Electric appliance specification generation method, storage medium, equipment and program product

Abstract

The invention provides an electric appliance specification generation method, a storage medium, equipment and a program product, and the method comprises the steps: for at least one operable component in a structure model of a target electric appliance, marking the component name of the operable component by using a first large language model according to the component type and / or visual feature of the operable component, using a second large language model to mark the functional state of the operable part; for one or more operable components, based on the component names and the functional states of the operable components, generating at least one operation task of the operable components by using a third large language model; and according to one or more of the component name, the function state and the operation task, generating at least part of the electric appliance specification in the document format by using a fourth language model. All large language models are organically matched in links of name labeling, state labeling, task generation and document arrangement, so that manual participation is effectively reduced, and the generation efficiency and accuracy of the electric appliance specification are greatly improved.

Description

Technical Field

[0001] The present application relates to the field of AIGC (Artificial Intelligence Generated Content) technology, and in particular to a method, storage medium, device and program product for generating an electrical appliance manual. Background Art

[0002] With the continuous enrichment and upgrading of smart appliances, users have higher and higher requirements for the manuals of electrical appliances. The relevant electrical appliance manuals are often manually written and drawn by designers, which is not only time-consuming and labor-intensive, but also may face difficulties in content updating and maintenance. In addition, due to the large number of electrical appliance models, complex functions and diverse user needs, the relevant manuals have many deficiencies in the accuracy of operating guidance.

[0003] Based on this, the present application provides a method, storage medium, device and program product for generating an electrical appliance manual to improve the relevant technology. Summary of the invention

[0004] The purpose of this application is to provide a method, storage medium, device and program product for generating an electrical appliance manual, which can effectively reduce manual participation and improve the efficiency and accuracy of generating an electrical appliance manual.

[0005] The purpose of this application is achieved by the following technical solutions:

[0006] In a first aspect, the present application provides a method for generating an electrical appliance manual, the method comprising: for at least one operable component in a structural model of a target electrical appliance, according to the component type and / or visual features of the operable component, using a first language model to label the component name of the operable component, and using a second language model to label the functional status of the operable component; for one or more operable components, based on the component name and functional status of the operable component, using a third language model to generate at least one operation task of the operable component; and according to one or more of the component name, the functional status and the operation task, using a fourth language model to generate at least part of the electrical appliance manual in a document format.

[0007] In some embodiments, the structural model comprises a CAD model; and / or, the operational task specifies component names and functional states in a parsable format.

[0008] In some embodiments, the method also includes: generating graphic information of the target appliance, the graphic information including one or more of a cover image, an overview image, a control panel image, and an operation guide image; generating at least part of the appliance manual in a document format using a fourth language model based on one or more of the component name, the functional status, and the operating task, includes: generating at least part of the appliance manual in a document format using the fourth language model based on one or more of the component name, the functional status, the operating task, and the graphic information.

[0009] In some embodiments, the process of generating the overview diagram includes: determining a posture with the largest total area of ​​operable components from a plurality of predefined candidate camera postures as a first rendering posture; rendering the structural model under the first rendering posture to obtain the overview image; and drawing the component names corresponding to the overview image in a dotted line annotation manner to generate the overview diagram.

[0010] In some embodiments, the generation process of the cover image includes: using a text-to-image generation model to add a background image to the overview image to generate the cover image; the corresponding space type of the background image is determined according to the appliance type of the target appliance.

[0011] In some embodiments, the generation process of the control panel diagram includes: clustering the operable components of the structural model based on geometric centers; determining, based on the clustering results, an area containing the operable components as a control panel; cropping a control panel image corresponding to the control panel from the overview image; and drawing the component names corresponding to the control panel image using dot-line annotation to generate the control panel diagram.

