizing a large number of teaching resources. It ensures that each handout is properly stored and facilitates subsequent retrieval and use. In addition, a clear file management system helps improve work efficiency and reduce time wasted due to not being able to find the required materials.

[0111] Step S750, in response to the trigger instruction of the batch export control 405, the batch export window is displayed, and according to the video teaching file specified by the user, based on the handout export format, content options and layout parameters, the document generation engine is called to generate teaching handouts, and they are exported in batches according to the storage path.

[0112] In step S750, the automated generation and output of teaching handouts is achieved. Once the user activates the batch export control 405, the system will enter the batch export window and call the document generation engine to create teaching handouts based on all previously set parameters (including the selected export format, specific content options, and personalized layout settings). These files are then automatically saved according to the user-specified storage path. This process greatly simplifies the workload of handout production, especially when handouts need to be generated for multiple video files, significantly improving work efficiency. At the same time, this also means that teachers can focus on teaching itself rather than tedious technical operations.

[0113] In summary, the large-model-based intelligent interactive method for film and television teaching provided in the embodiments of the present application has the following technical effects.

[0114] This method achieves efficient management and operation of video teaching files in various formats through an integrated main interface and various controls. It uses a large model to automatically parse video content, and automatically generates subject labels and knowledge point labels. It then associates them with the big data knowledge base and exercise database to generate extended knowledge indexes and extended exercise indexes, greatly enriching teaching resources. The system also supports personalized interactive settings, allowing teachers to set teaching interaction trigger points as needed, including knowledge point expansion, instant questions, and exercise practice, thereby enhancing interactivity and participation in the learning process. In addition, this method provides a complete set of teaching handout export solutions, covering export format selection, content screening, typesetting settings, and other functions, which greatly improves the efficiency of handout production.

[0115] Furthermore, the introduction of a multi-screen split-screen feature allows for the simultaneous display of video and film teaching files and a whiteboard, and supports user-adjustable split-screen ratios and cross-screen interaction, enhancing the flexibility and real-time interactivity of classroom presentations. Overall, this approach not only simplifies the teaching resource management process and improves the efficiency of educational resource utilization, but also promotes improved educational quality through intelligent and personalized teaching design. This provides strong technical support for modern education and helps achieve a more efficient and engaging teaching experience.

[0116] Secondly, refer to Figure 9 The present embodiment provides a large-scale model-based intelligent interactive system for film and television teaching, which is used to implement the aforementioned large-scale model-based intelligent interactive method for film and television teaching. The system includes a main interface module 910, a teaching file management module 920, a playback setting module 930, and a teaching handout export module 940.

[0117] The main interface module 910 is used to display the intelligent interactive main interface of film and television teaching based on the large model; wherein the main interface includes teaching file management controls, playback setting controls and teaching handout export controls.

[0118] The teaching file management module 920 is used to respond to the trigger instruction of the teaching file management control, display the teaching file management sub-interface, upload film and television teaching files in various formats, and use the big model to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files, automatically generate subject labels and knowledge point labels based on the parsed content of the film and television teaching files, and associate them with the big data knowledge base and the big data exercise library to generate an extended knowledge index and an extended exercise index.

[0119] The play setting module 930 is used to respond to the trigger instruction of the play setting control, display the play setting sub-interface, and set the teaching interaction trigger point according to the extended knowledge index and the extended exercise index.

[0120] The teaching handout export module 940 is used to respond to the trigger instruction of the teaching handout export control, display the teaching handout export sub-interface, and automatically generate and export teaching handouts based on the parsed content, extended knowledge index, extended exercise index and teaching interaction trigger point information of the film and television teaching files.

[0121] Similarly, the technical effects of the system embodiments provided in this application are the same as the technical effects of the above-mentioned method embodiments, and will not be elaborated here.

[0122] In some optional embodiments, the functions / operations mentioned in the block diagram may not occur in the order mentioned in the operation schematic diagram. For example, depending on the functions / operations involved, the two boxes shown in succession may actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiments presented and described in the flow chart of the present application are provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logical flows presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0123] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present application. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art will be able to implement the present application as set forth in the claims using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the appended claims and their equivalents.

[0124] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs that enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0125] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable programs for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can retrieve and execute a program from a program execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, a program execution system, apparatus, or device.

[0126] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in a suitable manner as necessary, and then storing it in a computer memory.

