Image content automatic generation-based smart education system

Abstract

The present invention relates to an image content automatic generation-based smart education system capable of providing a speech learning service for a user on the basis of an image generation model and an evaluation solution in the field of artificial intelligence-based education platforms and, more specifically, to an image content automatic generation-based smart education system capable of providing a smart education platform for a user to view image content provided from a smart education platform server, construct sentences (English, Korean, and various other languages) for creating images similar to the image content, and provide the sentences to the smart education platform, wherein the platform generates and provides images on the basis of the sentences of the user, and evaluates pronunciation fluency and provides feedback regarding same so that the user can learn speaking using the corresponding sentences.

Description

Smart education system based on automatic image content generation

[0001] The present invention relates to a smart education system based on automatic generation of image content, and more particularly, in the field of artificial intelligence-based education platforms, to a smart education system based on automatic generation of image content that can provide a user with a speaking learning service based on an image generation model and an evaluation solution.

[0002] In recent years, the rapid development of artificial intelligence technology has led to the rapid development of innovative AI-based educational services in the education sector.

[0003] However, in the field of language education (various language services such as Korean conversation education services and English conversation services), artificial intelligence technology has limitations in speech recognition functions, and the creation and evaluation of various contents still mainly relies on passive system structures that rely on pre-built data or specialized personnel (instructors, teachers, etc.).

[0004] In this context, the present invention proposes an innovative system and service that actively utilizes AI technology in areas such as content creation, speaking fluency assessment, and sentence structure assessment to provide learners (users) with diverse speaking training. Technological advancements in recent years have brought about revolutionary changes in online education and learning methods. Online education platforms have significantly increased educational accessibility and played a crucial role in expanding the learning experience. This has provided learners with the opportunity to study a variety of topics and improve their skills regardless of time and location.

[0005] However, maintaining learner engagement and participation has emerged as a significant challenge in online education. Language learning, especially speaking practice, is considered one of these challenges. Speaking practice plays a crucial role in improving language skills, but effectively assessing and providing feedback on learners' pronunciation accuracy, fluency, and speaking rate in an online environment is technically complex.

[0006] Furthermore, educational methods utilizing images are effective in stimulating learners' interest and engagement through visual stimulation and aiding conceptual understanding. However, the process of learners viewing images and describing them in sentences, or creating images based on sentences, through the interaction between images and text, is an area that existing technologies have not adequately addressed.

[0007] Therefore, the present invention seeks to overcome these problems, provide an innovative method to increase learner participation and engagement, and improve language and speaking skills.

[0008] The technical problem to be solved by the present invention is to provide a smart education system based on automatic generation of image content that can induce learner participation and engagement and provide an effective method for improving language learning and speaking skills.

[0009] Another technical challenge that the present invention seeks to address is to provide a smart education system based on automatic generation of image content that can maximize learning effectiveness in an online education environment while allowing learners to participate more actively and improve their language and speaking skills by making the learning experience more interesting and effective through interaction with images and sentences.

[0010] Another technical challenge that the present invention seeks to address is to provide a smart education system based on automatic generation of image content that can provide an image-based speaking learning system and service based on solutions for creating various contents based on artificial intelligence, various sentence evaluation and correction solutions, and speaking fluency solutions.

[0011] Another technical problem that the present invention seeks to solve is to provide a smart education system based on automatic generation of image content, including a service for registering and sharing images created by learners to provide efficient education services and learning motivation to learners (users).

[0012] The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0013] According to an embodiment of the present invention for achieving the above technical task, a smart education system based on automatic generation of image content is characterized by including a smart education platform that provides first image content, evaluates a learning result of a learner performed based on the provided first image content, and provides the evaluated result to the learner; and a learner platform that generates text or records voice based on the first image content provided from the smart education platform, transmits the generated text or voice to the smart education platform, and receives an evaluation result for the text or voice.

[0014] In addition, the smart education platform is characterized by including a learning content management unit that generates sentences to be provided to learners and the first image content and transmits them to the learner platform; and a similarity evaluation unit that evaluates the accuracy of text and voice generated by the learner based on the first image content transmitted to the learner platform.

[0015] In addition, the smart education platform further includes a learner image generation unit that generates second image content based on text or voice generated by the learner, and the similarity evaluation unit is characterized in that it evaluates accuracy by comparing the first image content and the second image content.

[0016] In addition, the smart education platform is characterized by including a sentence correction / recommendation unit that provides the learner with a correction sentence for a learner sentence generated based on text or voice generated by the learner, an expanded recommended sentence based on the learner sentence, and a correct sentence for the first image content.

