Systems and methods for generating immersive e-learning content

Abstract

Systems and methods for generating and presenting e-learning content are provided. The method includes receiving user input defining a plurality of lessons parameters, generating a core lesson outline comprising a plurality of lesson sub-topics selected based on the lesson parameters, generating sub-topic lesson data for each lesson sub-topic, generating a structured data file comprising a dynamic lesson graph and the sub-topic lesson data associated with nodes of the corresponding lesson sub-topics in the dynamic lesson graph and processing the structured data file to present the sub-topic lesson data via at least one presentation layer, the presentation layer being configured to provide a user interface allowing a user to navigate through the dynamic lesson graph and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

Description

[0001] SYSTEMS AND METHODS FOR GENERATING IMMERSIVE

[0002] E-LEARNING CONTENT

[0003] TECHNICAL FIELD

[0004] The present disclosure generally relates to the field of digital content curation and in particular to methods and systems for aggregating, generating and presenting e-leaming content to create immersive e-leaming experiences.

[0005] BACKGROUND

[0006] E-leaming refers to the use of digital technologies and the internet to facilitate education and training. It encompasses a variety of methods, including online courses, virtual classrooms, self-paced learning, and interactive digital platforms. By enabling learners to access educational materials and resources remotely, e-leaming provides flexibility and convenience, allowing individuals to study at their own pace and from any location. This approach is used in various fields, from academic institutions to corporate training, offering a scalable and accessible way to deliver education and skill development.

[0007] However, the Information age has led to a generative world overflowing with content, facts, opinions, and other data. It has become increasingly difficult to manage the vast amount of available data to find trustworthy information and allow discovery and exploration thereof without getting lost in the noise. This is particularly relevant in the field of e-leaming, where educators and students need the proper tools to manage and explore vast data to learn about the past, the present and help shape the future.

[0008] In addition, traditional modes of education are becoming rapidly outdated. Such modes are often confined to flat two-dimensional environments and unable to take advantage of the increasing amount of data. The limited immersive experiences that are available are cost prohibitive, expensive to produce, and often limited to specific platforms and devices.

[0009] There is therefore a need for methods and systems that allow managing and connecting the world’s wealth of knowledge and data, to generate and present immersive e-leaming content in a seamless and discoverable manner. SUMMARY

[0010] In the present disclosure, systems, methods, and novel processes for generating and presenting e-leaming content are provided. In the context of the present disclosure, e-leaming content refers to any material designed to inform, instruct, or facilitate e-leaming with respect to a particular topic or subject. As will be described in greater detail herein after, e-leaming content can take various forms, including written text, videos, podcasts, interactive activities, or visual aids, and is often created with the purpose of enhancing knowledge, developing skills, or encouraging critical thinking. The present technology may also structure and curate large and disparate stores of content to generate e-leaming content that suits the target audience’s age, expertise, and learning objectives.

[0011] As will be described herein after, the present technology may be used for generating, curating and presenting e-leaming content using 3D spatial technologies to create immersive e-leaming experiences. However, any system variation configured to enable generating and presenting digital content in various contexts, such as video gaming, professional training or any other suitable scenario, can be adapted to execute embodiment of the present technology, once teachings presented herein are appreciated.

[0012] In a first aspect, the present technology provides a computer-implemented method for generating and presenting e-leaming content. The method includes receiving user input defining a plurality of lessons parameters. The plurality of lesson parameters includes a student comprehension level, a subject or field of study, and an overall lesson topic. The method also includes generating a core lesson outline comprising a plurality of lesson sub-topics selected based on the lesson parameters and mapped as nodes in a dynamic lesson graph. The method also includes, for each lesson sub-topic, generating sub-topic lesson data by prompting a first large language model (LLM) to generate explanatory educational text for the sub-topic and identifying a plurality of media assets representative of the sub-topic by using the sub-topic and / or the explanatory educational text as keywords to query at least one database indexing media assets of a plurality of media types, and selecting resulting media assets based on an assigned relevance score indicative of a relevance of the media assets relative to the sub-topic. The method also includes generating a structured data file comprising the dynamic lesson graph, and the sub-topic lesson data associated with the nodes of the corresponding lesson subtopics in the graph and processing the structured data file to present the sub-topic lesson data via at least one presentation layer, the presentation layer being configured to provide a user interface allowing a user to navigate through the dynamic lesson graph, and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

[0013] In a second aspect, the present technology provides a system for generating and presenting e- leaming content. The system includes a conductor module comprising a processor and memory storing instructions which, when executed, cause the conductor module to receive user input defining a plurality of lessons parameters. The plurality of lesson parameters includes a student comprehension level, a subject or field of study, and an overall lesson topic. Execution of the instructions also causes the conductor module to generate a core lesson outline comprising a plurality of lesson sub-topics selected based on the lesson parameters and mapped as nodes in a dynamic lesson graph; generate, for each lesson sub-topic, sub-topic lesson data by prompting a first LLM to generate explanatory educational text for the sub-topic and identifying a plurality of media assets representative of the sub-topic by using the sub-topic and / or the explanatory educational text as keywords to query at least one database indexing media assets of a plurality of media types, and selecting resulting media assets based on an assigned relevance score indicative of a relevance of the media assets relative to the sub-topic. Execution of the instructions also causes the conductor module to generate a structured data file comprising the dynamic lesson graph, and the sub-topic lesson data associated with the nodes of the corresponding lesson sub-topics in the graph. The system also includes a presentation module comprising a processor and memory storing instructions which, when executed, cause the presentation module to process the structured data file to present the sub-topic lesson data via at least one presentation layer, the presentation layer being configured to provide a user interface allowing a user to navigate through the dynamic lesson graph, and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

[0014] BRIEF DESCRIPTION OF THE DRAWINGS

[0015] For a better understanding of the implementations described herein and to show more clearly how they may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings which show at least one exemplary implementation.

[0016] FIG. 1 is a schematic representation of a system for generating and presenting e-leaming content according to an embodiment. FIG. 2 is a flowchart illustrating a method for generating and presenting e-leaming content according to an embodiment.

[0017] FIG. 3 is a schematic representation of a dynamic lesson graph according to an embodiment.

[0018] FIG. 4 is a schematic representation of dynamic lesson graph according to another embodiment.

[0019] FIG. 5 is a schematic representation of a structured data fde including the dynamic lesson graph of FIG. 3, according to an embodiment.

[0020] FIGS. 6-10 are schematic representations of portions of a dynamic lesson graph as a user navigates therethrough, in accordance with an embodiment.