[0012] In some embodiments, the control panel icon includes a first control panel icon in RGB style and a second control panel icon in sketch style, and the step of drawing the component names corresponding to the control panel image in a dotted line annotation manner to generate the control panel icon includes: drawing the component names corresponding to the control panel image in a dotted line annotation manner to generate the first control panel icon in RGB style; performing edge detection on the first control panel icon in RGB style to create the second control panel icon in sketch style.

[0013] In some embodiments, the process of generating the operation guidance diagram includes: rendering the operable component in a second rendering posture to obtain a component image of the operable component; and drawing the operation guidance on the component image to obtain the operation guidance diagram.

[0014] In some embodiments, the second rendering pose is determined according to one or more of a component type, a visible surface area, and a joint axis of the operable component.

[0015] In some embodiments, the first large language model is a multimodal large language model; and / or, the second large language model is a large language model; and / or, the third large language model is a large language model; and / or, the fourth large language model is a large language model.

[0016] In some embodiments, generating at least part of the appliance manual in a document format using a fourth language model based on one or more of the component name, the functional status, and the operating task includes: using the fourth language model to process one or more of the component name, the functional status, and the operating task to generate a code for the appliance manual; and compiling the code for the appliance manual into a document format for the appliance manual.

[0017] In a second aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements any of the above methods when executed by a processor.

[0018] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements any one of the above methods when executing the computer program.

[0019] In a fourth aspect, the present application provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, any one of the above methods is implemented.

[0020] The present application provides a method, storage medium, device and program product for generating an electrical appliance manual, which uses large language models to work together to automatically generate high-quality electrical appliance manuals. Specifically, by analyzing the structural model of the target electrical appliance, at least one operable component is extracted from it, and the component name is annotated using the first large language model according to the component type and / or visual features of the operable component; on this basis, the functional state of the operable component is identified and described using the second large language model. Subsequently, based on the annotated component name and functional state, at least one feasible operation task description is formulated for the operable component using the third large language model. Afterwards, the above information (including one or more of the component name, functional state and operation task description) is automatically integrated into a document format according to the document layout logic through the fourth large language model, and can be further integrated with visual elements, tables and diagrams, thereby realizing the fully automated generation and presentation of the manual. In the embodiment of the present application, each large language model organically cooperates in the links of name labeling, state labeling, task generation and document layout, thereby effectively reducing manual participation and greatly improving the generation efficiency and accuracy of the electrical appliance manual. At the same time, with the help of multi-model information fusion and dynamic generation capabilities, not only can we quickly generate highly adaptable, unified format and detailed manuals for the complex structures of multiple categories of electrical appliances, but we can also flexibly select different texts, icons and layout styles, thereby providing users with more intuitive, more readable and more consistent instruction documents, greatly improving the quality of manual generation and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present application is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Figure 1 It is a flowchart of generating a manual reference data set provided in an embodiment of the present application.

[0023] Figure 2 It is a flowchart of a method for generating an electrical appliance manual provided in an embodiment of the present application.

[0024] Figure 3 It is a structural block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in this application to clearly and completely describe the technical solutions in the implementation mode of this application. Obviously, the described implementation mode is only a part of the implementation mode of this application, not all the implementation modes. Based on the implementation mode in this application, all other implementation modes obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0026] In the description of the embodiments of the present application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0027] See also Figure 1 , Figure 1 This is a flow chart of the generation of a manual check dataset provided by an embodiment of the present application. In order to create an open source manual dataset, the applicant designed a large model (e.g., a large language model) based on the home appliance CAD model to drive the manual check generation workflow. The generation workflow of the CheckManual dataset will be described in detail below. CAD stands for Computer Aided Design.

[0028] like Figure 1 As shown, in the leftmost part, the applicant analyzes real manuals to understand their formats and collects CAD models of appliances of different categories. The middle part shows the creation process of manual materials, including appliance creation, task generation, and graphic design. Each step can be manually verified to ensure correctness. Based on this information, the rightmost part generates appliance manuals with diverse formats through LATEX code.

