Web-based virtual digital person teaching method and system, storage medium and equipment

Abstract

The invention discloses a Web-based virtual digital person teaching method and system, a storage medium and equipment, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out micro-course content production based on pre-provided courseware data, obtaining micro-course content data, and storing the micro-course content data in a JSON format; the method comprises the following steps: acquiring micro-class content data, performing classified storage on the micro-class content data to obtain a plurality of JSON sub-files, and outputting the JSON sub-files according to a preset teaching process; according to a teaching process, performing picture content identification and video content identification on the JSON sub-file in sequence, and outputting teaching text content; and controlling the virtual digital person to carry out lip shape matching and action matching according to the teaching text content, and outputting audio content corresponding to the micro-class content data. According to the invention, the information interaction logic of the teaching of the virtual digital person is optimized, the expression of the teaching of the virtual digital person is improved, and the teaching effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a Web-based virtual digital human teaching method, system, storage medium and device. Background Art

[0002] With the rapid development of information technology, in some specific cases, students will use computer devices to take online courses, the content of which is more extensive and more selective than in the classroom.

[0003] At present, in order to improve the experience, such online learning platforms usually have the function option of virtual digital humans. Students can choose from several character images set by the system to take classes. However, the current virtual digital humans are not rich and realistic in non-verbal interactions, such as natural expressions of body language and facial expressions, which reduces the naturalness and intimacy of communication. In addition, AI-driven virtual digital humans may show delayed or incoherent responses in continuous interactions, affecting the fluency of the learning experience, thereby reducing the overall teaching effect. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a Web-based virtual digital human teaching method, system, storage medium and device, which optimizes the information interaction logic of the virtual digital human teaching and improves the expression of the virtual digital human during teaching, thereby improving the teaching effect.

[0005] The first invention of the present invention provides a Web-based virtual digital human teaching method, the method comprising: Produce micro-course content based on pre-provided courseware data, obtain micro-course content data and store it in JSON format; Acquire the micro-course content data, classify and store the micro-course content data, obtain multiple JSON sub-files, and output the JSON sub-files according to a preset teaching process; According to the teaching process, the image content recognition and video content recognition are performed on the JSON sub-file in turn, and the teaching text content is output; The virtual digital human is controlled to perform lip shape matching and action matching according to the teaching text content, and output audio content corresponding to the micro-course content data.

[0006] According to one aspect of the above technical solution, the steps of obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining a plurality of JSON sub-files, and outputting the JSON sub-files according to a preset teaching process include: Obtain the micro-course content data, and use the AI ​​model to read the micro-course content data; Identify the content in the micro-course content data and its corresponding teaching process through the AI ​​model; Classifying the micro-course content data through the AI ​​model to obtain multiple knowledge point labels, including label content, blackboard content, and exercise content; The label content, blackboard content and exercise content are stored according to the teaching process to obtain multiple JOSN sub-files; Control the multiple JOSN sub-files to be output according to the teaching process.

[0007] According to one aspect of the above technical solution, according to the teaching process, the steps of performing image content recognition and video content recognition on the JSON sub-file in turn and outputting the teaching text content include: Based on the teaching process, the AI ​​model is used to sequentially identify the plurality of JSON sub-files; Perform image content recognition and video content recognition on the JSON sub-file to extract feature content in the image and video; Generating teaching text content for the virtual digital person to teach according to the characteristic content, and continuously outputting the teaching text content.

[0008] According to one aspect of the above technical solution, in the step of controlling the virtual digital human to perform lip matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform lip matching according to the teaching text content includes: Based on the pre-built generative adversarial network, the audio-image mapping relationship is learned to obtain the lip shape matching parameters; The virtual digital human is controlled to perform lip matching according to the audio features and lip matching parameters in the teaching text content.

[0009] According to one aspect of the above technical solution, in the step of controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform action matching according to the teaching text content includes: Generate an action database based on the training actions of the virtual digital human in the preset training process; The virtual digital human is controlled to match corresponding action features in the action database according to the content of the teaching text.

[0010] According to one aspect of the above technical solution, when the virtual digital human receives an interaction instruction at any time point, the method further includes: Pause outputting the audio content corresponding to the micro-course content data; Parsing the interaction instruction based on the AI ​​model to generate a reply text content corresponding to the interaction instruction; Lip shape matching and action matching are performed based on the reply text content, and the reply audio content is output.

