Scene generation method, electronic equipment and storage medium

Abstract

The invention provides a scene generation method, electronic equipment and a storage medium, and relates to the technical field of text processing.The scene generation method comprises the steps that target scene description is generated based on original scene description input by a user; wherein the original scene description refers to fuzzy scene description, and the target scene description comprises key elements of the scene; and generating a target scene based on the target scene description. The scene generation efficiency and accuracy can be improved.

Description

Technical Field

[0001] This application relates to the technical field of text processing, and particularly to a scenario generation method, an electronic device, and a storage medium. Background Art

[0002] In the software development process, scenario generation is a key task, especially in professional fields such as the Internet of Vehicles (IoV). The IoV field involves communication between vehicles and vehicles, vehicles and infrastructure, as well as related data processing and service provision. Therefore, scenarios in professional fields need to be precise, comprehensive, and compliant with industry standards.

[0003] Currently, product managers need to manually write scenarios based on vague scenario ideas, combined with professional knowledge and historical data, which is not only time-consuming and laborious but also error-prone, resulting in low efficiency and accuracy in scenario generation. Summary of the Invention

[0004] This application provides a scenario generation method, an electronic device, and a storage medium to improve the efficiency and accuracy of scenario generation.

[0005] According to the first aspect of the embodiments of this application, a scenario generation method is provided, including:

[0006] Generating a target scenario description based on the original scenario description input by the user; wherein, the original scenario description refers to a vague scenario description, and the target scenario description includes key elements of the scenario;

[0007] Generating a target scenario based on the target scenario description.

[0008] According to the second aspect of the embodiments of this application, an electronic device is provided, including a memory and a processor;

[0009] The memory is connected to the processor and is used to store programs;

[0010] The processor is used to implement the scenario generation method as described in the first aspect by running the programs in the memory.

[0011] According to the third aspect of the embodiments of this application, a storage medium is provided, on which a computer program is stored, and when the computer program is run by a processor, it implements the scenario generation method as described in the first aspect.

[0012] In this application, based on the original scene description input by the user, a target scene description is generated. Here, the original scene description refers to a vague scene description, and the target scene description includes the key elements of the scene. Based on the target scene description, a target scene is generated. Compared with directly manually writing the scene according to the original scene description, in this application, the vague original scene description input by the user is first supplemented and improved to obtain a relatively standardized target scene description including the key elements of the scene, and then based on the target scene description, a target scene is generated, dividing a complex task into two relatively simple tasks, which can improve the efficiency and accuracy of scene generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the provided drawings.

[0014] Figure 1 It is a schematic flowchart of a scene generation method provided in an embodiment of the present application;

[0015] Figure 2 It is a schematic flowchart of step 101 provided in an embodiment of the present application;

[0016] Figure 3 It is a schematic flowchart of step 102 provided in an embodiment of the present application;

[0017] Figure 4 It is a schematic diagram of an exemplary AI scene assistant provided in an embodiment of the present application;

[0018] Figure 5 It is a schematic diagram of a page of a specific scene generation operation provided in an embodiment of the present application;

[0019] Figure 6 It is a schematic diagram of a page for selecting a similar scene after generating a scene provided in an embodiment of the present application;

[0020] Figure 7 It is a schematic diagram of a page for selecting "change another one" after generating a scene provided in an embodiment of the present application;

[0021] Figure 8 It is a schematic flowchart of step 102 provided in an embodiment of the present application;

[0022] Figure 9 It is a schematic flowchart of step 801 provided in an embodiment of the present application;

[0023] Figure 10 It is a schematic flowchart of a process for determining an example of scenario description transformation provided in an embodiment of the present application;

[0024] Figure 11 It is a schematic flowchart of step 102 provided in an embodiment of the present application;

[0025] Figure 12 It is a schematic structural diagram of a scenario generation device provided in an embodiment of the present application;

[0026] Figure 13 It is a schematic structural diagram of an electronic device provided in an embodiment of the present application. Detailed implementation manners

[0027] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0028] Exemplary implementation environment

[0029] In the process of software development, scenario generation is a key task, especially in professional fields such as the Internet of Vehicles. The Internet of Vehicles field involves communication between vehicles and vehicles, vehicles and infrastructure, as well as related data processing and service provision. Therefore, scenarios in professional fields need to be precise, comprehensive, and compliant with industry standards.

[0030] Currently, product managers need to manually write scenarios based on vague scenario ideas, combined with professional knowledge and historical data, which is not only time-consuming and laborious but also error-prone, resulting in low efficiency and accuracy of scenario generation.

[0031] Therefore, in order to improve the efficiency and accuracy of scenario generation, the inventors propose a solution of first supplementing and perfecting the vague original scenario description input by the user to obtain a relatively standardized target scenario description including key elements of the scenario, and then generating a target scenario based on the target scenario description. The following embodiments detail this solution.

[0032] The scenario generation method according to the embodiment of the present application can be executed by an electronic device such as a terminal device or a server. The terminal device can be a user device, a mobile device, a computing device, a wearable device, etc. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a cloud server capable of performing cloud computing. This method can be implemented by a processor calling computer-readable program instructions stored in a memory.

[0033] Exemplary method

[0034] See also Figure 1 In an exemplary embodiment, a scene generation method is provided. Figure 1 As shown, the process of the scene generation method mainly includes:

[0035] Step 101: Generate a target scene description based on an original scene description input by a user.

[0036] The original scene description refers to a fuzzy scene description, and the target scene description includes the key elements of the scene.