[0021] DETAILED DESCRIPTION

[0022] It will be appreciated that, for simplicity and clarity of illustration, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements or steps. In addition, numerous specific details are set forth in order to provide a thorough understanding of the exemplary implementations described herein. However, it will be understood by those of ordinary skill in the art that the implementations described herein may be practised without these specific details. In other instances, well-known methods, procedures and components have not been described in detail so as not to obscure the implementations described herein. Furthermore, this description is not to be considered as limiting the scope of the implementations described herein in any way but rather as merely describing the implementation of the various implementations described herein.

[0023] One or more systems described herein may be implemented in computer program(s) executed on processing device(s), each comprising at least one processor, a data storage system (including volatile and / or non-volatile memory and / or storage elements), and optionally at least one input and / or output device. “Processing devices” encompass computers, servers and / or specialized electronic devices which receive, process and / or transmit data. As an example, “processing devices” can include processing means, such as microcontrollers, microprocessors, and / or CPUs, or be implemented on FPGAs. For example, and without limitation, a processing device may be a programmable logic unit, a mainframe computer, a server, a personal computer, a cloud based program or system, a laptop, a personal data assistant, a cellular telephone, a smartphone, a wearable device, a tablet, a video game console or a portable video game device.

[0024] Each program is preferably implemented in a high-level programming and / or scripting language, for instance an imperative e.g., procedural or object-oriented, or a declarative e.g., functional or logic, language, to communicate with a computer system. However, a program can be implemented in assembly or machine language if desired. In any case, the language may be a compiled or an interpreted language. Each such computer program is preferably stored on a storage media or a device readable by a general or special purpose programmable computer for configuring and operating the computer when the storage media or device is read by the computer to perform the procedures described herein. In some embodiments, the system may be embedded within an operating system running on the programmable computer.

[0025] Furthermore, the system, processes and methods of the described embodiments are capable of being distributed in a computer program product comprising a computer readable medium that bears computer-usable instructions for one or more processors. The computer-usable instructions may also be in various forms including compiled and non-compiled code.

[0026] The processor(s) are used in combination with storage medium, also referred to as “memory” or “storage means”. Storage medium can store instructions, algorithms, rules and / or trading data to be processed. Storage medium encompasses volatile or non-volatile / persistent memory, such as registers, cache, RAM, flash memory, ROM, diskettes, compact disks, tapes, chips, as examples only. The type of memory is of course chosen according to the desired use, whether it should retain instructions, or temporarily store, retain or update data. Steps of the proposed method are implemented as software instructions and algorithms, stored in computer memory and executed by processors.

[0027] In the following disclosure, various modules will be described. As can be appreciated, such modules can be implemented as part of one or more of the above-described computer program(s) and / or processing device(s). In some embodiments, the modules can be provided as part of one or more non-transitory computer-readable media containing instructions, which when executed cause a processing device to implement the functionality of the modules.

[0028] The present technology aims, in one aspect, at providing a platform for spatial (i.e three- dimensional) discovery-based e-leaming experiences. In use, a user is provided with an enjoyable and engaging way to learn known concepts, view rich multimedia, and discover concepts in a rich 3D environment. As will be described in greater detail herein after, these learning experiences may be curated, customized, and delivered in a cost and time efficient manner.

[0029] In one aspect of the present technology, a system provides a conductor module configured to select and assemble media assets from various and disparate stores of content to generate a dynamic lesson graph for a given topic / subject that includes sub-topics mapped as nodes, with each sub-topic being associated with corresponding media assets. The system also provides an interface enabling a first user, such as a teacher, to customize the structured data file. The customization options include, but are not limited to, modifying nodes, edges, labels, media assets associated with the sub-topics and associated data attributes. Such customization may be carried out during a set-up phase of an immersive e-leaming experience. The system further provides a presentation module for processing the structured data file and allowing a second user, such as a student, to navigate through nodes of the dynamic lesson graph following predefined or adaptive pathways and be presented with media assets associated with the subtopics (e.g., through an immersive 3D environment). The system is configured to dynamically adjust the lesson graph based on user interactions, enabling the second user to explore new nodes, uncover related content, and engage in a guided learning experience. Such exploration can be carried out during an execution phase of the immersive e-leaming experience.

[0030] In the context of the present disclosure, a device can correspond to an electronic device, such as a computing device that may comprise a personal computer, a laptop, a smartphone, a tablet, among others. In some implementations, a device includes various hardware components including one or more single or multi-core processors, a solid-state drive, random-access memory (RAM), persistent memory and an input / output interface.

[0031] In some implementations, functionalities of a device may be distributed amongst multiple systems. As a person in the art of the present technology may appreciate, multiple variations as to how a device is implemented may be envisioned without departing from the scope of the present technology.

[0032] Communication between the various components of a device may be enabled by one or more internal and / or external buses (e.g. a PCI bus, universal serial bus, IEEE 1394 “Firewire” bus, SCSI bus, Serial-ATA bus, ARINC bus, etc.), to which the various hard-ware components are electronically coupled. The input / output interface may provide networking capabilities such as wired or wire-less access. As an example, the input / output interface may comprise a networking interface such as, but not limited to, one or more network ports, one or more network sockets, one or more network interface controllers and the like. Multiple examples of how the networking interface may be implemented will become apparent to the person skilled in the art of the present technology. For example, but without being limitative, the networking interface may implement specific physical layer and data link layer standard such as Ethernet, Fibre Channel, Bluetooth, NFC or Wi-Fi.

[0033] Further, a device may include a screen or display. In some implementations, the display may comprise and / or be housed with a touchscreen to permit users to input data via some combination of virtual keyboards, icons, menus, or other Graphical User Interfaces (GUIs). In some embodiments, the display may be implemented using a Liquid Crystal Display (LCD) display or a Light Emitting Diode (LED) display, such as an Organic LED (OLED) display. A device may be, for example and without being limitative, a handheld computer, a personal digital assistant, a cellular phone, a network device, a camera, a smart phone, an enhanced general packet radio service (EGPRS) mobile phone, a network base station, a media player, a navigation device, an e-mail device, a game console, or a combination of two or more of these data processing devices or other data processing devices.

[0034] In the context of the present disclosure, the term “media asset” refers to digital or physical resources used to create, enhance, or distribute content across various platforms. These assets include a wide range of formats, such as image fdes, video fdes, audio fdes, graphics, animations, three-dimensional (3D) objects, 3D point clouds, quizzes and documents.