[0011] According to one aspect of the above technical solution, in the process of stopping outputting the audio content corresponding to the micro-course content data and generating the reply text content corresponding to the interactive instruction, the method further includes: The preset lip shape feature set and motion feature set are called, and the lip shape feature set and the motion feature set are executed during the provisional teaching process.

[0012] The preset lip shape feature set includes a plurality of continuous lip shape features, and the preset motion feature set includes a plurality of continuous motion features.

[0013] The second aspect of the present invention is to provide a Web-based virtual digital human teaching system, the system is applied to the method described in the above technical solution, the method comprising: The first content processing module is used to produce micro-course content based on the pre-provided courseware data, obtain the micro-course content data and store it in JSON format; The second content processing module is used to obtain the micro-course content data, classify and store the micro-course content data, obtain multiple JSON sub-files, and output the JSON sub-files according to a preset teaching process; A teaching text content output module is used to perform image content recognition and video content recognition on the JSON sub-file in turn according to the teaching process, and output the teaching text content; The teaching execution module is used to control the virtual digital human to perform lip shape matching and action matching according to the teaching text content, and output audio content corresponding to the micro-course content data.

[0014] A third aspect of the present invention is to provide a readable storage medium having a computer program stored thereon, which implements the method described in the above technical solution when executed by a processor.

[0015] The fourth aspect of the present invention is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method described in the above technical solution when executing the computer program.

[0016] Compared with the prior art, the Web-based virtual digital human teaching method, system, storage medium and device shown in the present invention have the following beneficial effects: By producing micro-course content based on pre-provided courseware data, obtaining micro-course content data and storing it in JSON format, then obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining multiple JSON sub-files, and making the JSON sub-files output according to a pre-set teaching process, then according to the teaching process, sequentially performing image content recognition and video content recognition on the JSON sub-files, outputting the teaching text content, and finally controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content, and outputting the audio content corresponding to the micro-course content data. The method shown in the present invention can improve the teaching demeanor effect of the virtual digital human by transforming the micro-course content and controlling the virtual digital human to perform lip shape and action matching to explain the micro-course content, thereby improving the teaching effect, and solving the problem that the virtual digital human's demeanor is relatively rigid when teaching in the prior art, which makes it difficult to promote the virtual digital human teaching on a large scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 It is a flowchart of a Web-based virtual digital human teaching method shown in one embodiment of the present invention; Figure 2 It is a structural block diagram of a Web-based virtual digital human teaching system shown in one embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. Several embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0020] Embodiment 1 See also Figure 1The first embodiment of the present invention provides a Web-based virtual digital human teaching method, which is mainly used for virtual digital human teaching in the Web segment of the browser in the computer device, and students study the course in the Web segment. The method includes steps S10 to S40: Step S10, producing micro-course content based on pre-provided courseware data, obtaining micro-course content data and storing it in JSON format.

[0021] In this embodiment, the pre-provided courseware data includes courseware content collected in advance according to the teaching content, such as picture content and video content recorded by the teacher himself, etc. By collecting these picture content and video content, the courseware data used for teaching is obtained. Then, in order to achieve the purpose of teaching based on virtual digital people in this embodiment, it is also necessary to produce micro-courses based on the courseware data, so as to obtain micro-course content data. After the micro-course content data is produced, the micro-course content data is stored in JSON format in this embodiment, so as to obtain a JSON file corresponding to the courseware data.

[0022] Step S20, obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining a plurality of JSON sub-files, and outputting the JSON sub-files according to a preset teaching process; In this embodiment, the steps of obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining a plurality of JSON sub-files, and outputting the JSON sub-files according to a preset teaching process include: Obtain the micro-course content data, and use the AI ​​model to read the micro-course content data; Identify the content in the micro-course content data and its corresponding teaching process through the AI ​​model; Classifying the micro-course content data through the AI ​​model to obtain multiple knowledge point labels, including label content, blackboard content, and exercise content; The label content, blackboard content and exercise content are stored according to the teaching process to obtain multiple JOSN sub-files; Control the multiple JOSN sub-files to be output according to the teaching process.

[0023] Specifically, after obtaining the pre-produced micro-course content data, the pre-trained AI model will be used to obtain the micro-course content data, and then the AI ​​model will be used to identify the teaching content in the micro-course content data, as well as the teaching process corresponding to the teaching content. The AI ​​model will then be used to classify the micro-course content data according to the knowledge point labels, and label content, blackboard content, and exercise content will be obtained respectively. The label content, blackboard content, and exercise content mentioned above will be classified and stored according to the pre-set teaching process to obtain multiple JOSN sub-files, each of which is a part of the teaching content. The multiple JOSN sub-files will then be controlled to be output sequentially according to the teaching process.