[0037] In an exemplary embodiment, the original scene description input by the user may refer to a vague scene description, which usually only contains some basic concepts or general directions and needs to be further refined and supplemented.

[0038] In an exemplary embodiment, the original scene description may be input by the user through voice or by typing.

[0039] In an exemplary embodiment, the original scene description can be obtained by obtaining a user's voice signal through voice input by the user, and performing voice recognition on the user's voice signal to obtain the original scene description. Specifically, performing voice recognition on the user's voice signal to obtain the original scene description can be implemented by a large voice recognition model.

[0040] In an exemplary embodiment, the target scene description may refer to a normative scene description.

[0041] In an exemplary embodiment, the target scene description may determine what content needs to be included according to the requirements of the historical scene library, that is, the generated target scene description may be directly added to the historical scene library.

[0042] In an exemplary embodiment, the key elements of a scene may include the scene name, scene type, stage, role, environment, appeal and event behavior. The key elements of a scene may also include other key elements, which are not limited in this application.

[0043] For example, the original scenario description entered by the user is: When an abnormality occurs in the vehicle computer, the operator can remotely send log collection instructions to the vehicle, and then collect data for fault analysis.

[0044] The corresponding target scenario is described as:

[0045] Scenario name: Remote log collection

[0046] Scenario type: Operational scenario

[0047] Stage: Operational

[0048] Role: Operator

[0049] Environment: When the in-vehicle infotainment system malfunctions

[0050] Request: The operator needs to remotely collect in-vehicle infotainment system logs for fault analysis

[0051] Event actions: 1. The operator selects the faulty vehicle on the operation platform; 2. Issues a log collection command; 3. The in-vehicle infotainment system receives the command and starts collecting logs; 4. The log data is uploaded to the operation platform.

[0052] To further improve the efficiency and accuracy of scenario generation, the following embodiments introduce a solution for automatically generating target scenario descriptions through large models. In some embodiments, as Figure 2 shown, step 101 includes:

[0053] Step 201, splice the original scenario description input by the user and a preset prompt word template to obtain a first prompt word.

[0054] Among them, the preset prompt word template includes the requirements for scenarios, output specifications, and sample cases in the vehicle networking field; the output specifications include the key elements of the scenario; the sample cases in the vehicle networking field include the original scenario description sample cases and target scenario description sample cases in the vehicle networking field.

[0055] For example, the output specifications may include: the text output includes the scenario name, scenario type, belonging stage, role, environment, request, and event actions.

[0056] In an exemplary embodiment, several typical cases of the company's vehicle networking can be used in the sample cases in the vehicle networking field, combined with the company's vehicle networking private domain knowledge.

[0057] Step 202, input the first prompt word into the scenario description generation large model to obtain the target scenario description.

[0058] Among them, the first prompt word is used to instruct the scenario description generation large model to generate the target scenario description based on the original scenario description input by the user, referring to the requirements for scenarios, output specifications, and sample cases in the vehicle networking field.

[0059] The preset prompt template includes the requirements for scenarios in the vehicle networking field, output specifications, and examples in the vehicle networking field. The output specifications include the key elements of the scenarios. The examples in the vehicle networking field include the original scenario description examples and target scenario description examples in the vehicle networking field. The preset prompt template is highly relevant to the vehicle networking field and can guide the scenario description generation large model to output target scenario descriptions that meet the requirements of the vehicle networking field. Through the scenario description generation large model, the automatic generation of target scenario descriptions can be realized. Compared with manually writing target scenario descriptions, it can improve the efficiency of generating target scenario descriptions, and thus improve the efficiency of scenario generation. Moreover, based on the original scenario description input by the user, referring to the requirements for scenarios in the vehicle networking field, output specifications, and examples in the vehicle networking field, the generation of target scenario descriptions can make the target scenario descriptions more in line with the requirements of the vehicle networking field, improve the accuracy of generating target scenario descriptions, and thus improve the accuracy of scenario generation.

[0060] If you are not satisfied with the generated target scenario description, you can also use the scenario description replacement instruction to replace it with other scenario descriptions. The following embodiments introduce the specific solutions for scenario description replacement. In some embodiments, based on the original scenario description input by the user, generating a target scenario description includes: based on the original scenario description input by the user, generating at least two candidate scenario descriptions at one time; wherein, the candidate scenario descriptions include the key elements of the scenario; and determining the target scenario description from each of the candidate scenario descriptions. The scenario generation method further includes: after receiving the scenario description replacement instruction, determining a candidate scenario description other than the target scenario description from each of the candidate scenario descriptions for display.

[0061] In an exemplary embodiment, based on the original scenario description input by the user, generating at least two candidate scenario descriptions at one time may include: based on the original scenario description input by the user, generating a fourth prompt; inputting the fourth prompt into the scenario description generation large model to obtain at least two candidate scenario descriptions; wherein, the fourth prompt is used to instruct the scenario description generation large model to generate at least two candidate scenario descriptions at one time based on the original scenario description input by the user.

[0062] In an exemplary embodiment, the fourth prompt may include the original scenario description input by the user, the requirements for scenarios in the vehicle networking field, output specifications, the number of candidate scenario descriptions to be output, and candidate output examples in the vehicle networking field; the output specifications include the key elements of the scenario; the candidate output examples in the vehicle networking field include the original scenario description examples and candidate scenario description examples in the vehicle networking field, and one original scenario description example corresponds to multiple candidate scenario description examples. According to the number of candidate scenario descriptions to be output and the candidate output examples in the vehicle networking field, the scenario description generation large model is guided to generate at least two candidate scenario descriptions at one time.