[0035] In the context of the present disclosure, a “lesson graph” is a data structure used to represent relationships between lesson sub-topics. A lesson graph may be represented as nodes that represent the lesson sub-topics and links that connect pairs of nodes to show the relationships or interactions between them. A lesson graph may be stored in computer memory (e.g., Random Access Memory or persistent memory such as a database) as an adjacency matrix where rows and columns represent nodes, and each entry indicates whether an edge exists between two or more nodes, an adjacency list, and edge list or an incidence matric under the form of a .CSV or .JSON fde. Broadly speaking, the lesson graph may be a simplified knowledge graph that is derived from a more complex knowledge graph and / or from a plurality of more complex knowledge graphs of content, to provide pathways for exploring content according to topic-based relationships. For example, a plurality of complex knowledge graphs can be provided to index content according to various interrelated parameters. The lesson graph can be thought of as a specialized super-graph that relates the plurality of complex knowledge graphs by defining links between content mapped in those graphs based on how that content relates to different topics.

[0036] The user may navigate through a given lesson graph by interacting with a user interface. Broadly speaking, the user interface includes, in the context of the present technology, a visual layout, commands, input devices, and feedback mechanisms that enable the user to transmit user instructions, navigate through a given lesson graph, and access information efficiently. A user interface may include a Graphical User Interface (GUI) in the form of a 2D and / or 3D spatial immersive experience. It is appreciated, however, that in other implementations the user interface can be provided in the form of a Command Line Interface (CLI), a Voice User Interface (VUI), a Touch User Interface and / or a Device Motion Tracked Interface and / or a Hand Tracked Interface and / or a Gaze Based Interface.

[0037] With reference to FIG. 1, a system 10 for generating and presenting immersive e-leaming content is shown according to an embodiment. The system 10 includes a back-end server 100, in operative communication with at least one user device 200 and a plurality of data stores 50 storing media assets. Broadly described, the back-end server 100 includes a conductor module 110 in operative communication with an interaction module 112 and a media asset database 300. The conductor module is configured to generate e-leaming content in the form of a structured data file 118. For example, the structured data file 118 may be implemented as a .YAML or a .JSON data file . The e-leaming content is generated based on user instructions received from the user device 200 via interaction module 112, and based on media assets from the plurality of data stores 50 indexed by the database 300. The user device 200 is operable to receive the structured data file 118, and to process the same via a presentation module 230 to present the e-leaming content to a user.

[0038] Although the conductor module 110 is depicted as communicably connected to the database 300 on FIG. 1, it is appreciated that the conductor module 110 may interface with other sources of data, data stores or APIs such as knowledge graphs and / or LLM APIs as will be described in greater detail herein after. In alternative implementations, the conductor module 110 may be directly implemented in the user device 200. In yet other implementations, the functionality of the conductor module 110 may be distributed and may be partly implemented on the back-end server 100 and the user device 200 concurrently.

[0039] In some implementations, the system 10 may include a plurality of user devices, similar to the user device 200 such that different users may concurrently interact with a same e-leaming content and be collectively part of a shared immersive e-leaming experience. In this scenario, one of the user devices can be flagged as a host user device onto which the conductor module 110 is executed. The outputs of the conductor module 110 may be shared with the other user devices, which adjust a presentation thereof to provide individual experience for each of the users.

[0040] In the illustrated embodiment, the user device 200 is a desktop computer. It should be appreciated, however, that in other embodiments, the user device 200 can be implemented as other types of computing and / or processing devices, such as a laptop computer, tablet, smartphone, augmented reality device, virtual reality device, a handheld computer, a personal digital assistant, a cellular phone, a network device, a camera, a media player, a game console, or a combination of two or more of these processing devices or other processing devices etc.

[0041] The user device 200 includes a communications interface configured to provide networking capabilities such as wired or wireless access. As an example, the communications interface may include a networking interface that may implement specific physical layer and data link layer standards such as Ethernet, Fibre Channel, Wi-Fi or Token Ring. The communications interface may use the networking interface to allow operative communication between the back-end server 100 and data stores 50 via a local or wide area network, such as over the internet.

[0042] Further, the user device 200 includes an I / O module 210 for receiving input from, and providing input to, a user. The I / O module 210 can include screen or display capable of rendering a user interface. For example, the display may render images, including 3D images, videos, audio recordings 3D point clouds, Augmented Reality (AR) images, program outputs, etc. As will be described in greater detail hereinafter, the I / O module 210 may be used to display e-leaming content based on the structured data fde 118 received from the conductor module 110. In some embodiments, the I / O module 210 may include a touchscreen to permit users to input data via some combination of virtual keyboards, icons, menus, or other Graphical User Interfaces (GUIs).

[0043] In some embodiments, the I / O module 210 may include a display implemented using a Liquid Crystal Display (LCD) display or a Light Emitting Diode (LED) display, such as an Organic LED (OLED) display. In other embodiments, the display may be remotely communicatively connected to the user device 200 via a wired or a wireless connection, so that e-leaming content may be displayed at a location different from the location of the user device 200. In this configuration, I / O module 210 may include a display that is operationally coupled to, but housed separately from, other functional units and systems in the user device 200.

[0044] Upon receiving user input via interaction module 112, the back-end server 100 is configured to generate a structured data file 118 based on the user input and media assets indexed by the database 300 and selected based on the user input. Operation of the back-end server 100 including the conductor module 110 will be described in more detail hereinafter. In the illustrated embodiment, the back-end server 100 is implemented as a single server, but it is appreciated that in other embodiments, the functionality of the server 100 may be distributed and may be implemented via multiple virtual and / or physical servers.

[0045] In operation, the conductor module 110 is configured to query the database 300 for media assets 51 based on the user input received via interaction module 112. The database 300 may index any structured or unstructured collections of media assets stored on or more data stores 50. The database 300 may reside on the same hardware as the server 100 or it may reside on separate hardware, such as a dedicated server or plurality of servers. In the illustrated embodiment, the data stores 50 are external data stores comprising third party content accessible via one or more application data interfaces (APIs). It is appreciated, however, that in some embodiments, at least some of the data stores can be located on the server 100.