[0127] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0128] In the above description of this specification, reference to the terms "one embodiment / implementation," "another embodiment / implementation," or "certain embodiments / implementations" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in the embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0129] Although the embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0130] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. The intelligent interactive method of film and television teaching based on large models is characterized by: The following steps are involved: Displaying a main interface for intelligent interaction of film and television teaching based on a large model; wherein the main interface includes a teaching file management control, a playback setting control, and a teaching handout export control; In response to a trigger instruction on the teaching file management control, a teaching file management sub-interface is displayed, and film and television teaching files in various formats are uploaded. The video images, voice signals, and subtitle texts of the film and television teaching files are automatically parsed using the large model. Subject labels and knowledge point labels are automatically generated based on the parsed content of the film and television teaching files, and the labels are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index. In response to a trigger instruction on the play setting control, a play setting sub-interface is displayed, and a teaching interaction trigger point is set according to the extended knowledge index and the extended exercise index; In response to the trigger instruction of the teaching handout export control, the teaching handout export sub-interface is displayed, and the teaching handout is automatically generated and exported based on the parsed content of the film and television teaching file, the extended knowledge index, the extended exercise index and the information of the teaching interaction trigger point.

2. The intelligent interactive method for film and television teaching based on a large model according to claim 1 is characterized in that: The teaching file management sub-interface includes a file upload control, a one-key parsing control, a label generation control, and a teaching association control; The display teaching file management sub-interface is used to upload film and television teaching files in various formats, and the large model is used to automatically parse the video images, voice signals, and subtitle texts of the film and television teaching files. Subject labels and knowledge point labels are automatically generated based on the parsed content of the film and television teaching files, and the labels are associated with the big data knowledge base and the big data exercise database to generate an extended knowledge index and an extended exercise index, including the following steps: In response to a trigger instruction for the file upload control, a multi-format file selection window is displayed to receive the video teaching file uploaded by the user; In response to a trigger instruction of the one-key parsing control, a file parsing window is displayed, the large model is called to automatically parse the video image, voice signal, and subtitle text of the film and television teaching file, and the parsed content is generated and displayed; In response to a trigger instruction for the label generation control, a label generation window is displayed, and based on the parsed content, the large model is called to generate the subject label and the knowledge point label; In response to the trigger instruction of the teaching association control, the teaching association window is displayed, and based on the subject label and the knowledge point label, it is associated with the big data knowledge base and the big data exercise library to generate the extended knowledge index and the extended exercise index.

3. The intelligent interactive method for film and television teaching based on a large model according to claim 2 is characterized in that: Also includes: The subject label and the knowledge point label are displayed in the label generation window, and the user is supported to manually adjust the subject label and the knowledge point label. The adjusted subject label and the knowledge point label are associated with the parsed content and stored.

4. The intelligent interactive method for film and television teaching based on a large model according to claim 2 is characterized in that: The teaching association window includes a label display area, a knowledge base matching control, an exercise bank matching control and an index generation control; The displaying of the teaching association window, based on the subject label and the knowledge point label, associating them with the big data knowledge base and the big data exercise database to generate the extended knowledge index and the extended exercise index, includes the following steps: In the label display area, select the required film and television teaching file, and display its corresponding subject label and knowledge point label; In response to a trigger instruction of the knowledge base matching control, the associated extended knowledge is retrieved from the big data knowledge base based on the subject label and the knowledge point label, and a mapping relationship between the extended knowledge and the time axis or content location of the film and television teaching file is established as a first mapping relationship; In response to a trigger instruction for the exercise library matching control, based on the knowledge point tag, extended exercises containing the same or derived knowledge points are retrieved from the big data exercise library and associated with exercise attributes, and a mapping relationship between the extended exercises and the timeline or content location of the film and television teaching file is established as a second mapping relationship; the exercise attributes include the subject tag, the knowledge point tag, the difficulty tag, and the usage scenario tag; In response to a trigger instruction for the index generation control, the subject label, the knowledge point label, the extended knowledge, the extended exercises and their corresponding exercise attributes, the first mapping relationship and the second mapping relationship are integrated to generate the extended knowledge index and the extended exercise index.