[0063] For example, based on the original scene description input by the user, three candidate scene descriptions are generated simultaneously at one time, namely candidate scene description A, candidate scene description B, and candidate scene description C. Candidate scene description A is selected from the three candidate scene descriptions for display. After receiving the scene description replacement instruction, one candidate scene description is selected from candidate scene description B and candidate scene description C for display. It can be candidate scene description B for display, or candidate scene description C for display.

[0064] In an exemplary embodiment, receiving the scene description replacement instruction can be detecting a click operation on the "Change" button, or receiving a scene description replacement instruction input by the user through voice, or receiving a scene description replacement instruction in other ways. This application does not limit this.

[0065] In the prior art, for "Change", after the input content is input into the large model, the large model generates one output content at a time. When the user clicks "Change", the large model regenerates new output content according to the input content, and the time for regenerating new output content is relatively long and the efficiency is relatively low. In this application, based on the original scene description input by the user, at least two candidate scene descriptions are generated at one time, and the target scene description is determined from each candidate scene description. After receiving the scene description replacement instruction, one candidate scene description other than the target scene description is determined from each candidate scene description for display. Multiple candidate scene descriptions are generated simultaneously at one time. After receiving the scene description replacement instruction, one candidate scene description other than the target scene description is selected from the multiple candidate scene descriptions that have been generated for display, without regenerating the scene description. Compared with the solution of regenerating new output content, the processing duration of replacing with other scene descriptions through the scene description replacement instruction is significantly reduced, and the user satisfaction is improved.

[0066] Step 102, generate a target scene based on the target scene description.

[0067] In some embodiments, the target scene can be structured data generated according to the target scene description, which can be directly stored in the historical scene library and only has text descriptions; the target scene can also include both structured data generated according to the target scene description and a target scene schematic diagram, having both text descriptions and pictures; the target scene can also include other content. This application does not limit this.

[0068] After generating the target scene description, the similar scene description of the target scene description can also be found according to the method based on vector retrieval, and then the target scene is generated based on the similar scene description. The following embodiments give a similar introduction to the solution of finding the similar scene description of the target scene description according to the method based on vector retrieval. In some embodiments, such as Figure 3As shown, step 102 includes:

[0069] Step 301, generating a first vector of the target scenario description and second vectors corresponding to each historical scenario description in the historical scenario library.

[0070] Among them, the historical scenario description includes the key elements of the scenario.

[0071] In an exemplary embodiment, the historical scenario library is a database storing past successful cases and typical scenarios, providing reference for the generation of new scenarios. Here, the scenario refers to a standardized scenario description. The historical scenario library may include typical historical scenario descriptions manually entered by users, or historical scenario descriptions recognized by users generated by a scenario description generation large model.

[0072] In an exemplary embodiment, the historical scenario library adopts a MySQL database (a relational database).

[0073] In an exemplary embodiment, the target scenario description and each historical scenario description in the historical scenario library can be batch vectorized through a vectorization large model to obtain a first vector of the target scenario description and second vectors corresponding to each historical scenario description in the historical scenario library, and then the first vector of the target scenario description and the second vectors corresponding to each historical scenario description in the historical scenario library are stored in a vector database.

[0074] Step 302, determining the semantic similarity between the first vector and each second vector.

[0075] In an exemplary embodiment, the semantic similarity may include cosine similarity, or other types of similarity, and this application does not limit this.

[0076] Step 303, determining a similar scenario description of the target scenario description from each historical scenario description based on each semantic similarity.

[0077] In an exemplary embodiment, step 303 may include: determining the historical scenario description corresponding to the semantic similarity greater than a preset similarity threshold as the similar scenario description of the target scenario description. Step 303 may also include: sorting the semantic similarities from large to small to obtain a sorting result; determining the historical scenario descriptions corresponding to the top N semantic similarities in the sorting result as the similar scenario descriptions of the target scenario description; where N is a positive integer.

[0078] In an exemplary embodiment, in the process of determining the similar scenario description of the target scenario description from each historical scenario description, the Reranker technology can be adopted. A re - ranker is a model or algorithm that re - ranks a preliminary retrieved or generated candidate result set.

[0079] Step 304: Generate a target scenario based on the similar scenario descriptions.

[0080] Based on each semantic similarity, determine the similar scenario descriptions of the target scenario description from each historical scenario description. Generate a target scenario based on the similar scenario descriptions. The target scenario is generated through the similar scenario descriptions determined from the historical scenario library to be similar to the target scenario description. Since the target scenario is directly generated using the similar scenario descriptions in the historical scenario library, and the historical scenario library stores typical historical scenario descriptions manually entered by users or historical scenario descriptions recognized by users generated by a scenario description generation large model, all of which are historical scenario descriptions confirmed by users. The similar scenario descriptions determined from the historical scenario library are historical scenario descriptions confirmed by users and similar to the target scenario description. Compared with the directly generated target scenario description, the accuracy of the similar scenario descriptions is higher. Generating a target scenario based on the similar scenario descriptions can improve the accuracy of scenario generation.