[0046] In the illustrated implementation, the database 300 is configured to index media assets of a plurality of extendable media types such as image files, video files, audio files, graphics, animations, three-dimensional (3D) objects, 3D point clouds and gaussian splatters, quizzes, and documents. The database 300 is a graph database that indexes each media asset in association with corresponding data attributes indicative of a content of the media asset, and according to different organizations or relationships with other media assets. For example, the data attributes may include a title of the media asset, a creator / author of the media asset, a file type of the media asset, a textual description of the media asset, keywords and / or tags indicative of a topic or a theme of the media asset, etc. In this implementation, the data attributes of a given media asset may also include the following attributes: “HAS ADMIN SCORE FOR KEYWORD” that includes a human admin relevance score relative to a specific keyword for the given media asset, “HAS CV SCORE FOR KEYWORD” that includes a computer generated relevance score relative to a specific keyword for the given media asset, “CONTAINED_IN” including information about a folder in which the given media asset is located and information about a layer corresponding to the given media asset within a video project files and composition files of a digital visual effects and motion graphics software such as AFTER EFFECTS from ADOBE, "EPISODE SCRIPT FOR" including indications of an episode script related to the given media asset within a video project file and / or a composition file, "AE PROJECT FOR" including indications of an episode script (e.g., an AFTER EFFECTS project) based on the given media asset, "HAS FS PARENT AEP" including indications of an Adobe Illustrator file related to the given media asset and used in a video project file and / or a composition file, "DERIVED FROM" including indication of an original source file of the media asset, "EPISODE SCENE FOR" including indications of a layer of a video project file and / or a composition file for which the given media asset is a base for video media assets, "AE LAYER FOR" including indications a layer of a video project file and / or a composition file for which the given media asset is a base for text media assets, "CAPTION FOR" including indications an episode script scene of a video project file and / or a composition file for which the given media asset is a base for caption media assets, and "VIDEO FOR" including indications that the given media asset is the final result for a given video project file and / or a composition file for video media assets.

[0047] Different media assets can be indexed as part of a knowledge graph, for example where media assets relating to similar topics, themes, authors, video project file and / or a composition file etc., are related. The database 300 can be queried via an API to locate relevant media assets (for example using a keyword or other search parameters) and can output a list of media assets ranked by relevance. Although a single database 300 is illustrated in the present embodiment, it is appreciated that in other embodiments the conductor module 110 can be configured to query a plurality of databases configured to index different types of media assets and / or different collections of media assets across different data stores 50. In some implementations, the database 300 is communicably connected with servers or other databases via local or wide area networks, such as over the internet, and ingest new media assets therefrom. For example, the database 300 may access media assets from websites (e.g. WIKIPEDIA), online streaming or video platforms (e.g., YOUTUBE), etc.

[0048] With additional reference now to FIG. 2, a method 400 for generating and presenting e-leaming content using the above-described system 10 will be described in accordance with an embodiment. In one or more aspects, the method 400 or one or more steps thereof may be performed by one or more processors and / or one or more computers (such as the back-end server and user device) part of system 10. The method 400 or one or more steps thereof may be embodied in computer-executable instructions that are stored in a computer-readable medium, such as a non-transitory mass storage device, loaded into memory and executed by a CPU. Some steps or portions of steps in the flow diagram may be omitted or changed in order.

[0049] The method 400 starts with receiving, at operation 410, a user input defining a plurality of lessons parameters. In operation, a user associated with the user device 200, such as a teacher or a lesson planner, may operate I / O module 210 to interact with the back-end server 100 via the interaction module 112 to provide input that defines the plurality of lesson parameters. In the present embodiment, the lesson parameters include a student comprehension level (e.g “grade 9”, “grade 12”, “first year university”), a subject or field of study (e.g., “Chemistry”, “Astronomy”, “Artificial Intelligence”), and an overall lesson topic (e.g., “Elements and the Periodic Table”, “Celestial Bodies”), although it is appreciated that in other embodiments, other lesson parameters and / or combinations of lesson parameters can be provided.

[0050] As can be appreciated, the user input can define the lesson parameters using different techniques. For example, the interaction module 112 may implement a Large Language Model (LLM) configured to prompt the user via conversation through the I / O module 210 to determine the target comprehension level. In some embodiments, the interaction module 112 may implement trained artificial intelligence (Al) algorithms to identify the comprehension level from conversational input provided by the user. For example, the user input can comprise conversational text from which the Al algorithms are trained to identify and / or extract a corresponding comprehension level. In some implementations, the LLM chatbot may simply prompt the user to select a student comprehension level from a predefined list of comprehension levels. The interaction module 112 may similarly implement a LLM to determine the subject or field of study, and an overall lesson topic from the user input. For example, the interaction module 112 can implement a LLM chatbot configured to prompt the user to select the subject or field of study from a pre-defined list of subjects or fields of study. The LLM can also be configured to converse with the user to determine the target lesson topic by allowing the user to suggest a topic and / or by suggesting possible topics if none are provided by the user.

[0051] In some implementations, the LLM chatbot may prompt the user for the subject or field of study corresponding to subjects or fields of study relating to the media assets indexed by the database 300. The LLM chatbot may also prompt the user for the overall lesson topic corresponding to lesson topics relating to the media assets indexed by the database 300. For example, the subjects or fields of study and / or the lesson topics may be stored as a knowledge graph in the database 300, and the LLM chatbot may query the knowledge graph and generate corresponding prompts and responses via Retrieval-Augmented Generation (RAG) or other suitable techniques. In an implementation, the knowledge graph can be representative of a course curriculum, such that the user can be prompted for fields of study and / or lesson topics that are limited to those defined in the curriculum.

[0052] The method 400 continues with generating, at operation 420, a core lesson outline. The core lesson outline is generated by the conductor module 110 by selecting a plurality of lesson subtopics based on the lesson parameters, and mapping the selected lesson sub-topics as nodes in a lesson graph. In an embodiment, the lesson parameters can be provided as input to the conductor module 110, and the conductor module 110 is configured to query a LLM to propose sub-topics based on the comprehension level, the subject of study, and the overall lesson topic. For example, the input provided to the conductor module 110 can include a comprehension level of “Grade 9”, a subject of study of “Astronomy” and an overall lesson topic of “Celestial Bodies”, and the proposed sub-topics can include: “Stars; Planets; Moons; Asteroids; Comets”. As can be appreciated, the LLM can propose a plurality of sub-topics, and the conductor module 110 can be configured to select a predetermined number of sub-topics for building the lesson graph, such as 3-5 sub-topics. In some embodiments, lesson sub-topics can be stored as a knowledge graph or as part of a recommendation database such as elasticSearch or openSearch, and the LLM can be configured to generate responses comprises sub-topic recommendations by utilizing RAG to query the knowledge graph and / or recommendation databases. Following selection of sub-topics, the conductor module 110 is configured to map the selected sub-topics as nodes in a lesson graph. An exemplary lesson graph 500 is depicted in FIG. 3. In this example, the selected sub-topics are mapped as a sequence defining a linear graph, but alternative structures are possible. As can be appreciated, the lesson graph 500 generated in this manner defines the core lesson outline that users will be able to initially navigate and explore. The number of sub-topics included in the core lesson plan may vary, for example according to the overall lesson topic. As can be further appreciated, a given sub-topic may be associated with one or more secondary sub-topics that may be represented in the lesson graph as nodes that are linked to the given sub-topic.