[0029] Although some embodiments of this document are described using household appliances as an example, this application is not only applicable to household appliances, but also to other appliances other than household appliances. Although some embodiments of this document are described using manuals as an example, this application is not only applicable to generating manuals, but also to other instructions other than generating manuals. The reference manual generated above is an example of an appliance manual and should not be regarded as a limitation of this application.

[0030] The implementation methods of the present application are described in detail below.

[0031] See also Figure 2 , Figure 2 It is a flowchart of a method for generating an electrical appliance manual provided in an embodiment of the present application.

[0032] The embodiment of the present application provides a method for generating a manual of an electrical appliance, the method comprising steps S101 to S103.

[0033] Step S101: for at least one operable component in the structural model of the target electrical appliance, according to the component type and / or visual features of the operable component, use the first language model to label the component name of the operable component, and use the second language model to label the functional status of the operable component.

[0034] Step S102: for one or more operable components, based on the component name and functional status of the operable component, using a third language model to generate at least one operation task of the operable component.

[0035] Step S103: Generate at least part of the appliance manual in a document format using a fourth language model according to one or more of the component name, the functional status and the operation task.

[0036] In some embodiments, the first large language model may be a multimodal large language model (MLLM).

[0037] In some embodiments, the second large language model may be a large language model (ie, Large Language Model, LLM).

[0038] In some embodiments, the third largest language model may be a large language model.

[0039] In some embodiments, the fourth largest language model may be a large language model.

[0040] In some embodiments, the target appliance may be a household appliance.

[0041] In some embodiments, the target appliance is a microwave oven, a dishwasher, a refrigerator, or a coffee machine.

[0042] Before generating appliance manuals, real appliance manuals can be analyzed to understand the format of real manuals and collect available detailed appliance CAD models for manual creation.

[0043] Exemplarily, the applicant collected 110 home appliance manual PDFs from the Internet, which covered 11 categories of home appliances involved in the data set (i.e., the manual reference data set). Then, each manual was manually analyzed from three aspects: functional component annotation type, functional component operation guide, and task expression format. The applicant found that the functional component can be explained in a dot-line annotation format. These lines connect the points on the home appliance components and the functional names or IDs of the components (i.e., functional components). The instructions for the operation of the functional components are expressed, for example, in the following format: Plain text description, such as "Turn the temperature knob 90° clockwise". Close-up of the operated component (i.e., the operated functional component), such as the closed and open states of the red microwave door. Component motion decomposition diagram, showing the dynamic changes of the component (i.e., functional component) during the motion process. Taking the opening of the white blender lid as an example, the motion trajectory of the component contact point, for example, is to move the blue blender slider from right to left. In addition, the operation tasks in the real manual can be described in the form of text, bulleted lists, ordered lists, and single-column or multi-column tables.

[0044] In the above embodiments, the structural model may be, for example, a three-dimensional structural model. In some embodiments, the structural model may include a CAD model. The above embodiments do not limit the method for obtaining the CAD model. In some embodiments, the CAD model in the PartNet-Mobility Dataset may be used. Taking home appliances as an example, the home appliance CAD model may come from the PartNet-Mobility Dataset, which contains common categories of home appliance CAD models, including microwave ovens, dishwashers, and refrigerators, and the home appliance CAD models in the data set are annotated with operable parts. The applicant has collected a total of 11 categories and 182 CAD models, covering home appliances widely used in daily life.

[0045] As described above, in some embodiments, the operable component may include one or more of a button, a knob, a lever, a slider, a drawer, a door, a lid, and a container.

[0046] It should be noted that, in actual applications, since some functional components are operable and some functional components are inoperable, the operable components may include some or all functional components.

[0047] For example, in step S101, using the CAD model, a specific electrical appliance, such as a household appliance, can be created by labeling each operable component with a specific component name (e.g., a temperature knob) and a corresponding functional state (e.g., 60° clockwise: 120°C, 120° clockwise: 160°C).