[0081] In an exemplary embodiment, as Figure 4 shown, it is a schematic diagram of an exemplary AI (Artificial Intelligence) scenario assistant. The page of the AI scenario assistant can be displayed in the sidebar of the front-end page of the scenario library module of the web (World Wide Web) system of the cloud platform product workstation, or can be displayed at other positions. This application does not limit this. Figure 4 The AI scenario assistant in includes 2 entrances, namely "Generate Scenario" and "Search Scenario". "Generate Scenario" means generating a target scenario description based on the original scenario description input by the user, that is, generating a standardized scenario description. "Search Scenario" means performing a similarity search of scenarios, which can mean searching for scenario descriptions similar to the target scenario description, or searching for scenario descriptions similar to the original scenario description input by the user. Figure 4 In , you can select the button in the shape of a microphone to input the original scenario description by voice, or input the original scenario description by typing. After the input is completed, click the circular send button to send the original scenario description, and then perform scenario generation or similarity search of scenarios. Embedding the AI scenario assistant into the web (World Wide Web) system of the cloud platform product workstation can seamlessly connect to the online work process of the original product manager.

[0082] In an exemplary embodiment, as Figure 5 shown, it is a schematic diagram of the page for the specific operation of generating a scenario. Select the "Generate Scenario" entrance on the page of the AI scenario assistant in , input and send the original scenario description: When the vehicle-mounted computer has an abnormality, the operator can remotely send a log collection instruction to the vehicle, and then collect data for fault analysis. The AI scenario assistant will generate the target scenario description corresponding to the original scenario description: Figure 4 In ,

[0083] Scenario Name: Remote Log Collection

[0084] Scenario Type: Operation Scenario

[0085] Belonging Stage: Operation

[0086] Role: Operator

[0087] Environment: When the in-vehicle computer has an abnormality

[0088] Requirement: The operator needs to remotely collect the in-vehicle computer logs for fault analysis

[0089] Event Behavior: 1. The operator selects the faulty vehicle on the operation platform; 2. Issues a log collection instruction; 3. The in-vehicle computer receives the instruction and starts to collect logs; 4. The log data is uploaded to the operation platform.

[0090] There are 3 buttons below the target scenario description, namely "Copy and New", "Change", and "Similar Scenarios". If "Copy and New" is clicked, the target scenario description can be stored in the historical scenario library. If "Change" is clicked, a new standardized scenario description can be output. If "Similar Scenarios" is clicked, the similar scenario descriptions of the target scenario description in the historical scenario library can be queried. In an exemplary embodiment, as Figure 6 shown, it is a schematic diagram of the page for selecting similar scenarios after generating the scenario. In an exemplary embodiment, as Figure 7 shown, it is a schematic diagram of the page for selecting "Change" after generating the scenario.

[0091] In order to more intuitively display the scenario, the following embodiments provide a solution for generating a target scenario schematic diagram based on the target scenario description. In some embodiments, as Figure 8 shown, step 102 includes:

[0092] Step 801, generate a target scenario schematic diagram based on the target scenario description.

[0093] In an exemplary embodiment, the target scenario schematic diagram refers to a schematic diagram that can reflect the target scenario description. For example, a schematic diagram of the car owner opening the in-vehicle refrigerator by operating the mobile phone APP (Application). The target scenario schematic diagram facilitates users to intuitively understand the target scenario description.

[0094] In an exemplary embodiment, for one target scenario description, one target scenario schematic diagram can be generated, or at least two target scenario schematic diagrams can be generated for the user to select. This application does not limit this.

[0095] To improve the efficiency and accuracy of generating a schematic diagram of a target scenario, the following embodiments propose a solution of first transforming the target scenario description into a text description more suitable for generating a scenario schematic diagram, and then obtaining the scenario schematic diagram through a text-to-image model. In some embodiments, as Figure 9 shown, step 801 includes:

[0096] Step 901, generating a second prompt based on the target scenario description and an example of scenario description transformation.

[0097] Among them, the example of scenario description transformation includes the pre-transformation scenario description and the post-transformation scenario description, and the post-transformation scenario description is used to generate a scenario schematic diagram.

[0098] In an exemplary embodiment, the example of scenario description transformation may include an example of scenario description transformation in the field of vehicle networking, which improves the accuracy of generating a schematic diagram of a target scenario in the vehicle networking field.

[0099] In an exemplary embodiment, the pre-transformation scenario description is not suitable for directly using text-to-image, and the post-transformation scenario description is more suitable for directly using text-to-image to generate a scenario schematic diagram. The post-transformation scenario description is a scenario description obtained by transforming the pre-transformation scenario description.

[0100] To obtain higher-quality examples of scenario description transformation, the following embodiments are proposed. In some embodiments, as Figure 10 shown, the process of determining the example of scenario description transformation includes:

[0101] Step 1001, generating a third prompt based on the sample scenario description and the example of sample scenario description transformation.

[0102] In an exemplary embodiment, the sample scenario description and the example of sample scenario description transformation may include the sample scenario description and the example of sample scenario description transformation in the field of vehicle networking, which can obtain higher-quality examples of scenario description transformation in the vehicle networking field and improve the accuracy of generating a schematic diagram of a target scenario in the vehicle networking field.

[0103] Step 1002, inputting the third prompt into the scenario description transformation large model to obtain the optimized sample scenario description.

[0104] Among them, the third prompt is used to instruct the scenario description transformation large model to refer to the example of sample scenario description transformation and transform the sample scenario description into the optimized sample scenario description.

[0105] Step 1003, inputting the optimized sample scenario description into the text-to-image model to obtain the sample scenario schematic diagram.

[0106] In an exemplary embodiment, an optimized sample scenario description can generate a sample scenario schematic diagram or at least two sample scenario schematic diagrams. Specifically, an optimized sample scenario description can generate a plurality of sample scenario schematic diagrams, and the present application does not limit this.