[0053] For example, FIG. 4 depicts a lesson graph 500A for which each node 510 corresponding to sub-topics is associated with four secondary sub-topics, mapped as nodes 520. More specifically, the node 510 of the sub-topic “Stars” is associated with nodes 520 of secondary sub-topics “Constellation”; “Red giant”; “Supernova” and “White dwarf’. The nodes 510 form an initial “Lesson Path” for the user. When exploring a node 510 corresponding to a sub-topic of the core lesson plan (e.g., “Planets”) a user may decide to navigate to one of the secondary sub-topics associated with the current node (e.g., “Red giant”). The lesson graph 500A can be characterized as being “dynamic” in that the conductor module 110 can be configured add and / or remove nodes as the user navigates the lesson graph. For example, if a user chooses to navigate to a secondary sub-topic, the conductor module 110 can generate a new lesson section built around that secondary sub-topic, for example by generating further sub-topics that are related to the secondary sub-topic and that are mapped as nodes that branch from that secondary sub-topic in the lesson graph. Examples of dynamically modifying a lesson graph during user exploration will be described in more detail hereinbelow.

[0054] In FIG. 4, tertiary sub-topics that are related to the secondary sub-topic but that expand beyond the overall lesson topic are also depicted. This encourages the discovery-based learning exploration, and provides new opportunities to find unique connections between topics that are not typically within the same curriculum. This may provide possibilities for the user to derive from the initial “Lesson Path” into a dynamic “Trailblazing” mode and navigate to secondary, tertiary and more distant nodes, thereby forming a personalized lesson path.

[0055] In some implementations, the lesson graph may be manually constrained to avoid exploration that is too far off the initial subject / topic. For example, a teacher may define a limit on the number and / or depth of additional sub-topics that can be dynamically generated from the initial lesson graph. In some implementations, additional exploration can be provided as a reward. For example, a minimal number and / or set of lesson-essential nodes can be defined, and additional nodes will be generated in the lesson graph only for users (e.g., students) who have visited each of the lesson-essential nodes.

[0056] Once the core lesson outline is generated in the form of a lesson graph, the method 400 continues with generating, at operation 430, sub-topic lesson data sub-topics in the lesson graph. In an embodiment, the lesson data can include educational text, educational media, and assessments for each sub-topic, although it is appreciated that other lesson data is also possible. The lesson data can be generated by the conductor module 110 through generative techniques and / or by searching for and selecting appropriate content from existing data stores.

[0057] Educational text can correspond to explanatory text to explain a sub-topic to a user at a grade- appropriate level. In an embodiment, the conductor module 110 can be configured to generate the educational text by prompting a LLM to generate explanatory educational text for a subtopic at the comprehension level defined in the lesson parameters. In some embodiments, the core lesson outline can be provided to the LLM as part of the prompt, for example to provide context such that the text between each sub-topic flows more logically. In some implementations, the educational text can include notes visible only to a given user. For example, the educational text can include notes that are only visible to a teacher to guide the teacher in presenting content and providing a desired e-leaming experience for students. Such notes can be manually provided by the teacher and / or can be generated by an LLM, for example by prompting the LLM to generate a lesson presentation plan.

[0058] E-leaming content can include various types of media assets that can assist in explaining and / or otherwise presenting aspects of the sub-topic. The e-leaming content can be generated for a sub-topic by the conductor module 110 by identifying a plurality of media assets representative of the sub-topic in the database 300. For example, the conductor module 110 can query the database 300 to obtain a plurality of potentially relevant media assets and select the media assets that are considered to be the most relevant. In some embodiments, the conductor module 110 may use an API to search / query the data-base 300 for media assets 51 using different search parameters. A resulting list of media assets 51 can be received by the conductor module 110, with each media asset being assigned a corresponding relevance score by the search / query that is indicative of the relevance of the media to that specific sub-topic or query. The conductor module 110 can subsequently select media assets 51 based on their assigned relevance scores (e.g., by selecting a predetermined number of media assets having the higher relevance score). In some embodiments, the conductor module 110 can select media assets while favoring a variety of media types, for example selecting at least one video, at least one 3D model, at least one audio file, etc., to accommodate different learning types of the user. In such embodiments, the conductor module 110 can be configured to select media assets having the highest relevance score from each of a predetermined set of media types.

[0059] As can be appreciated, different search parameters can be used to query the database 300 for relevant media assets. For example, the database 300 can be queried using the sub-topic and / or the explanatory educational text generated for the sub-topic as keywords. The relevance score associated with located media assets can be based on a variety of different factors.

[0060] In some embodiments, the relevance score can be at least partially based on a human admin score. Such a score can, for example, be manually assigned by a human operator to indicate how strongly a media asset relates to a specific keyword.

[0061] In an embodiment, the relevance score can be at least partially based on an LLM score. Such a score can be generated by prompting an LLM to indicate how strongly a media asset relates to a specific keyword. In some implementations, a plurality of candidate media assets can be provided to the LLM along with the sub-topic and / or the educational text generated for the subtopic, in order to prompt the LLM to choose which media assets are the most suitable for a lesson directed to the sub-topic. Where the media asset comprises a visual component, the LLM can employ computer vision to recognize objects in the media asset and assist with determining how strongly the media asset relates to a keyword or educational text.

[0062] In an embodiment, the relevance score can be at least partially based on knowledge graph distance score. Such a score can be generated by measuring a distance between a media asset and the nearest node that matches the keyword in a knowledge graph.

[0063] In an embodiment, the relevance score can be at least partially based on a similarity between the media asset and the sub-topic. For example, such a score can be generated by calculating a vector distance between a first vector representing data attributes of the media assets and a second vector representing the sub-topic.

[0064] In an embodiment, the relevance score can be at least partially based on a similarity between the media asset and user preferences. In an implementation, the user preferences can comprise lesson preferences explicitly defined by the user. For example, a student or a teacher may indicate that they prefer visual content over audio content. In an implementation, the user preferences can be defined based on historic user input. In particular, historic user input received via interaction module 112 can be recorded, and such historic input can be used to rank results such that they are tuned to the preferences of a specific user.