[0017] In addition, the present invention is characterized by further including a speaking evaluation unit that selects one of the sentences generated through the voice generated by the learner, the correction sentence, the recommended sentence, and the correct sentence, and when the learner records the selected sentence, evaluates the learner's speaking fluency based on the selected sentence and the recording file.

[0018] In addition, the smart education platform is characterized by including a personal content management unit that stores all matters performed through the learner platform.

[0019] In addition, the learner platform is characterized by including a learner sentence generation unit that generates text or voice created by the learner based on the first image content received from the learning content management unit and transmits the generated text or voice to the smart education platform; and a learner image display unit that receives the evaluation result through the similarity evaluation unit.

[0020] In addition, the learner platform is characterized by including a speaking learning unit that checks the sentences created by the learner, the corrected sentences, the recommended sentences, and the correct sentences, selects one sentence for speaking practice, and then records the selected sentence as a voice; and an evaluation screen management unit that receives feedback on the evaluation results through the speaking evaluation unit and provides guidance on whether to relearn, end learning, save learning results, and share learning results.

[0021] The above embodiments of the present invention are only some of the preferred embodiments of the present invention, and various embodiments reflecting the technical features of the present invention can be derived and understood by a person having ordinary knowledge in the relevant technical field based on the detailed description of the present invention described below.

[0022] According to the present invention, the following effects are achieved.

[0023] First, the present invention provides a service that evaluates learners' descriptive sentences based on images generated by AI, and conversely, generates images based on learner-generated sentences and evaluates their consistency. Furthermore, by assessing the pronunciation fluency of learners' generated sentences and providing feedback to improve pronunciation in real-life speaking (free speech), the service can improve learners' sentence construction and speaking skills by combining sentence construction for image generation with speaking practice.

[0024] Additionally, by interacting with images and sentences, it can enhance the imagination for sentence practice and improve the ability to construct sentences, and by repeating this process, it can maximize and develop the learner's language skills.

[0025] Furthermore, the present invention provides a personalized learning experience, that is, a customized learning path utilizing the learner's voice and text data, and provides an opportunity to identify and improve the learner's development direction, thereby providing a learning experience optimized for the abilities and needs of each learner, thereby maximizing learning efficiency.

[0026] In addition, the present invention can have a positive effect on improving learning motivation and learning performance by promoting learners' interest and participation through a learning method utilizing images and by using visual stimuli to stimulate learners' interest and increase the enjoyment of learning.

[0027] In addition, the present invention can provide a picture-based word learning solution or support effective learning through images and sentences in the process of preparing for a language test.

[0028] Furthermore, the present invention can propose an innovative change in the field of language education by providing learners with various language education services, thereby improving learning efficiency and providing an engaging learning experience.

[0029] The effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by those skilled in the art to which the present invention pertains from the description below.

[0030] Figure 1 is a procedure diagram according to an embodiment of a smart education system based on automatic image content generation of the present invention.

[0031] Figure 2 is a configuration diagram according to an embodiment of a smart education system based on automatic image content generation of the present invention.

[0032] Figure 3 is a configuration diagram of a learning content management unit according to the present invention.

[0033] Figure 4 is a configuration diagram of a learner image generation unit according to the present invention.

[0034] Figure 5 is a configuration diagram of a similarity evaluation unit according to the present invention;

[0035] Figure 6 is a configuration diagram of a sentence correction / recommendation unit according to the present invention.

[0036] Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. When designating components in each drawing, it should be noted that, where possible, identical components will be given the same reference numerals, even if they appear in different drawings. Furthermore, when describing embodiments of the present invention, if a detailed description of a related known structure or function is deemed to hinder understanding of the embodiments of the present invention, such detailed description will be omitted.

[0037] Additionally, terms such as first, second, A, B, (a), (b), etc. may be used to describe components of embodiments of the present invention. These terms are only intended to distinguish the components from other components, and the nature, order, or sequence of the components are not limited by the terms. When it is described that a component is "connected," "coupled," or "connected" to another component, it should be understood that the component may be directly connected or connected to the other component, but components may also be "connected," "coupled," or "connected" between each component.

[0038] FIG. 1 is a diagram showing a procedure configuration according to an embodiment of a smart education system based on automatic generation of image content according to the present invention, FIG. 2 is a diagram showing a configuration concept according to an embodiment of a smart education system based on automatic generation of image content according to the present invention, FIG. 3 is a diagram showing a configuration of a learning content management unit according to the present invention, FIG. 4 is a diagram showing a configuration of a learner image generation unit according to the present invention, FIG. 5 is a diagram showing a configuration of a similarity evaluation unit according to the present invention, and FIG. 6 is a diagram showing a configuration of a sentence correction / recommendation unit according to the present invention.