[0107] Step 1004: Based on the user feedback data of the sample scenario schematic diagram, screen out the optimized sample scenario description transformation examples from the optimized sample scenario description and the sample scenario description.

[0108] In an exemplary embodiment, a mapping relationship can be established among the sample scenario description, the optimized sample scenario description, and the sample scenario schematic diagram. The user feedback data can include whether the user is satisfied with the sample scenario schematic diagram and the degree of satisfaction, and can also include which parts of the sample scenario schematic diagram the user is satisfied with and which parts the user is not satisfied with. The present application does not limit this. Step 1004 can include: Based on the user feedback data of the sample scenario schematic diagram, determine the sample scenario schematic diagram that the user is satisfied with, and determine the sample scenario description and the optimized sample scenario description corresponding to the sample scenario schematic diagram that the user is satisfied with as the optimized sample scenario description transformation examples, where the sample scenario description corresponding to the sample scenario schematic diagram that the user is satisfied with is the scenario description before transformation, and the optimized sample scenario description corresponding to the sample scenario schematic diagram that the user is satisfied with is the scenario description after transformation.

[0109] Step 1005: Determine whether the loop stop condition is satisfied. If so, execute Step 1006; otherwise, execute Step 1007.

[0110] In an exemplary embodiment, the loop stop condition can include that the number of loops reaches a preset number, or can include other loop stop conditions. The present application does not limit this.

[0111] Step 1006: Determine the optimized sample scenario description transformation example as the scenario description transformation example.

[0112] Step 1007: Determine the optimized sample scenario description transformation example as the sample scenario description transformation example, and return to execute Step 1001.

[0113] Determine an optimized sample scenario description transformation example through user feedback data, and then generate a sample scenario schematic diagram through the optimized sample scenario description transformation example. Further obtain user feedback data, continuously optimize the sample scenario description transformation example through the user feedback data, and then obtain a higher-quality scenario description transformation example. Apply the higher-quality scenario description transformation example to generate an optimized scenario description corresponding to the target scenario description, improve the quality of the optimized scenario description corresponding to the target scenario description, and then improve the quality of the target scenario schematic diagram obtained according to the optimized scenario description. Moreover, user feedback data is adopted in the optimization process, which can make the target scenario schematic diagram obtained according to the optimized scenario description more in line with user preferences and improve user satisfaction.

[0114] Step 902: Input the second prompt into the scenario description transformation large model to obtain an optimized scenario description.

[0115] Among them, the second prompt is used to instruct the scenario description transformation large model to refer to the scenario description transformation example and transform the target scenario description into an optimized scenario description.

[0116] Step 903: Input the optimized scenario description into the text-to-image model to obtain a target scenario schematic diagram.

[0117] First, transform the target scenario description through the scenario description transformation large model to obtain an optimized scenario description more suitable for text-to-image. Then, generate a target scenario schematic diagram corresponding to the optimized scenario description through the text-to-image model, which can realize the automatic generation of the target scenario schematic diagram and improve the efficiency of generating the target scenario schematic diagram. Moreover, first transform the target scenario description to obtain an optimized scenario description more suitable for text-to-image, and then perform text-to-image, which can improve the accuracy of generating the target scenario schematic diagram and further improve the efficiency and accuracy of scenario generation.

[0118] Step 802: Generate a target scenario based on the target scenario description and the target scenario schematic diagram.

[0119] Generate a target scenario schematic diagram based on the target scenario description, and generate a target scenario based on the target scenario description and the target scenario schematic diagram. The target scenario includes both the target scenario description and the target scenario schematic diagram, which is convenient for users to intuitively understand the target scenario and improve user satisfaction with the target scenario.

[0120] In an exemplary embodiment, each large model involved in the present application can be deployed as a private domain model, that is, the large model is deployed locally to avoid information leakage and improve security.

[0121] The target scenario can be used to generate a target product requirements document or for other application scenarios. The following embodiments introduce the application scenario where the target scenario can be used to generate a target product requirements document. In some embodiments, the scenario generation method further includes: generating a target product requirements document based on the target scenario.

[0122] In an exemplary embodiment, the product requirements document is a detailed document that defines the goals, scope, functions, user interface, performance requirements, etc. of the product and serves as the basis for guiding product design, development, and testing.

[0123] In an exemplary embodiment, the product requirements document in a professional field needs to be precise, comprehensive, and compliant with industry standards. The professional field can include the Internet of Vehicles or other professional fields. The Internet of Vehicles refers to a system that uses advanced information and communication technologies to achieve interconnection and communication between vehicles and between vehicles and infrastructure, providing various intelligent transportation services.

[0124] In an exemplary embodiment, the target scenario can be directly used to generate a target product requirements document, improving the efficiency and accuracy of writing the product requirements document. It can also serve as a basic version for the product manager and development team to refine and adjust on this basis to ensure the consistency and standardization of the document. This application does not limit this.

[0125] To enrich the presentation form of the target scenario, the following embodiments propose a solution for generating a target product chart based on the target scenario description. In some embodiments, as Figure 11 shown, step 102 includes:

[0126] Step 1101, generating a target product chart based on the target scenario description.

[0127] Among them, the target product chart is used to display the behavior execution logic of the product.

[0128] In an exemplary embodiment, the target product chart can include product charts such as sequence diagrams, data flow diagrams, flowcharts, architecture diagrams, etc., or other types of product charts. This application does not limit this.