[0065] For example, the historic user input can correspond to manual adjustment of a selection of media assets, such as a teacher performing a manual adjustment of lesson graph and / or media assets included therein when constructing a lesson with the help of conductor 110 during a setup phase. Future media assets can then be ranked such that they more closely resemble media assets preferred by the teacher as indicated through their manual adjustment. As another example, the historic user input can correspond to previous interactions with media assets, also referred to as “active object selection triggers”, such as a student navigating the lesson graph and interacting with media assets included therein during an execution phase of a lesson. In some implementations, the historic user input also includes information about passive viewport observation frustum directionality. More precisely, the frustum is a truncated pyramid shape that defines the field of view of a virtual camera corresponding to the field of view of the user upon being provided by the media assets in a 3D environment. It is used to determine which parts of a 3D scene are visible to the camera and thus should be rendered. The historic user input may thus include information about elements seen and observed by the user during the immersive 3D e-leaming experience.

[0066] Future media assets can then be ranked such that they more closely resemble media assets that the student prefers to interact with. In an implementation, the user preferences can be stored as a vector, and the relevance score can be generated by calculating a vector distance between a first vector representing data attributes of the media asset and a second vector corresponding to the user preferences vector. In some implementations, reinforced learning can be carried out using the user preferences vector to tune results to the specific user.

[0067] Assessments can correspond to questions or tests (i.e., “quizzes”) to confirm a student’s understanding of a sub-topic. In an embodiment, the conductor module 110 can be configured to generate questions or tests based on the educational text and / or the e-leaming content generated for the sub-topic. In some implementations, the assessments can be procedurally generated. For example, the conductor module 110 can be configured to generate a plurality of true or false questions based around different media assets 51 associated with the sub-topic. In particular, the conductor module 110 may identify a given media asset from among the media assets selected for the sub-topic, and a description of the given media asset may be swapped with the description of another one of the media assets. The true or false question can be formulated by asking the user “Does this media match this description?”. A similar process can be repeated for the remaining media assets selected for the sub-topic to generate a plurality of questions. In some implementations, the assessments can be generated with the help of a LLM. For example, the conductor module 110 can be configured to prompt the LLM to generate questions relating to one or more media assets associated with the sub-topic. The prompt can include the lesson educational text and topics to provide contextual data to the LLM.

[0068] In some embodiments, generating the lesson data may also include generating text-to-speech audio for at least some of the explanatory text and text associated with at least some of the selected media assets 310, the assessment questions, and media descriptions associated with the selected media assets 51. The text-to-speech audio utterances may be generated by selecting, by the conductor module 110, a text-to-speech voice from a plurality of text-to- speech voices based at least in part on the student comprehension level. The conductor module 110 may also identify one or more words or phrases that are important to the e-leaming content based on the lesson sub-topic, and apply vocal emphases in the text-to-speech audio utterances to the one or more words and / or phrases containing the one or more words.

[0069] In some embodiments, generating the lesson data can also include generating environment media that can be used to decorate an immersive environment in which the sub-topic lesson data can be rendered. The conductor module 110 can be configured to identify environment media representative of the sub-topic by using the sub-topic and / or the explanatory e-leaming text as keywords to query at least one database indexing environmental media, and select resulting media assets based on an assigned relevance score indicative of a relevance of the environmental media relative to the sub-topic. For example, in response to the field of study being “Ancient Egypt”, the conductor module 110 may select a background image and / or animations related to the Pyramids, pharaohs etc. to be displayed as a background of the selected media assets during the execution phase of a lesson.

[0070] As can be appreciated, operation 430 can be repeated for each of the sub-topics in the lesson graph. In some implementations, operation 430 can also be performed for one or more secondary sub-topics included in the lesson graph. In some implementations, operation 430 can be repeated each time a new node is added to the lesson graph, for example in response to a user navigating to the secondary sub-topic, and nodes for new sub-topics being generated. The method 400 continues with generating, at operation 440, a structured data file 118 comprising the dynamic lesson graph, and the sub-topic lesson data associated nodes of respective subtopics in the dynamic lesson graph. FIG. 5 depicts an illustrative structured data file 500B generated from the dynamic lesson graph 500. More specifically, the structured data file 500B includes the nodes of the dynamic lesson graph 500, and sub-topic lesson data for each of the sub-topic. In this example, the node “Stars” is associated with explanatory text Tl, media assets Ml, M2, M3 and M4, and assessment question QI, the node “Planets” is associated with explanatory text T2, media assets M5, M6 and M7, and assessment question Q2, the node “Moons” is associated with explanatory text T3, media assets M8 and M9, and assessment question Q3, the node “Asteroids” is associated with explanatory text T4, media assets M10, Mi l, M12 and M13, and assessment question Q4, and the node “Comets” is associated with explanatory text T5, media assets M14, M15, M16, M17 and M18, and assessment question Q5. Although not illustrated, it is appreciated that each node can also be associated with its corresponding environmental media, text-to-speech audio, and / or any other generated subtopic lesson data. The structured data file can include any information that is needed to allow rendering an environment that enables a student to explore sub-topics in a lesson, and to allow rendering or otherwise presenting sub-topic lesson data. For example, it can include copies of sub-topic lesson data, and / or references that allow accessing the sub-topic lesson data (e.g., such as URLs that allows accessing the various media assets). As can be appreciated, the structured data file can be generated in a presentation agnostic data format such that the same data file can be used to render or play back the lesson through various different presentation forms, such as a 3D presentation, 2D presentation, rendered video, dynamic video, etc.

[0071] In some embodiments, the method can include allowing a user to modify the structure data file in order to generate a customized lesson. For example, a teacher can operate user device 200 to provide input to the conductor module 110 via the interaction module 112. Such input can include instructions to modify the lesson graph 500, for example by adding, removing, or changing nodes and / or by adding, removing, or changing relationships between nodes. The input can further include instructions to modify sub-topic lesson data associated with nodes, for example to add, remove, or change media assets associated with a node, etc. In response to the lesson graph 500 being modified, an updated structured data file 118 can be generated. In some implementations, the media attributes of the selected media assets 310 are adjusted upon retrieval of the selected media assets 310 from the database 300. For example, a language (e.g. the lexicon) of the textual description of the media asset, keywords and / or tags indicative of the topic or the theme of the media asset may be adjusted based on the student comprehension level determined at operation 410. For example, the description associated with an image may be simplified in response to the student comprehension level being “grade 6” compared to “grade 9”.