[0039] With reference to FIGS. 1 to 6, the smart education system based on automatic image content generation according to the present invention is described as follows.

[0040] The present invention relates to a smart education system based on automatic generation of image content, a learner speaking service based on image generation for solving problems occurring in online education systems and education services and for learner participation.

[0041] The procedure and concept of the service proposed in the present invention are as illustrated in Fig. 1, and the smart education platform server side and terminal side functions for providing the service are as illustrated in Fig. 2.

[0042] First, referring to Fig. 1, the procedure configuration according to the smart education system based on automatic image content generation of the present invention is described as follows.

[0043] Learners can select a picture to learn from among the pictures provided by the service of the present invention.

[0044] Learners can view images (primary image content) displayed on the terminal and interpret them into sentences. The interpreted sentence, for example, "cat and tree," can be entered using text, voice recording, or other input methods.

[0045] The sentences composed by the learner are transmitted to the smart education platform server, and the server can evaluate the similarity between the image (second image content) created based on the sentence composed by the learner and the exposed picture (first image content) and provide it to the learner.

[0046] Learners can view the images and sentences they have created and choose whether to create images again or learn to speak.

[0047] When selecting a re-challenge to generate an image again, the learner can return to the step of generating a sentence that interprets the image while the image is again exposed to the terminal and generate a sentence that describes the image exposed to the terminal.

[0048] When a learner selects speaking learning, the screen of the terminal may present a corrected sentence that corrects errors in the sentence composed by the learner, a recommended sentence expanded based on the sentence composed by the learner, and the correct sentence for the problem.

[0049] Learners can select sentences for speaking practice from the above-mentioned correction sentences, recommended sentences, and correct sentences. Once a sentence is selected, learners can read and record it.

[0050] The learner's recorded voice and the sentences selected by the learner are transmitted to the smart education platform server, and the server can show the results of the speaking evaluation (including pronunciation evaluation, fluency evaluation, speaking speed, rhythm, stress, etc.) to the learner through the terminal.

[0051] Learners can choose a different sentence to practice speaking again. Upon selecting speaking practice again, they can return to the content display screen to begin learning. The content display screen refers to the state in which the terminal screen displays corrected sentences, expanded recommended sentences based on the learner's original sentences, and the correct sentences for the questions.

[0052] After completing their speaking practice, learners can register (save) and manage the sentences and images they create in their photo album, and share their own images and sentences. A photo album can refer to a space where users can store and manage images and other media content. This feature, available within certain apps or platforms, allows users to save images to their computers, smartphones, tablets, etc., and easily access them when needed.

[0053] Sharing images and text from your photo album can be done in a variety of ways.

[0054] First, learners can select images and sentences created within the learning platform and share them with other users or groups using the internal sharing (within the app) function. This also allows them to share materials with other platform users.

[0055] Additionally, after completing the learning, learners can upload the images and sentences saved in their photo album to social media platforms (e.g., Facebook, Instagram, Twitter, etc.) or create a shareable link to share them with others.

[0056] Additionally, learners can create shared links for the images and sentences they create after completing a course and share them with others. This allows learners to view shared content even if they aren't signed up for a specific platform or have access to their photo album.

[0057] Additionally, learners can share the images and sentences they create with others via email or messaging apps after completing their learning.

[0058] Information included when registering or sharing may include learner-generated sentences (or corrected sentences), images, learner information, date and time, sharing settings, sharing links, feedback or comments, social media information, similarity scores, speaking fluency scores, etc.

[0059] Next, the following description will be given with reference to FIGS. 2 to 6.

[0060] FIG. 2 is a conceptual diagram of a system configuration according to the present invention for providing the service illustrated in FIG. 1.

[0061] First, the smart education system based on automatic image content generation of the present invention can be configured to include a smart education platform and a learner platform.

[0062] The above smart education platform can provide first image content, evaluate the learning outcomes of learners based on the provided first image content, and provide the evaluated results to the learners. The smart education platform can include a learning content management unit, a learner character processing unit, a learner image generation unit, a similarity evaluation unit, a sentence correction / recommendation unit, a speaking evaluation unit, and a personal content management unit.