[0048] Considering that the component function name (i.e., component name) is closely related to the appliance layout and the visual features of the component, the labeling process can be guided by MLLM. Before labeling, the component identification (e.g., component name) and type of the controllable component (i.e., operable component) are automatically marked on the overview image of the appliance, for example, by a dotted line labeling method (i.e., dotted line annotation method). Using these figures as visual cues, MLLM can be guided to design the functional name (i.e., component name) of each component. As an example, the multimodal large model considers the appliance type and carefully analyzes the visual information (e.g., visual features) of the component to create reasonable labeling.

[0049] In some embodiments, the method further includes: when the number of annotated components does not match the total number of operable components of the target electrical appliance, re-executing step S101. That is, when the number of annotated components does not match the total number of operable components (for example, the two are not equal), a regeneration mechanism will be triggered. By parsing the predictions of the MLLM using regular expressions, a dictionary that records the one-to-one relationship between component IDs and corresponding component names can be obtained. As an example, the dictionary may include the following: 0. Lens; 1. Capture Button; 2. Power Button. Such a dictionary can be used to generate an electrical appliance. By repeating the above annotation process, one or more electrical appliances can be created from a CAD model.

[0050] In order to mark the functional state of the operable components, the available states are first sampled according to the type (i.e., component type) of the operable components (i.e., operable components). For buttons, the available number of presses can be randomly sampled within a predefined numerical range as available states. For knobs and levers, several rotation angles can be randomly sampled at intervals of 60° (or, 30°, 45°, 90°) as available states. Other operable components (i.e., operable components) such as doors, lids, and containers can be assigned "open" and "closed" as available states. Through the component function name (i.e., component name) and the available states, the LLM can be guided to design specific functions for one or more available states. In this way, each operable component on the appliance can be labeled with one or more functional states. As an example, the labeling results of the functional state of an operable component may include the following.

[0051] "Capture Button":{

[0052] "Press 1time":"take a photo",…

[0053] "Press 5times":"activate burst mode"}……

[0054] Among them, the component name of the operable component is Capture Button, and the two functional states of the operable component include: ["Press 1time":"take a photo"] and ["Press 5times":"activate burstmode"], Press 1time means pressing once, take a photo means taking a photo, Press 5times means pressing 5 times, and activate burst mode means activating the continuous shooting mode.

[0055] As an example, an operable component has five available states, and two of the available states are selected for functional design, thereby obtaining the labeling results of the two functional states of the operable component.

[0056] The annotation results of large models (including MLLM and LLM) can be manually verified. In the annotation of part names, MLLM sometimes fails to recognize tiny visual elements on the CAD model, such as drawn button logos or scales, which may cause the annotated function name to not match the visual appearance of the part. In the annotation of function status, LLM may create counterintuitive designs, such as "Oven temperature knob 500℃". Therefore, the part names and function status annotated in large models can be checked through the human verification website of the design to correct the erroneous situations.

[0057] In actual applications, the applicant has found that the manual generally includes several representative operation tasks to introduce how to use the appliance. These operation tasks are also the core of the manual-based operation evaluation. As an example, executable operation tasks can be created based on the component names and functional states of the appliance.

[0058] In some embodiments, the operational tasks may include short-term operational tasks and / or long-term operational tasks.

[0059] In some embodiments, the operational tasks may specify component names and functional states in a parsable format.

[0060] As an example, LLM is used to propose short-term and long-term operation tasks for each appliance based on the component names and functional states of the operable components on the appliance. The operation steps corresponding to each operation task specify the component name and functional state in a parsable format, such as "1. Front door, open", "2. Power button, press once", etc. As an example, in order to improve task quality and reduce duplication, LLM only needs to design a new task based on the created tasks in one reasoning.

[0061] During the task creation process, in order to ensure the correctness of the evaluation, after LLM completes the task generation, the operation steps of each operation task can be manually checked to adjust the unreasonable step sequence and correct the wrong steps.