[0072] Once the structured data file is generated 118, the method 400 continues with operation 450 where the structured data file is processed to present the sub-topic lesson data via at least one presentation layer. In the present embodiment, the data file 118 is processed and the presentation layer is implemented via presentation module 230 of user device 200. The presentation module 230 provides a user interface allowing the user of the user device 200 to navigate through the dynamic lesson graph, and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

[0073] The presentation module 230 is configured to provide an intuitive and interactive user interface (UI) that allows the users to seamlessly navigate through the dynamic lesson graph and be provided by the sub-topic lesson data (e.g. the selected media assets) associated with each subtopic of the dynamic lesson graph upon navigating to a correspond node thereof. It can be said that each node thus represents a distinct lesson or learning activity, linked based on prerequisite knowledge or topic dependencies.

[0074] In an embodiment, the presentation layer is implemented as a web-based application, desktop software or as a native application designed for various platforms, including but not limited to mobile operating systems such as Android and iOS. In particular, the native application may be distributed through recognized app distribution platforms such as Google Play or the Apple App Store, enabling users to download and install the application directly onto their user devices. In use, the presentation layer provides the user with visual access to the dynamic lesson graph. The presentation layer can be implemented on the I / O module 210 of the user device 200, e.g. via a display. The user may interact on navigation controls via the I / O module 210 to access and navigate among nodes. When navigating to a node, the corresponding sub-topic lesson data can be rendered on the presentation layer. For example, the presentation module 230 can be configured to retrieve media assets 51 from corresponding data stores 50 (e.g., via references provided in the structured data file 118), and render such assets as part of the presentation layer. The interface may highlight certain paths or lessons that are recommended based on the user’s past performance or preferences.

[0075] The implementation of this presentation layer may utilize web technologies, such as HTML, CSS, and JavaScript, and their accompanying frameworks (such as React), for rendering static and / or dynamic user interfaces on a web browser page for example. In some embodiments, the presentation layer could incorporate augmented reality (AR) or virtual reality (VR) technologies to create an immersive, 3D lesson graph navigation experience. The flexibility of the presentation layer allows for seamless integration with various content types, such as text, video, and interactive exercises (e.g. assessment questions), ensuring that users can engage with lessons in the most effective format for their learning style.

[0076] As described above, the lesson graph can be dynamically updated as a user explores the lesson graph via the presentation module 230. FIGS. 6 to 10 depicts an illustrative scenario of dynamically updating a lesson graph in response to navigation of a user through nodes thereof. FIG. 6 shows a portion 600A of the dynamic lesson graph, the user being currently at the node of the sub-topic “Stars”. The user is thus provided with media assets associated with the subtopic “Stars”. FIG. 6A shows that a consecutive node of the node “Stars” in the dynamic lesson graph, or a primary sub-topic of the node “Stars”, is the node “Planets”. The node “Stars” also has two secondary sub-topics illustrated as nodes “Constellations” and “Supernovas”.

[0077] Broadly speaking, the conductor module 110 may generate one or more secondary sub-topics of a given sub-topic in response to the user (e.g., the student) navigating to the node of the given sub-topic. The one or more secondary sub-topics are mapped as nodes connected to the node of the given sub-topic in the dynamic lesson graph. The generation of sub-topic lesson data is further performed for each secondary sub-topics.

[0078] In this illustrative scenario, the user navigates to the secondary sub-topic “Constellations”, as shown on FIG. 7 depicting a portion 600B of the dynamic lesson graph. In response, the conductor module 110 generates three secondary sub-topics “Zodiac”, “Star navigation” and “Ancient Mythology” of the node “Constellations”. The sub-topics “Zodiac”, “Star navigation” and “Ancient Mythology” may also be referred to as tertiary sub-topics of the sub-topic “Stars”. It can thus be said that the dynamic lesson graph is dynamically adjusted based on a current navigation of the user through the dynamic lesson graph. Upon generating additional nodes, the conductor module 110 also generates corresponding sub-topic lesson data for each new sub-topic as described herein above.

[0079] In this illustrative scenario, the user navigates to the tertiary sub-topic “Star navigation”, as shown on FIG. 8 depicting a portion 600C of the dynamic lesson graph. In response, the conductor module 110 generates three quaternary sub-topics “Celestial Coordinate system”, “Maritime History” and “Ancient Mythology” of the node “Constellations”.

[0080] In this illustrative scenario, the user navigates to the quaternary sub-topic “GPS technology”, as shown on FIG. 9 depicting a portion 600D of the dynamic lesson graph. In response, the conductor module 110 generates three quinary sub-topics “Satellite systems”, “Triangulation” and “Earth’s rotation” of the node “GPS technology” and associates a connection sub-topic “Relative Theory” selected from the pre-defined list of sub-topics.

[0081] In this illustrative scenario, the user navigates to a node “Supernova” which a tertiary sub-topic of the connection node “Relative Theory”, as shown on FIG. 10 depicting a portion 600E of the dynamic lesson graph. In this implementation, the media assets associated with the node “Supernova” reached through the connection node “Relativity theory” may be different from the media assets associated with the node “Supernova” corresponding to the secondary subtopic “supernova” depicted on FIG. 6. It can thus be said the system 100 provide dynamic content adaptation where each node would contain specific content that adapts based on the path taken through the dynamic lesson graph.

[0082] In some implementations, the conductor module 110 can be configured to prune nodes in response to the user navigating through the lesson graph. For example, where a lesson graph includes a plurality of secondary sub-topic and the user navigates to a given one of the secondary sub-topics, the remaining secondary sub-topics can be removed from the dynamic lesson graph. A similar process can be carried out for tertiary, quaternary, and subsequent nodes to remove nodes that are not relevant to the user’s journey.

[0083] It is clear that a person skilled in the art can make various modifications and variations to this application without departing from the scope of this disclosure. This disclosure is intended to cover these modifications and variations of this application provided that they fall within the scope of protection defined by the following claims and their equivalent technologies.

Claims

CLAIMS1. A computer-implemented method for generating and presenting e-learning content, the method comprising: i) receiving user input defining a plurality of lessons parameters, the plurality of lesson parameters comprising: a student comprehension level, a subject or field of study, and an overall lesson topic; ii) generating a core lesson outline comprising a plurality of lesson sub-topics selected based on the lesson parameters and mapped as nodes in a dynamic lesson graph; iii) for each lesson sub-topic, generating sub-topic lesson data by: prompting a first large language model (LLM) to generate explanatory educational text for the sub-topic; and identifying a plurality of media assets representative of the sub-topic by using the sub-topic and / or the explanatory educational text as keywords to query at least one database indexing media assets of a plurality of media types, and selecting resulting media assets based on an assigned relevance score indicative of a relevance of the media assets relative to the sub-topic; iv) generating a structured data file comprising the dynamic lesson graph, and the subtopic lesson data associated with the nodes of the corresponding lesson sub-topics in the graph; and v) processing the structured data file to present the sub-topic lesson data via at least one presentation layer, the presentation layer being configured to provide a user interface allowing a user to navigate through the dynamic lesson graph, and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

2. The method of claim 1 , wherein the user input is received by prompting the user via an LLM chatbot.

3. The method of claim 2, wherein the LLM chatbot is configured to prompt the user for the student comprehension level from a predefined list of comprehension levels.