[0129] In an exemplary embodiment, step 1101 can be implemented through Mermaid technology. Mermaid technology is a text-based chart drawing language or tool that allows users to describe charts through simple text syntax and then automatically convert them into visual graphics. The roles, environments, and event behaviors in the target scenario description can be combined with the vehicle networking case, and Mermaid source code can be generated through a text-to-text model, and then visualized and drawn on the front end based on the Mermaid source code. It is also possible to generate Mermaid source code through a text-to-text model by combining the event behaviors in the target scenario description with the vehicle networking case, and then visualize and draw on the front end based on the Mermaid source code. This application does not limit this.

[0130] Step 1102, generate a target scenario based on the target scenario description and the target product chart.

[0131] Generating a target product chart based on the target scenario description and then generating a target scenario based on the target scenario description and the target product chart not only enriches the expression form of the target scenario but also effectively improves work efficiency and ensures the accuracy and intuitiveness of information transmission.

[0132] In summary, in this application, based on the original scenario description input by the user, a target scenario description is generated. Here, the original scenario description refers to a vague scenario description, and the target scenario description includes the key elements of the scenario. Based on the target scenario description, a target scenario is generated. Compared with directly manually writing a scenario according to the original scenario description, in this application, the vague original scenario description input by the user is first supplemented and improved to obtain a relatively standardized target scenario description including the key elements of the scenario, and then based on the target scenario description, a target scenario is generated, dividing a complex task into two relatively simple tasks, which can improve the efficiency and accuracy of scenario generation.

[0133] Exemplary device

[0134] Correspondingly, an embodiment of this application also provides a scenario generation device, as Figure 12 shown. The scenario generation device includes:

[0135] A first generation unit 1201, configured to generate a target scenario description based on the original scenario description input by the user; where the original scenario description refers to a vague scenario description, and the target scenario description includes the key elements of the scenario;

[0136] A second generation unit 1202, configured to generate a target scenario based on the target scenario description.

[0137] Optionally, the first generation unit 1201 is specifically configured to:

[0138] The original scene description input by the user is concatenated with a preset prompt template to obtain a first prompt; wherein, the preset prompt template includes requirements for the scene in the vehicle networking field, output specifications, and examples in the vehicle networking field; the output specifications include key elements of the scene; the examples in the vehicle networking field include original scene description examples and target scene description examples in the vehicle networking field;

[0139] The first prompt is input into a scene description generation large model to obtain a target scene description;

[0140] Among them, the first prompt is used to instruct the scene description generation large model to generate a target scene description based on the original scene description input by the user, referring to the requirements for the scene in the vehicle networking field, the output specifications, and the examples in the vehicle networking field.

[0141] Optionally, the first generation unit 1201 is specifically configured to:

[0142] Based on the original scene description input by the user, at least two candidate scene descriptions are generated at one time; wherein, the candidate scene descriptions include key elements of the scene;

[0143] The target scene description is determined from each of the candidate scene descriptions;

[0144] The scene generation device further includes:

[0145] A replacement unit, configured to, after receiving a scene description replacement instruction, determine a candidate scene description other than the target scene description from each of the candidate scene descriptions for display.

[0146] Optionally, the second generation unit 1202 is specifically configured to:

[0147] Generate a first vector of the target scene description and second vectors corresponding to each historical scene description in the historical scene library; wherein, the historical scene descriptions include key elements of the scene;

[0148] Determine the semantic similarity between the first vector and each of the second vectors;

[0149] Based on each of the semantic similarities, determine a similar scene description of the target scene description from each of the historical scene descriptions;

[0150] Generate a target scene based on the similar scene description.

[0151] Optionally, the second generation unit 1202 includes:

[0152] A first generation subunit, configured to generate a target scene schematic diagram based on the target scene description;

[0153] A second generation subunit, configured to generate a target scenario based on the target scenario description and the target scenario schematic diagram.

[0154] Optionally, the first generation subunit is specifically configured to:

[0155] Generate a second prompt based on the target scenario description and a scenario description transformation example, where the scenario description transformation example includes a pre-transformation scenario description and a post-transformation scenario description, and the post-transformation scenario description is used to generate a scenario schematic diagram;

[0156] Input the second prompt into a scenario description transformation large model to obtain an optimized scenario description, where the second prompt is used to instruct the scenario description transformation large model to refer to the scenario description transformation example and transform the target scenario description into the optimized scenario description;

[0157] Input the optimized scenario description into a text-to-image model to obtain a target scenario schematic diagram.

[0158] Optionally, the first generation subunit is further configured to:

[0159] Generate a third prompt based on a sample scenario description and a sample scenario description transformation example;

[0160] Input the third prompt into the scenario description transformation large model to obtain a sample optimized scenario description, where the third prompt is used to instruct the scenario description transformation large model to refer to the sample scenario description transformation example and transform the sample scenario description into the sample optimized scenario description;

[0161] Input the sample optimized scenario description into the text-to-image model to obtain a sample scenario schematic diagram;

[0162] Based on the user feedback data of the sample scenario schematic diagram, screen out an optimized sample scenario description transformation example from the sample scenario description transformation examples;

[0163] Determine the optimized sample scenario description transformation example as the sample scenario description transformation example, and return to execute the step of generating a third prompt based on the sample scenario description and the sample scenario description transformation example until a loop stop condition is met, and determine the optimized sample scenario description transformation example as the scenario description transformation example.

[0164] Optionally, the scenario generation device further includes:

[0165] A product requirement document generation unit, configured to generate a target product requirement document based on the target scenario.