[0062] In some embodiments, the method may further include: generating graphic information of the target appliance, the graphic information including one or more of a cover image, an overview image, a control panel image, and an operation guide image; generating at least part of the appliance manual in a document format using a fourth language model based on one or more of the component name, the functional status, and the operating task, including: generating at least part of the appliance manual in a document format using the fourth language model based on one or more of the component name, the functional status, the operating task, and the graphic information.

[0063] According to the applicant's analysis of the real manual, the cover image, overview image, control panel image and component operation guide (ie, operation guide image) are the main images in the manual. The generation process of these images will be described below.

[0064] In some embodiments, the generation process of the overview diagram may include: determining a posture with the largest total area of ​​operable components from a plurality of predefined candidate camera postures as a first rendering posture; rendering the structural model under the first rendering posture to obtain the overview image; and drawing the component names corresponding to the overview image in a dotted line annotation manner to generate the overview diagram. For example, a set of candidate camera postures may be predefined to capture electrical appliances from different perspectives, and the perspective with the largest total area of ​​operable components (i.e., the first rendering posture) may be selected to render the overview image. Furthermore, the overview diagram may display the functional names of the operable components on the appliance. As an example, the overview image may be an RGB image.

[0065] In some embodiments, the generation process of the cover image may include: using a text-to-image generation model to add a background image to the overview image to generate the cover image; the corresponding space type of the background image is determined according to the appliance type of the target appliance.

[0066] As an example, the cover image of a manual can show the appearance and typical usage scenarios of the appliance. In order to synthesize a custom usage scenario, a background image (e.g., an indoor background image) can be added to the overview image of the appliance through a text-to-image generation model (e.g., an SDXL-Lightning model). Among them, the spatial type of the background image (e.g., room type) can be selected according to the appliance category (i.e., appliance type). For example, an oven is usually placed in the kitchen, while a washing machine is usually arranged in the laundry room. Therefore, the background image corresponding to the oven is, for example, a kitchen scene image, and the background image corresponding to the washing machine is, for example, a laundry room scene image.

[0067] In some embodiments, the generation process of the control panel diagram may include: clustering the operable components of the structural model based on geometric centers; determining, based on the clustering results, an area containing the operable components as a control panel; cropping a control panel image corresponding to the control panel from the overview image; and drawing the component names corresponding to the control panel image using dot-line annotation to generate the control panel diagram.

[0068] In the above embodiment, the control panel diagram can show the function names of the operable components on the electrical appliance. For example, before annotation, the operable components (e.g., buttons and knobs) are first clustered according to the geometric center. The clustering algorithm can use the DBSCAN algorithm. If clustered buttons and knobs are found, the area containing them is selected as the control panel and cut out from the overview image as a close-up of the control panel (i.e., the control panel image). Then, a dotted line annotation method is used to draw a control panel diagram according to the component function name (i.e., the component name).

[0069] Considering that most real manuals present overview diagrams and control panel diagrams in the form of line drawings, edge detection can be performed on the RGB style control panel diagram (e.g., RGB normal map) to create sketch style overview diagrams and control panel diagrams. In this way, these two RGB style and sketch style diagrams can be generated.

[0070] In some embodiments, the overview icon may include a first overview icon in RGB style and a second overview icon in sketch style, and the step of drawing the component names corresponding to the overview image in a dotted line annotation manner to generate the overview icon may include: drawing the component names corresponding to the overview image in a dotted line annotation manner to generate the first overview icon in RGB style; performing edge detection on the first overview icon in RGB style to create the second overview icon in sketch style.

[0071] In some embodiments, the control panel icon may include a first control panel icon in RGB style and a second control panel icon in sketch style, and the step of drawing the component names corresponding to the control panel image in a dotted line annotation manner to generate the control panel icon may include: drawing the component names corresponding to the control panel image in a dotted line annotation manner to generate a first control panel icon in RGB style; performing edge detection on the first control panel icon in RGB style to create a second control panel icon in sketch style.