5. The intelligent interactive method for film and television teaching based on a large model according to claim 2 is characterized in that: The calling of the large model to automatically analyze the video images, voice signals, and subtitle texts of the film and television teaching file, and generating and displaying the analyzed content, includes the following steps: Sampling key frames of the video images and extracting visual features through the visual processing module of the large model; Performing endpoint detection and noise reduction processing on the speech signal, and extracting the speech transcription text through the speech processing module of the large model; Perform word segmentation and entity recognition on the subtitle text, combine it with the speech transcription text, and use the text processing module of the large model to extract the semantic features of the speech text; Through the cross-modal fusion module of the large model, the visual features and the speech text semantic features are fused to generate and display structured parsed content, including keywords associated with timestamps and their knowledge point descriptions, key picture annotations and speech summaries.

6. The intelligent interactive method for film and television teaching based on a large model according to claim 2 is characterized in that: The step of calling the large model to generate the subject label and the knowledge point label based on the parsed content includes the following steps: Based on the analyzed content, outputting the subject label through the subject classification module of the large model; Through the educational knowledge graph module of the large model, the keywords in the parsed content are semantically matched with the preset educational knowledge graph to generate preliminary labels; The clustering and filtering module of the large model is used to perform semantic clustering and redundant filtering on the preliminary labels, retaining preliminary labels with a confidence level higher than 0.85 as the knowledge point labels, and supplementing the hierarchical relationship between the knowledge point labels.

7. The intelligent interactive method for film and television teaching based on a large model according to claim 1 is characterized in that: The playback setting sub-interface includes an interactive content marking control, an interactive configuration selection control, and an index association control; The display playback setting sub-interface sets the teaching interaction trigger point according to the extended knowledge index and the extended exercise index, including the following steps: In response to a trigger instruction for the interactive content marking control, an interactive content marking window is displayed, and based on the extended knowledge index and the extended exercise index, a mapping relationship between the extended exercises and the extended knowledge and the time axis or content location of the film and television teaching file is extracted, and the teaching interaction trigger point is automatically marked on the time axis of the film and television teaching file or the corresponding content location; Wherein, the teaching interaction trigger points include the time axis mark points and content positioning suspension mark points of the film and television teaching file; In response to a trigger instruction of the interaction configuration selection control, an interaction configuration window is displayed to receive the interaction type and trigger mode of the teaching interaction trigger point selected by the user; The interaction types include knowledge point expansion, instant questioning, and exercise practice; the triggering modes include automatic triggering and manual triggering; the automatic triggering refers to automatic triggering when the video teaching file is played to the timeline mark point; the manual triggering refers to triggering when the user manually clicks the content positioning floating mark point; In response to a trigger instruction for the index association control, the extended exercises and the extended knowledge are bound to the corresponding teaching interaction trigger points according to the extended knowledge index and the extended exercise index.

8. The intelligent interactive method for film and television teaching based on a large model according to claim 1 is characterized in that: The main interface also includes a multi-screen split-screen control; In response to the trigger instruction of the multi-screen split-screen control, the multi-screen split-screen sub-interface is displayed, and the screen and whiteboard of the specified film and television teaching file are displayed synchronously. The user can drag the dividing line to adjust the split-screen ratio, and cross-screen interaction is supported.

9. The intelligent interactive method for film and television teaching based on a large model according to claim 1 is characterized in that: The teaching handout export sub-interface includes an export format selection control, an export content screening control, an export layout setting control, an export path selection control, and a batch export control; The display teaching handout export sub-interface automatically generates and exports teaching handouts based on the parsed content of the film and television teaching file, the extended knowledge index, the extended exercise index, and the teaching interaction trigger point information, including the following steps: In response to a trigger instruction on the export format selection control, a format selection window is displayed to receive a handout export format selected by the user; In response to a trigger instruction for the export content screening control, a content check list is displayed to receive content options selected by the user; the content check list includes an analysis content option, a subject label and knowledge point label option, an expanded knowledge option, and an expanded exercise option; In response to a trigger instruction for the export layout setting control, a layout setting window is displayed to receive layout parameters set by the user; In response to a trigger instruction for the export path selection control, displaying a path selection window and receiving a storage path selected by a user; In response to the trigger instruction of the batch export control, a batch export window is displayed, and according to the video teaching file specified by the user, based on the handout export format, the content options and the typesetting parameters, the document generation engine is called to generate the teaching handouts, and they are exported in batches according to the storage path.

10. The intelligent interactive system for film and television teaching based on large models is characterized by: Used to execute the intelligent interactive method for film and television teaching based on a large model as described in any one of claims 1 to 9.

Images