4. The method of claim 2, wherein the LLM chatbot is configured to prompt the user for the subject or field of study corresponding to subjects or fields of study relating to the media assets indexed in the at least one database.

5. The method of claim 4, wherein the subjects or fields of study relating to the media assets are stored as a knowledge graph, and the LLM chatbot is configured to query the knowledge graph and generate corresponding prompts and responses via Retrieval-Augmented Generation (RAG).

6. The method of claim 2, wherein the LLM chatbot is configured to prompt the user for the overall lesson topic corresponding to lesson topics relating to the media assets indexed in the at least one database.

7. The method of claim 6, wherein the lesson topics are stored as a knowledge graph, and the LLM chatbot is configured to query the knowledge graph and generate corresponding prompts and responses via RAG.

8. The method of claim 1, wherein the core lesson outline is generated by prompting a second LLM to generate the plurality of lesson sub-topics based on the lesson parameters.

9. The method of claim 8, wherein lesson sub-topics are stored as a knowledge graph, and the second LLM is configured to query the knowledge graph and generate corresponding responses via RAG.

10. The method of claim 1, wherein prompting the first LLM to generate the explanatory educational text for the sub-topics comprises providing the core lesson outline as part of the prompt.

11. The method of claim 1 , wherein the media types comprise at least one of: audio files, image files, video files, three-dimensional (3D) objects, 3D point clouds and quizzes.

12. The method of claim 1, further comprising recording lesson preferences indicative of historic user input indicative of manual adjustment of the selection of media assets, wherein the relevance score is further based on the lesson preferences.

13. The method of claim 1, wherein the relevance score is based at least in part on one or more of: a human admin score manually entered by a human operator, a LLM score generated by the first LLM based on comparison of the explanatory educational text for the sub-topic and descriptive data of the media assets, a first vector distance between a first vector representing the descriptive data of the media assets and a second vector representing the sub-topic, and a second vector distance between a third vector representing user data indicative of a user’s preference and the second vector.

14. The method of claim 1, wherein the at least one database indexes media assets in association with data attributes comprising: one or more keywords descriptive of a content of the media file, weights associated with one or more keywords, textual descriptions, metadata and video extracts.

15. The method of claim 1, further comprising recording navigation data indicative of historic user interactions with the dynamic lesson graph and lesson sub-topics; and wherein the relevance score of corresponding media assets is based at least in part on the recorded navigation data.

16. The method of claim 1, further comprising: generating one or more secondary sub-topics of a given sub-topic, the one or more secondary sub-topics being mapped as nodes connected to the node of the given sub-topic in the dynamic lesson graph; and repeating step iii) for each of the one or more secondary sub-topics.

17. The method of claim 16, wherein the generating the one or more secondary sub-topics is performed in response to the user navigating to the node corresponding to the given subtopic.

18. The method of claim 16, further comprising, upon navigation of the user to the node of a given secondary sub-topics, pruning nodes of the other secondary sub-topics in the dynamic lesson graph and generating one or more tertiary sub-topics of a given secondary subtopic, the one or more tertiary sub-topics being mapped as nodes connected to the node of the given secondary sub-topic in the dynamic lesson graph.

19. The method of claim 1, wherein the generating the lesson data further comprises generating assessment questions based on at least some of the selected media assets.

20. The method of claim 19, wherein the assessment questions are generated by prompting the first LLM to generate questions for the at least some of the selected media assets using a prompt that includes the explanatory text and the lesson sub-topics as contextual input.

21. The method of claim 19, wherein the generating the lesson data further comprises generating text-to-speech audio for at least one of the explanatory text, the assessment questions, and media descriptions associated with the selected media assets.

22. The method of claim 21, wherein generating the text-to-speech audio comprises selecting a text-to-speech voice from a plurality of text-to-speech voices based at least in part on the student comprehension level.

23. The method of claim 21, comprising identifying one or more words or phrases that are important to the e-learning content based on the lesson sub-topic, and wherein generating the text-to-speech audio comprises vocally emphasizing in the text-to-speech audio the one or more words and / or phrases containing the one or more words.

24. The method of claim 1, wherein the generating the lesson data further comprises identifying environment media representative of the sub-topic by using the sub-topic and / or the explanatory educational text as keywords to query at least one database indexing environmental media, and selecting resulting media assets based on an assigned relevance score indicative of a relevance of the environmental media relative to the sub-topic.

25. The method of claim 1, wherein generating the sub-topic lesson data further includes generating: environment media configured to be executed by the presentation layer and display a decorative environment upon rendering at least some of the sub-topic lesson data, assessment questions, and audio utterances by using a text-to-speech module provided to the user upon rendering at least some of the sub-topic lesson data.

26. A system for generating and presenting e-learning content, the system comprising: a conductor module comprising a processor and memory storing instructions which, when executed, cause the conductor module to: i) receive user input defining a plurality of lessons parameters, the plurality of lesson parameters comprising: a student comprehension level, a subject or field of study, and an overall lesson topic; ii) generate a core lesson outline comprising a plurality of lesson sub-topics selected based on the lesson parameters and mapped as nodes in a dynamic lesson graph; iii) for each lesson sub-topic, generate sub-topic lesson data by: prompting a first LLM to generate explanatory educational text for the subtopic; and identifying a plurality of media assets representative of the sub-topic by using the sub-topic and / or the explanatory educational text as keywords to query at least one database indexing media assets of a plurality of media types, and selecting resulting media assets based on an assigned relevance score indicative of a relevance of the media assets relative to the sub-topic; andiv) generate a structured data file comprising the dynamic lesson graph, and the subtopic lesson data associated with the nodes of the corresponding lesson sub-topics in the graph; and a presentation module comprising a processor and memory storing instructions which, when executed, cause the presentation module to process the structured data file to present the subtopic lesson data via at least one presentation layer, the presentation layer being configured to provide a user interface allowing a user to navigate through the dynamic lesson graph, and to render at least some of the sub-topic lesson data upon the user navigating to corresponding nodes of the dynamic lesson graph.

Images