[0072] In some embodiments, the process of generating the operation guidance diagram may include: rendering the operable component in a second rendering posture to obtain a component image of the operable component; and drawing the operation guidance on the component image to obtain the operation guidance diagram.

[0073] In some embodiments, the second rendering pose may be determined according to one or more of a component type, a visible surface area, and a joint axis of the operable component.

[0074] Component operation guidance diagrams (i.e., operation guidance diagrams) can introduce the physical interaction methods of the operable components of the electrical appliance. In order to generate such diagrams, for example, the posture of rendering the operable component (i.e., the second rendering posture) can be determined based on the component type and the visible surface area. In addition, the corresponding rendering postures of some operable components can be pre-set. For example, for buttons, knobs, levers, and drawers, the camera axis can be rendered parallel to their joint axes. For other operable components, including doors, sliders, lids, and containers, they are rendered in a posture where the camera axis is perpendicular to their joint axes while ensuring that the visible surface area is maximized. Then, an operation guidance diagram is drawn on the rendered component image to show how to operate them. With reference to the common representations in the real manual, the above-mentioned embodiments use the above-mentioned four different strategies to draw the operation guidance diagrams. In order to generate the operation guidance diagrams, the applicant manually annotated the 3D positions of one or more contact points of each operable component in the CAD model of the electrical appliance, and calculated its motion trajectory based on the joint axis to which the operable component is connected.

[0075] When rendering the overview image of the appliance based on the largest visible surface area, the pose (i.e., the first rendering pose) may not be suitable for viewing each component. In addition, the sketch-style second control panel illustration may be missing lines or contain unusual shadows. Therefore, the control panel illustration may be manually verified to trigger regeneration when these errors occur.

[0076] In some embodiments, generating at least part of the appliance manual in a document format using a fourth language model based on one or more of the component name, the functional status, and the operating task may include: using the fourth language model to process one or more of the component name, the functional status, and the operating task to generate a code for the appliance manual; and compiling the code for the appliance manual into a document format for the appliance manual.

[0077] In some embodiments, the use of the fourth language model to process one or more of the component name, the functional state, and the operation task to generate the code for the appliance manual may include: using the fourth language model to process the component name, the functional state, the operation task, and the graphic resource to generate the code for the appliance manual. The graphic resources may include, for example, fragile reminder icons, temperature warning icons, environmental protection icons, etc. In order to increase diversity, cover images, charts, and operation guide icons of different formats may be randomly extracted from the graphic part to add to the graphic resources.

[0078] In some embodiments, the document format may be in PDF format. In other embodiments, the document format may be in DOC format.

[0079] In some embodiments, the code of the appliance manual may be LaTeX code. LaTeX is a document programming language that provides the dual advantages of standardization and flexibility. Not only can the same document style be maintained between pages, but different types of graphics and tables can also be easily inserted into the document through LaTeX code. The large language model has a powerful code generation capability. As an example, the above embodiment uses LLM to generate the LaTeX code of the appliance manual, and then compiles it into the appliance manual in PDF format.

[0080] As an example, a set of LaTeX packages (e.g., multirow, xcolor, caption) can be predefined, some commonly used icons (e.g., fragile reminder icons, temperature warning icons, environmental protection icons) can be prepared as graphic resources, and then cover images, charts, and operation guide icons in different formats can be randomly extracted from the graphic part to add to the graphic resources, and a text style (e.g., text, bullet list, table) can be randomly selected from the real manual for task description. During the generation process, the functional status, operation tasks, and image paths (i.e., the paths of the above graphic resources) in text form are synthesized into prompts and input into the large language model, which is required to continue to generate a part of LaTeX code based on the previous content. In order to reduce the generation of uncompilable code, a compilation mechanism based on chapter granularity can be adopted during the compilation process. The newly generated LaTeX code will be appended to the end of the previous correct code and compiled together. If there is an error in the code compilation, regeneration will be triggered. Figure 1 An example of a manual from the CheckManual dataset is shown in .