[0024] More specifically, in this embodiment, micro-course production is performed based on courseware data to obtain micro-course content data in JSON format, and classification and storage are performed based on the micro-course content data to obtain multiple JSON sub-files. The purpose is to output materials for various teaching forms so that virtual digital humans can be used to call data in the future to realize virtual digital human teaching.

[0025] Step S30, according to the teaching process, perform image content recognition and video content recognition on the JSON sub-file in turn, and output the teaching text content; In this embodiment, according to the teaching process, the steps of performing image content recognition and video content recognition on the JSON sub-file in turn and outputting the teaching text content include: Based on the teaching process, the AI ​​model is used to sequentially identify the plurality of JSON sub-files; Perform image content recognition and video content recognition on the JSON sub-file to extract feature content in the image and video; Generating teaching text content for the virtual digital person to teach according to the characteristic content, and continuously outputting the teaching text content.

[0026] In this embodiment, after the JSON file is classified by the AI ​​model to obtain multiple JSON sub-files, and output in sequence according to the teaching process, content recognition and video content recognition are performed on each JSON file to extract the characteristic content of the pictures and videos in the JSON sub-files, and then the teaching text content is generated according to the several characteristic contents, that is, the teaching text content that the virtual digital human needs to call.

[0027] For example, if the picture or video expresses the "solution of a linear equation of two variables", then through image recognition and content recognition, the specific details of the solution process of the "linear equation of two variables" can be identified, such as the derivation of unknown numbers X and Y, and the specific details of the solution. Then these specific details are connected in series to obtain the broadcast material of the virtual digital human. Based on the broadcast material, the virtual digital human can achieve the purpose of teaching.

[0028] Of course, the virtual digital human's teaching should be expressive and similar to the real person's teaching process, rather than being "rigid" and boring. Therefore, in order to enhance the fun of the virtual digital human's teaching, the virtual digital human's lip shape matching and action matching will be performed in this embodiment to make the virtual digital human's teaching closer to real person's teaching.

[0029] Step S40, controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content, and outputting audio content corresponding to the micro-course content data.

[0030] In this embodiment, in the step of controlling the virtual digital human to perform lip matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform lip matching according to the teaching text content includes: Based on the pre-built generative adversarial network, the audio-image mapping relationship is learned to obtain the lip shape matching parameters; The virtual digital human is controlled to perform lip matching according to the audio features and lip matching parameters in the teaching text content.

[0031] In this embodiment, in the step of controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform action matching according to the teaching text content includes: Generate an action database based on the training actions of the virtual digital human in the preset training process; The virtual digital human is controlled to match corresponding action features in the action database according to the content of the teaching text.

[0032] Specifically, in order to achieve lip shape matching of virtual digital humans, we first learn the mapping relationship between audio and image based on the pre-built generative adversarial network to obtain numerous lip shape matching parameters, such as the lip shape matching parameters corresponding to many keywords including "binary linear", "equation", "solution", etc., that is, each keyword has a corresponding lip shape matching parameter. By running the code, the lip shape matching parameters can be called to enable the virtual digital human to perform the lip shape expression.

[0033] Among them, after the teaching text content, which is the teaching material used for virtual digital people to teach, is output according to the AI ​​model, the virtual digital people will be controlled to perform lip matching according to the audio features and lip matching parameters in the teaching text content. For example, the explanation of a micro-class requires matching a large number of lip movements to complete the virtual digital people's mouth expression. In this way, the virtual digital people's mouths can be close to the expressions of real people when they are teaching micro-classes.

[0034] Similarly, in order to make the movements of the virtual digital human in line with the teaching content when giving micro-classes, this embodiment will first pre-train the virtual digital human so that the virtual digital human can complete common movements such as "nodding", "shaking head", "stretching out hand", "retracting hand", and movements that can be naturally expressed according to the teaching content, such as the action of "stretching out hand to point to knowledge points". After outputting the teaching text content based on the AI ​​model, this embodiment will control the virtual digital human to match the movements in the pre-established action database. The movements refer to movements other than mouth movements, so that the virtual digital human's teaching can be more vivid.

[0035] Finally, after matching the lip shape and the action, the audio content corresponding to the micro-course content data is output, that is, the content that the students can finally hear and see. This embodiment can make the virtual digital person more realistic when giving micro-course lectures by matching the lip shape and the action, which is conducive to improving the students' cooperation.