[0063] The above-mentioned learner platform can generate text or record voice based on the first image content provided from the smart education platform, transmit the generated text or voice to the smart education platform, and receive evaluation results for the text or voice. The above-mentioned learner platform can be configured to include an image content management unit, a learner sentence generation unit, a learner image display unit, a speaking learning unit, and an evaluation screen management unit.

[0064] The above learning content management unit is a content generation unit for providing the service of the present invention, and can be configured as illustrated in Fig. 3. The learning content management unit can periodically generate sentences and images (first image content) to be provided to learners.

[0065] Figure 3 illustrates the main function of the learning content management unit, including the ability to generate periodic content. When sentences for learners are pre-written and registered as learning sentences, the registered text (learning sentences) are provided to an AI-based image generation model to generate images. The text and images are then transferred to the learning data storage unit for management.

[0066] Additionally, you can build a database that stores images and sentences, and use pre-prepared sentence templates and image data to create new content. By combining sentences and images in the database, you can create a variety of learning content.

[0067] Additionally, natural language processing technology can be used to analyze text related to images and generate corresponding sentences. For example, this can be used to generate descriptions of images or generate sentences based on objects or features within the image.

[0068] Additionally, registered content sentences (including English, Korean, and other foreign languages) can be passed to a local image generation model to generate images, but the system can also include a section for generating and storing images externally by passing text sentences and image styles to an external image generation model (e.g., Dall-e). Furthermore, natural language processing technology can be used to analyze image-related text and generate corresponding sentences. For example, image descriptions can be generated, or sentences can be generated based on objects or features within the image.

[0069] External API processing conforms to external API specifications, and the present invention may include a section for generating commands and sentences for API utilization. If the image creation result is transmitted via a non-image path through external API processing, a function for receiving and storing the image from the path may be included.

[0070] Additionally, scripts can be developed to automate image and sentence generation tasks. For example, a script can be run periodically (daily, weekly, etc.) to generate and distribute sentences and images for learners.

[0071] Additionally, machine learning models can be used to predict sentences and images appropriate for a learner by analyzing the learner's previous activities or tendencies.

[0072] Learning content (images) generated by the above learning content management unit can be provided to a learner platform (terminal). At this time, information such as the learning level difficulty, creation date, and image style (e.g., cartoon style, illustrator style, watercolor style, sketch style, etc.) of the image can be transmitted together. The image content management unit of the above learner platform can filter and display to learners by difficulty level, creation date, and image style based on the information provided by the above smart education platform (server side). Here, the user can select the image he or she wishes to learn and begin learning.

[0073] The learner can view the image (first image content) selected from the image content management unit and generate sentences for generating the same image through the learner sentence generation unit. Here, the sentence generation method may include both voice recording using the terminal's microphone and text input using the terminal's typing function. Through the learner sentence generation unit, the learner can compose sentences using voice or text depending on the input method selected and transmit the voice file or text to the learner sentence processing unit of the smart education platform via the server.

[0074] The above learner sentence processing unit can generate a voice recognition result (text) through a voice recognizer when a voice file is transmitted from the terminal and transmit it to the learner image generation unit. Additionally, when only text is provided from the terminal, the provided text can be transmitted to the learner image generation unit.

[0075] Referring to FIG. 4, the learner image generation unit can generate an image (second image content) based on text or voice received from the learner sentence processing unit. Here, a local image generation model can be used to generate a text- or voice-based image, and, if necessary, an external API can be accessed to transmit text and commands to generate an image. The learner image and learner sentence generated by the learner image generation unit can be transmitted to the similarity evaluation unit.

[0076] The above similarity evaluation unit can evaluate the mutual similarity between the image generated by the learner (second image content) and the learning content image (first image content). Here, two methods can be utilized for similarity evaluation, as illustrated in Figure 5.

[0077] The first method evaluates the similarity between images by calculating the distance between the learner-generated image and the correct image through correlation analysis. In other words, this method calculates the distance by analyzing the correlation between images. This can be achieved by comparing the similarity of image characteristics or pixel values. For example, the learner-generated image and the correct image could each be expressed numerically, and the correlation between these numbers could be analyzed to determine the similarity between the images. Correlation analysis can then be used to numerically assess the degree of similarity.

[0078] Alternatively, AI models can be trained to assess how similar a new image is to the correct answer. Specifically, an AI model can be trained to evaluate the similarity between images. For example, multiple images and their corresponding correct answer images can be provided as training data, allowing an AI model to be trained. The trained model can then be used to assess the similarity between a new image and the correct answer. This model automatically learns the characteristics of the images and can then determine the similarity between the images based on these characteristics.