[0081] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the above methods is implemented.

[0082] An embodiment of the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, any of the above methods is implemented.

[0083] The computer program product may be a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the computer program product of the present application is not limited thereto, and the computer program product may be any combination of one or more computer readable media.

[0084] An embodiment of the present application further provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements any of the above methods when executing the computer program.

[0085] See also Figure 3 , Figure 3 It is a structural block diagram of a computer device provided in an embodiment of the present application.

[0086] The embodiments of the present application do not limit the computer device, which may be, for example, a local computer device, a cloud computer device, a distributed computer device, etc.

[0087] The computer device may include: a memory 110, a processor 120, and a communication interface 130. The memory 110, the processor 120, and the communication interface 130 are connected via an internal connection path.

[0088] The memory 110 is used to store computer programs. In some implementations, the computer programs may include codes for implementing the methods of the embodiments of the present application.

[0089] The processor 120 is used to execute the computer program stored in the memory 110 to control the communication interface 130 to receive input data and information and output data such as operation results. In some implementations, when the solution of the embodiment of the present application is implemented by software or firmware, the computer program for implementing the solution of the embodiment of the present application can be stored in the processor 120 and executed by the processor 120.

[0090] The memory 110 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (Programmable ROM, PROM), an erasable programmable read-only memory (Erasable PROM, EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory may be a random access memory (RAM). It should be noted that the memory 110 described herein is intended to include, but is not limited to, any memory of these and other suitable types. As an example, the memory 110 includes a random access memory (RAM), a cache memory and a read-only memory (ROM). Among them, the memory 110 stores a computer program, and the computer program can be executed by the processor 120 so that the processor 120 implements the steps of any of the above methods.

[0091] The processor 120 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor 120 may be any conventional processor, etc.

[0092] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 120 or an instruction in the form of software. The method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in the processor 120 for execution. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 110, and the processor 120 reads the information in the memory 110 and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0093] In some implementations, in addition to the hardware units described above, the computer device may also include software modules, where the software modules may be, for example, an operating system, a basic input output system (BIOS), application software, etc.

[0094] The operating system is used to manage the hardware and / or software resources of computer devices and is the core and cornerstone of computer devices. The operating system needs to handle basic tasks such as managing and configuring memory, determining the priority of system resource supply and demand, controlling input and output devices, operating networks, and managing file systems. In order to facilitate user operations, most operating systems will provide an operating interface for users to interact with the system.

[0095] BIOS is used to run hardware initialization during the power-on boot phase and provide runtime services for the operating system and application programs. In some implementations, BIOS can also monitor and display processor temperature and execute functions such as adjusting temperature protection strategies.

[0096] Application software is also called application program, which can be understood as software written for a specific application purpose of the user. It is one of the main categories of computer software. For example, application software can be a program used to achieve power control, temperature management, etc.

[0097] It should be understood that the specific examples in this specification are only intended to help those skilled in the art to better understand the implementation methods of the present application, rather than to limit the scope of protection of the present application.

[0098] It is understood that in the various implementations of this specification, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this application.

[0099] It can be understood that the various implementation modes described in this specification can be implemented individually or in combination, and this application is not limited to this.

[0100] Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as those generally understood by those skilled in the art of the technical field of this specification. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the scope of this specification. The term "and / or" used in this specification includes any and all combinations of one or more related listed items. The singular forms of "a", "above", and "the" used in this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0101] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this specification.

[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of each embodiment described above can refer to the corresponding process in other embodiments and will not be repeated here.

[0103] In the several embodiments provided in this specification, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the technical solution of the present application.

[0105] In addition, each functional unit in each embodiment of the present specification may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0106] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this specification, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of this specification. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0107] The above are only specific implementations of this specification, but the protection scope of this application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in this specification, which should be included in the protection scope of this specification. Therefore, the protection scope of this application should be based on the protection scope of the claims.