[0036] Therefore, the Web-based virtual digital human teaching method shown in this embodiment has the following beneficial effects: By producing micro-course content based on pre-provided courseware data, micro-course content data is obtained and stored in JSON format, then the micro-course content data is obtained, the micro-course content data is classified and stored, a plurality of JSON sub-files are obtained, and the JSON sub-files are output according to a pre-set teaching process, and then according to the teaching process, the image content recognition and video content recognition are performed on the JSON sub-files in turn, and the teaching text content is output, and finally the virtual digital human is controlled to perform lip shape matching and action matching according to the teaching text content, and the audio content corresponding to the micro-course content data is output. The method shown in this embodiment can improve the teaching demeanor effect of the virtual digital human by transforming the micro-course content and controlling the virtual digital human to perform lip shape and action matching to explain the micro-course content, thereby improving the teaching effect, and solving the problem that the virtual digital human's demeanor is relatively rigid when teaching in the prior art, which makes it difficult to promote the virtual digital human teaching on a large scale.

[0037] Embodiment 2 The second embodiment of the present invention also provides a Web-based virtual digital human teaching method. The virtual digital human teaching method shown in this embodiment is basically consistent with the method flow shown in the first embodiment, except that: In this embodiment, when the virtual digital human receives an interaction instruction at any time point, the method further includes: Pause outputting the audio content corresponding to the micro-course content data; Parsing the interaction instruction based on the AI ​​model to generate a reply text content corresponding to the interaction instruction; Lip shape matching and action matching are performed based on the reply text content, and the reply audio content is output.

[0038] Wherein, in the process of stopping outputting the audio content corresponding to the micro-course content data and generating the reply text content corresponding to the interactive instruction, the method further includes: The preset lip shape feature set and motion feature set are called, and the lip shape feature set and the motion feature set are executed during the provisional teaching process.

[0039] The preset lip shape feature set includes a plurality of continuous lip shape features, and the preset motion feature set includes a plurality of continuous motion features.

[0040] In this embodiment, although the virtual digital human teaches according to the teaching process, it can interact with students at any time point. This requires that the computer device held by the student has a voice input device or a video input device to interact with the virtual digital human online.

[0041] Specifically, at any time point, students trigger the interactive process through specific keywords, that is, send interactive instructions or interactive requests to the virtual digital human, then the virtual digital human teaching can pause the current teaching process and parse the interactive instructions based on the AI ​​model. For example, if the interactive instruction issued by the student records "how to verify the solution of a quadratic equation", the AI ​​model will think about the interactive instruction, output other content recorded in addition to the teaching content, and re-match the lip shape and movement based on the content, and output the reply audio content, so that students can learn the content in the Web segment.

[0042] Of course, interactive instructions are not necessarily used for learning, but can also be used to obtain information. For example, if you ask the virtual digital person "What will the weather be like tomorrow?", the virtual digital person shown in this embodiment can also serve as a voice assistant.

[0043] Embodiment 3 See also Figure 2The third embodiment of the present invention provides a Web-based virtual digital human teaching system, which is applied to the method described in the above embodiments. The system includes: a first content processing module 10, a second content processing module 20, a teaching text content output module 30 and a teaching execution module 40.

[0044] The first content processing module 10 is used to produce micro-course content based on pre-provided courseware data, obtain micro-course content data and store it in JSON format; The second content processing module 20 is used to obtain the micro-course content data, classify and store the micro-course content data, obtain multiple JSON sub-files, and output the JSON sub-files according to a preset teaching process; A teaching text content output module 30 is used to perform image content recognition and video content recognition on the JSON sub-file in sequence according to the teaching process, and output the teaching text content; The teaching execution module 40 is used to control the virtual digital human to perform lip matching and action matching according to the teaching text content, and output audio content corresponding to the micro-course content data.

[0045] The beneficial effects of using the Web-based virtual digital human teaching system shown in this embodiment are: By producing micro-course content based on pre-provided courseware data, obtaining micro-course content data and storing it in JSON format, then obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining multiple JSON sub-files, and making the JSON sub-files output according to the pre-set teaching process, then according to the teaching process, sequentially performing image content recognition and video content recognition on the JSON sub-files, outputting the teaching text content, and finally controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content, and outputting the audio content corresponding to the micro-course content data. The system shown in this embodiment can improve the teaching demeanor effect of the virtual digital human by transforming the micro-course content and controlling the virtual digital human to perform lip shape and action matching to explain the micro-course content, thereby improving the teaching effect, and solving the problem that the virtual digital human's demeanor is relatively rigid when teaching in the prior art, which makes it difficult to promote the virtual digital human teaching on a large scale.