[0079] The second method can include both a method that uses word similarity to evaluate the similarity between learner sentences and correct sentences, and a method that evaluates the semantic similarity of sentences through learning of an artificial intelligence model.

[0080] First, a method can be used to compare the similarity between the words in the sentences created by the learner and the correct sentences. For example, the similarity between the words in the sentences can be assessed by comparing their meanings and usage. Similarity between sentences can be measured by determining the semantic similarity between words.

[0081] Another method involves training an AI model to assess semantic similarity between sentences. The model is trained using information necessary to understand the structure and meaning of sentences, and can then be used to predict semantic similarity between new sentences.

[0082] These two methods can be used to evaluate the similarity between learner-generated sentences and the correct sentences. Depending on the learner's situation or the assessment system, the evaluation results from each method can be used independently. Alternatively, the evaluation scores from both methods can be combined to ultimately assess the learner's sentence similarity. This allows for assessing how similar the learner's sentences are to the correct sentences.

[0083] The final learner's similarity evaluation can be evaluated using the following formula.

[0084] (Formula) Similarity evaluation = (1-α) × image similarity + (α × sentence similarity)

[0085] At this time, α (alpha) is a constant that can be set to a value between 0 and 1, and plays a role in adjusting the weights of the image similarity and sentence similarity indices.

[0086] Image similarity, which is obtained by measuring the similarity between images, represents the similarity between the image generated by the learner and the correct image.

[0087] The sentence similarity obtained by measuring the similarity between the sentences generated by the learner and the correct sentence represents the semantic similarity between the sentences obtained through the method mentioned above.

[0088] And (1-α) represents the weight of image similarity, and 'α' represents the weight of sentence similarity. The 'α' value can be adjusted to control the relative importance of the two similarity metrics.

[0089] In this way, the final similarity assessment can be performed by combining the similarity between images and sentences. The above formula allows learners to independently evaluate and adjust the similarity of their generated content in terms of images and sentences.

[0090] Additionally, image similarity and sentence similarity each range from '0' to '1', with values ​​closer to 1 indicating greater similarity. In the above formula, the final similarity evaluation can be performed by combining the similarity values ​​of the image and sentence.

[0091] The similarity score and learner image generated by the above similarity evaluation unit are provided to the learner image display unit of the learner platform (terminal), and the learner can view the image and similarity score provided through the above learner image display unit and choose whether to generate the image again or perform speaking learning.

[0092] If the learner wants to generate an image again, the processing step of the learner sentence generation section described above can be returned.

[0093] When a learner selects speaking learning, the smart education platform (server-side) can generate corrected sentences based on the learner's spoken sentences and recommended sentences based on the learner's sentences, and deliver these to the learner platform. The sentence correction / recommendation unit, which generates the corrected and recommended sentences, can be configured as illustrated in Figure 6.

[0094] Referring to Fig. 6, the sentence correction / recommendation unit can transmit a correction sentence command for sentence correction and the learner's sentence to an external API processing unit (e.g., ChatGPT) to receive the corrected sentence. Furthermore, the unit can transmit a recommendation sentence command for generating recommended sentences and the learner's sentences to generate recommended sentences based on the learner's sentences. In this case, when generating recommended sentences to recommend sentences relevant to the learning content, the learner's sentences and the correct sentences can be transmitted together to provide more complete recommendations that are relevant to the learning problem.

[0095] Through the Speaking Learning section, learners can review their spoken sentences, corrected sentences, recommended sentences, and correct answers, and select sentences for speaking practice. They can record the selected sentences and send the audio file and the selected learning sentences to the Speaking Evaluation section. In other words, learners can review the sentences they have spoken through the Speaking Learning section. This includes their spoken sentences, corrected sentences, recommended sentences, and correct answers. Learners can then select the sentences they wish to practice speaking. They can then record the selected sentences and send the audio file and the selected learning sentences to the Speaking Evaluation section. In this way, learners can evaluate their pronunciation and speaking skills and receive feedback by pronouncing the selected sentences aloud.

[0096] The above speaking evaluation unit can provide an audio file and an answer file to a speaking fluency evaluation unit (artificial intelligence model), and generate learning results for speaking (fluency score, pronunciation accuracy score, rhythm, stress, speed score, etc.) and transmit them to the learner's evaluation screen management unit. That is, the speaking evaluation unit can receive the learner's audio file and answer file and utilize the artificial intelligence model to evaluate the fluency, pronunciation accuracy, rhythm, stress, speed, etc. of speaking. Thereafter, the evaluation results can be generated and sent to the learner's evaluation screen management unit, and since the evaluation results include various scores indicating the learner's speaking ability, the learner can check them and determine the direction of his or her own development.