Claims

1. A method for generating an electrical appliance manual, characterized in that: The method comprises: For at least one operable component in the structural model of the target electrical appliance, according to the component type and / or visual features of the operable component, use the first language model to label the component name of the operable component, and use the second language model to label the functional state of the operable component; For one or more operable components, based on the component name and functional status of the operable component, using a third language model to generate at least one operation task of the operable component; At least part of the appliance manual in a document format is generated using a fourth language model according to one or more of the component name, the functional status and the operation task.

2. The method for generating an electrical appliance manual according to claim 1, characterized in that: The structural model comprises a CAD model; and / or, the operation task specifies component names and functional states in a parsable format.

3. The method for generating an electrical appliance manual according to claim 1, characterized in that: The method further includes: generating graphic information of the target electrical appliance, the graphic information including one or more of a cover image, an overview image, a control panel image, and an operation guide image; The method of generating at least part of the appliance manual in a document format using a fourth language model based on one or more of the component name, the functional status, and the operating task includes: generating at least part of the appliance manual in a document format using the fourth language model based on one or more of the component name, the functional status, the operating task, and the graphic information.

4. The method for generating an electrical appliance manual according to claim 3, characterized in that: The generation process of the overview diagram includes: From a plurality of predefined candidate camera poses, determine a pose with a maximum total area of ​​an operable component as a first rendering pose; Rendering the structural model in the first rendering posture to obtain the overview image; The component names corresponding to the overview image are drawn in a dotted line annotation manner to generate the overview diagram.

5. The method for generating an electrical appliance manual according to claim 4, characterized in that: The generation process of the cover image includes: A background image is added to the overview image using a text-to-image generation model to generate the cover image; a corresponding space type of the background image is determined according to the appliance type of the target appliance.

6. The method for generating an electrical appliance manual according to claim 4, characterized in that: The generation process of the control panel diagram includes: Clustering the operational components of the structural model based on geometric centers; According to the clustering result, determining an area containing the operable components as a control panel; cropping a control panel image corresponding to the control panel from the overview image; The component names corresponding to the control panel image are drawn in a dotted line annotation manner to generate the control panel diagram.

7. The method for generating an electrical appliance manual according to claim 6, characterized in that: The control panel diagram includes a first control panel diagram in RGB style and a second control panel diagram in sketch style, and the method of drawing the component names corresponding to the control panel image in a dotted line annotation manner to generate the control panel diagram includes: Draw the component names corresponding to the control panel image in a dotted line annotation manner to generate a first control panel icon in RGB style; Edge detection is performed on the first control panel illustration in RGB style to create a second control panel illustration in sketch style.

8. The method for generating an electrical appliance manual according to claim 3, characterized in that: The generation process of the operation guidance diagram includes: rendering the operable component in a second rendering posture to obtain a component image of the operable component; The operation guide is drawn on the component image to obtain the operation guide diagram.

9. The method for generating an electrical appliance manual according to claim 8, characterized in that: The second rendering posture is determined according to one or more of a component type, a visible surface area, and a joint axis of the operable component.

10. The method for generating an electrical appliance manual according to claim 1, characterized in that: The first largest language model is a multimodal large language model; and / or, the second largest language model is a large language model; and / or, the third largest language model is a large language model; and / or, the fourth largest language model is a large language model.

11. The method for generating an electrical appliance manual according to claim 1, characterized in that: The step of generating at least part of the appliance manual in a document format using the fourth language model according to one or more of the component name, the functional state and the operation task comprises: Using the fourth language model to process one or more of the component name, the functional status and the operation task to generate a code for the appliance manual; The code of the electrical appliance manual is compiled into the electrical appliance manual in a document format.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 11 is implemented.

13. A computer device, characterized in that: The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 11 when executing the computer program.

14. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 11 is implemented.

Images