[0046] Embodiment 4 A fourth embodiment of the present invention provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method described in the above embodiment is implemented.

[0047] Embodiment 5 A fifth embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method described in the above embodiments when executing the computer program.

[0048] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples" or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0049] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A Web-based virtual digital human teaching method, characterized in that: The method comprises: Produce micro-course content based on pre-provided courseware data, obtain micro-course content data and store it in JSON format; Acquire the micro-course content data, classify and store the micro-course content data, obtain multiple JSON sub-files, and output the JSON sub-files according to a preset teaching process; According to the teaching process, the image content recognition and video content recognition are performed on the JSON sub-file in turn, and the teaching text content is output; Controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content, and outputting audio content corresponding to the micro-course content data; The steps of obtaining the micro-course content data, classifying and storing the micro-course content data, obtaining a plurality of JSON sub-files, and outputting the JSON sub-files according to a preset teaching process include: Obtain the micro-course content data, and use the AI ​​model to read the micro-course content data; Identify the content in the micro-course content data and its corresponding teaching process through the AI ​​model; Classifying the micro-course content data through the AI ​​model to obtain multiple knowledge point labels, including label content, blackboard content, and exercise content; The label content, blackboard content and exercise content are stored according to the teaching process to obtain multiple JOSN sub-files; Control the multiple JOSN sub-files to be output according to the teaching process.

2. The Web-based virtual digital human teaching method according to claim 1, characterized in that: According to the teaching process, the steps of performing image content recognition and video content recognition on the JSON sub-file in turn and outputting the teaching text content include: Based on the teaching process, the AI ​​model is used to sequentially identify the plurality of JSON sub-files; Perform image content recognition and video content recognition on the JSON sub-file to extract feature content in the image and video; Generating teaching text content for the virtual digital person to teach according to the characteristic content, and continuously outputting the teaching text content.

3. The Web-based virtual digital human teaching method according to claim 2, characterized in that: In the step of controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform lip shape matching according to the teaching text content includes: Based on the pre-built generative adversarial network, the audio-image mapping relationship is learned to obtain the lip shape matching parameters; The virtual digital human is controlled to perform lip matching according to the audio features and lip matching parameters in the teaching text content.

4. The Web-based virtual digital human teaching method according to claim 2, characterized in that: In the step of controlling the virtual digital human to perform lip shape matching and action matching according to the teaching text content and outputting the audio content corresponding to the micro-course content data, the step of controlling the virtual digital human to perform action matching according to the teaching text content includes: Generate an action database based on the training actions of the virtual digital human in the preset training process; The virtual digital human is controlled to match corresponding action features in the action database according to the content of the teaching text.

5. The Web-based virtual digital human teaching method according to any one of claims 1 to 4, characterized in that: When the virtual digital human receives an interactive instruction at any time point, the method further includes: Pause outputting the audio content corresponding to the micro-course content data; Parsing the interaction instruction based on the AI ​​model to generate a reply text content corresponding to the interaction instruction; Lip shape matching and action matching are performed based on the reply text content, and the reply audio content is output.

6. The Web-based virtual digital human teaching method according to claim 5, characterized in that: In the process of stopping outputting the audio content corresponding to the micro-course content data and generating the reply text content corresponding to the interactive instruction, the method further includes: The preset lip shape feature set and motion feature set are called, and the lip shape feature set and the motion feature set are executed during the provisional teaching process. The preset lip shape feature set includes a plurality of continuous lip shape features, and the preset motion feature set includes a plurality of continuous motion features.

7. A Web-based virtual digital human teaching system, characterized in that: The system is applied to the method described in any one of claims 1 to 6, the method comprising: The first content processing module is used to produce micro-course content based on pre-provided courseware data, obtain micro-course content data and store it in JSON format; The second content processing module is used to obtain the micro-course content data, classify and store the micro-course content data, obtain multiple JSON sub-files, and output the JSON sub-files according to a preset teaching process; A teaching text content output module is used to perform image content recognition and video content recognition on the JSON sub-file in turn according to the teaching process, and output the teaching text content; The teaching execution module is used to control the virtual digital human to perform lip matching and action matching according to the teaching text content, and output audio content corresponding to the micro-course content data.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of being executed on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images