[0097] Learners can receive their speaking assessment results through the assessment screen management section above. They can then receive feedback on these assessment results and choose whether to re-study for further learning, or end the course and save or share their results with others. Through this selection process, learners can determine the direction of their personal development and utilize their learning results as desired.

[0098] To save or share learning outcomes, a learner's device (e.g., a smartphone) can request registration with the smart education platform (server side). The personal content management unit within the smart education platform then receives the learner's personal content-related information and stores and manages images, sentences created by the learner, similarity scores, and learning dates in a database (DB). This ensures that the learner's learning outcomes and information are safely stored, accessible to the learner or shared with others when needed.

[0099] This invention can overcome existing limitations in the field of language education and provide learners with an innovative and effective learning experience. It can also contribute to revolutionizing the entire education system and helping learners improve their language skills.

[0100] Although all components constituting the embodiments of the present invention have been described above as being combined or operating in combination, the present invention is not necessarily limited to these embodiments. That is, within the scope of the purpose of the present invention, all components may be selectively combined and operated one or more times. In addition, terms such as "include," "comprise," or "have" described above, unless specifically stated to the contrary, mean that the corresponding component may be inherent, and therefore should be interpreted as including other components rather than excluding other components. All terms, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the art to which the present invention pertains, unless otherwise defined. Commonly used terms, such as terms defined in a dictionary, should be interpreted as being consistent with the contextual meaning of the relevant technology, and shall not be interpreted in an ideal or overly formal sense, unless explicitly defined in the present invention.

[0101] The above description is merely an illustrative illustration of the technical idea of ​​the present invention, and those skilled in the art will appreciate that various modifications and variations can be made without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate, rather than limit, the technical idea of ​​the present invention, and the scope of the technical idea of ​​the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within a scope equivalent thereto should be interpreted as being included within the scope of the rights of the present invention.

Claims

1. A smart education platform that provides first image content, evaluates the learning outcomes of learners based on the provided first image content, and provides the evaluated outcomes to learners; and A smart education system based on automatic generation of image content, including a learner platform that generates text or records voice based on the first image content provided from the smart education platform, transmits the generated text or voice to the smart education platform, and receives evaluation results for the text or voice.

2. In paragraph 1, The above smart education platform, A learning content management unit that generates sentences to be provided to learners and the first image content and delivers them to the learner platform; and A smart education system based on automatic generation of image content, including a similarity evaluation unit that evaluates the accuracy of text and voice generated by a learner based on the first image content transmitted to the learner platform.

3. In paragraph 2, The above smart education platform, It further includes a learner image generation unit that generates second image content based on text or voice generated by the learner. The above similarity evaluation unit is a smart education system based on automatic generation of image content that evaluates accuracy by comparing the first image content and the second image content.

4. In paragraph 1, The above smart education platform, A smart education system based on automatic generation of image content, comprising a sentence correction / recommendation section that provides the learner with a corrected sentence for a learner sentence generated based on text or voice generated by the learner, an expanded recommended sentence based on the learner sentence, and a correct sentence for the first image content.

5. In paragraph 4, A smart education system based on automatic generation of image content, which further includes a speaking evaluation section that evaluates the speaking fluency of a learner based on the selected sentence and the recording file, when the learner selects one of the sentences generated through the voice generated by the learner, the corrected sentence, the recommended sentence, and the correct sentence, and records the selected sentence as a voice.

6. In paragraph 1, The above smart education platform, A smart education system based on automatic generation of image content, including a personal content management unit that stores all matters performed through the above learner platform.

7. In paragraph 3, The above learner platform is, A learner sentence generation unit that generates text or voice created by the learner based on the first image content received from the learning content management unit and transmits it to the smart education platform; and A smart education system based on automatic generation of image content including a learner image display unit that receives evaluation results through the above similarity evaluation unit.

8. In paragraph 5, The above learner platform is, A speaking learning unit that checks the sentences created by the user, the above-mentioned corrected sentences, the above-mentioned recommended sentences, and the above-mentioned correct sentences, selects one sentence for speaking practice, and then records the selected sentence as a voice; and A smart education system based on automatic generation of image content, including an evaluation screen management section that receives feedback on evaluation results through the above-mentioned speaking evaluation section and provides guidance on whether to relearn, end learning, save learning results, and share learning results.

Images