[0166] The scenario generation device provided in this embodiment belongs to the same inventive concept as the scenario generation method provided in the foregoing embodiments of the present application, and can execute the scenario generation method provided in any of the foregoing embodiments of the present application, and has corresponding functional modules and beneficial effects for executing the scenario generation method. For technical details not described in detail in this embodiment, reference may be made to the specific processing content of the scenario generation method provided in the foregoing embodiments of the present application, which will not be elaborated here.

[0167] The functions implemented by the above first generation unit 1201 and second generation unit 1202 may be implemented by the same or different processors respectively, which is not limited in the embodiments of the present application.

[0168] It should be understood that the units in the above device may be implemented in the form of a processor calling software. For example, the device includes a processor, the processor is connected to a memory, instructions are stored in the memory, and the processor calls the instructions stored in the memory to implement any of the above methods or the functions of each unit of the device. The processor may be a general-purpose processor, such as a CPU or a microprocessor, etc., and the memory may be a memory inside or outside the device. Alternatively, the units in the device may be implemented in the form of a hardware circuit, and the functions of some or all of the units may be implemented through the design of the hardware circuit. The hardware circuit may be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are implemented through the design of the logical relationship of the components in the circuit. Again, for example, in another implementation, the hardware circuit may be implemented by a PLD. Taking an FPGA as an example, it may include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured through a configuration file, so as to implement the functions of some or all of the above units. All units of the above device may be all implemented in the form of a processor calling software, or all implemented in the form of a hardware circuit, or some implemented in the form of a processor calling software, and the remaining part implemented in the form of a hardware circuit.

[0169] In the embodiments of the present application, a processor is a circuit with signal processing capabilities. In one implementation, the processor may be a circuit with instruction reading and running capabilities, such as a CPU, a microprocessor, a GPU, or a DSP, etc. In another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or can be reconstructed. For example, the processor is a hardware circuit implemented by an ASIC or a PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the configuration of the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it may also be a hardware circuit designed for artificial intelligence, which may be understood as a type of ASIC, such as an NPU, a TPU, a DPU, etc.

[0170] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method. For example: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0171] In addition, each unit in the above device can be integrated in whole or in part, or can be independently implemented. In one implementation, these units are integrated together and implemented in the form of an SOC. The SOC can include at least one processor for implementing any of the above methods or the functions of each unit of the device. The types of the at least one processor can be different. For example, it can include a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0172] Exemplary system

[0173] Exemplarily, an embodiment of the present application further provides a scenario generation system, which mainly includes three parts: a web front end, an intermediate service, and a privatized container. The web front end includes a voice input module, a first scenario generation module, and a first similarity retrieval module; the intermediate service includes a voice recognition module, a second scenario generation module, and a second similarity retrieval module; the privatized container includes a large voice recognition model, a large text generation model, a large image generation model, and a vectorization processing large model;

[0174] The voice input module is used to obtain the voice signal input by the user and send the voice signal to the voice recognition module;

[0175] The voice recognition module is used to convert the voice signal into an original scenario description by calling the large voice recognition model and send the original scenario description to the first scenario generation module;

[0176] The first scenario generation module is used to send the original scenario description to the second scenario generation module after receiving a scenario generation instruction;

[0177] The second scenario generation module is used to call the large text generation model and the large image generation model, and generate a target scenario description and a target scenario schematic diagram based on the original scenario description;

[0178] The first similarity retrieval module is used to send the similarity retrieval instruction to the second similarity retrieval module after receiving a similarity retrieval instruction;

[0179] A second similarity retrieval module, configured to call a vectorized large model to generate a first vector of the target scenario description and second vectors corresponding to respective historical scenario descriptions in the historical scenario library; determine the semantic similarity between the first vector and each of the second vectors; and determine a similar scenario description of the target scenario description from each of the historical scenario descriptions based on each semantic similarity.

[0180] In an exemplary embodiment, the target scenario description and the target scenario schematic diagram can be used to generate a target scenario, and the similar scenario description of the target scenario description can also be used to generate a target scenario.

[0181] Exemplary electronic device

[0182] An embodiment of the present application provides an electronic device. Refer to Figure 13 as shown, the device includes:

[0183] a memory 200 and a processor 210;

[0184] wherein, the memory 200 is connected to the processor 210 and is configured to store programs;

[0185] The processor 210 is configured to implement the scenario generation method disclosed in any of the above embodiments by running the programs stored in the memory 200.

[0186] Specifically, the above electronic device may further include: a bus, a communication interface 220, an input device 230, and an output device 240.

[0187] The processor 210, the memory 200, the communication interface 220, the input device 230, and the output device 240 are interconnected through the bus. Among them:

[0188] The bus may include a path for transmitting information between various components of the computer system.

[0189] The processor 210 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0190] The processor 210 may include a main processor, and may also include a baseband chip, a modem, etc.

[0191] The program for implementing the technical solution of the present invention is stored in the memory 200, and the operating system and other key services may also be stored. Specifically, the program may include program code, and the program code includes computer operation instructions. More specifically, the memory 200 may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk memory, a flash memory, and so on.

[0192] The input device 230 may include devices for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor, etc.

[0193] The output device 240 may include devices for allowing information to be output to a user, such as a display screen, a printer, a speaker, etc.

[0194] The communication interface 220 may include devices of any transceiver type for communicating with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0195] The processor 210 executes the program stored in the memory 200 and calls other devices, and can be used to implement each step of any one of the scenario generation methods provided in the above embodiments of the present application.

[0196] Exemplary computer program product and storage medium

[0197] In addition to the above methods and devices, the embodiments of the present application may also be a computer program product, which includes computer program instructions. When the computer program instructions are run by a processor, the processor is caused to execute the steps in the scenario generation method according to various embodiments of the present application described in any of the above embodiments of this specification.

[0198] The computer program product may be written in any combination of one or more programming languages for programming code for performing the operations of the embodiments of the present application. The programming languages include object-oriented programming languages, such as Java, C++, etc., and also include conventional procedural programming languages, such as the "C" language or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, executed as an independent software package, partially on a user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0199] In addition, an embodiment of the present application may also be a storage medium, on which a computer program is stored. The computer program is executed by a processor to perform the steps in the scenario generation method according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically may implement the following steps:

[0200] Step 101: Generate a target scenario description based on the original scenario description input by the user.

[0201] The original scenario description refers to a vague scenario description, and the target scenario description includes key elements of the scenario.

[0202] Step 102: Generate a target scenario based on the target scenario description.

[0203] For the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0204] It should be noted that the embodiments in this specification are all described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0205] The steps in the methods of the embodiments of the present application can be adjusted, combined, and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.

[0206] The modules and sub-modules in the devices and terminals in the embodiments of the present application can be combined, divided, and deleted according to actual needs.

[0207] In several embodiments provided by the present application, it should be understood that the disclosed terminals, devices, and methods can be implemented in other ways. For example, the terminal embodiments described above are merely illustrative. For example, the division of modules or sub-modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple sub-modules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or modules can be in an electrical, mechanical, or other form.

[0208] A module or sub-module described as a separate component may or may not be physically separated. A component as a module or sub-module may or may not be a physical module or sub-module, that is, it may be located in one place, or may be distributed across multiple network modules or sub-modules. Some or all of the modules or sub-modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0209] In addition, each functional module or sub-module in various embodiments of the present application can be integrated in a processing module, or each module or sub-module can exist physically alone, or two or more modules or sub-modules can be integrated in one module. The above-mentioned integrated module or sub-module can be implemented in the form of hardware, or can be implemented in the form of a software functional module or sub-module.

[0210] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0211] The steps of the method or algorithm described in combination with the embodiments disclosed in this article can be directly implemented by hardware, a software unit executed by a processor, or a combination of the two. The software unit can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0212] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0213] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A scene generation method, characterized in that: include: Generate a target scene description based on an original scene description input by a user; wherein the original scene description refers to a fuzzy scene description, and the target scene description includes key elements of the scene; Based on the target scene description, a target scene is generated.

2. The scene generation method according to claim 1, characterized in that: The step of generating a target scene description based on an original scene description input by a user includes: The original scene description input by the user and the preset prompt word template are spliced ​​to obtain a first prompt word; wherein the preset prompt word template includes the requirements for the scene in the field of Internet of Vehicles, output specifications and examples in the field of Internet of Vehicles; the output specifications include key elements of the scene; the examples in the field of Internet of Vehicles include original scene description samples and target scene description samples in the field of Internet of Vehicles; Inputting the first prompt word into a scene description generation model to obtain a target scene description; Among them, the first prompt word is used to instruct the scene description generation model to generate a target scene description based on the original scene description input by the user, with reference to the requirements of the Internet of Vehicles field for the scene, the output specification and the examples of the Internet of Vehicles field.

3. The scene generation method according to claim 1, characterized in that: The step of generating a target scene description based on an original scene description input by a user includes: Based on the original scene description input by the user, at least two candidate scene descriptions are generated at one time; wherein the candidate scene descriptions include key elements of the scene; Determining the target scene description from each of the candidate scene descriptions; The method further comprises: After receiving the scene description change instruction, a candidate scene description other than the target scene description is determined from the candidate scene descriptions for display.

4. The scene generation method according to claim 1, characterized in that: The generating a target scene based on the target scene description includes: Generate a first vector of the target scene description and a second vector corresponding to each of the historical scene descriptions in the historical scene library; wherein the historical scene descriptions include key elements of the scene; determining a semantic similarity between the first vector and each of the second vectors; Based on each of the semantic similarities, determining a similar scene description to the target scene description from each of the historical scene descriptions; Based on the similar scene description, a target scene is generated.

5. The scene generation method according to claim 1, characterized in that: The generating a target scene based on the target scene description includes: Based on the target scene description, generate a target scene schematic diagram; A target scene is generated based on the target scene description and the target scene schematic diagram.

6. The scene generation method according to claim 5, characterized in that: The generating a target scene schematic diagram based on the target scene description includes: Based on the target scene description and the scene description transformation example, a second prompt word is generated; wherein the scene description transformation example includes a scene description before transformation and a scene description after transformation, and the scene description after transformation is used to generate a scene schematic diagram; Inputting the second prompt word into the scene description transformation model to obtain an optimized scene description; wherein the second prompt word is used to instruct the scene description transformation model to refer to the scene description transformation example and transform the target scene description into the optimized scene description; The optimized scene description is input into the text graph model to obtain a target scene schematic diagram.

7. The scene generation method according to claim 6, characterized in that: The method further comprises: Based on the sample scene description and the sample scene description modification example, generating a third prompt word; Input the third prompt word into the scene description transformation model to obtain the sample optimized scene description; wherein the third prompt word is used to instruct the scene description transformation model to transform the sample scene description into the sample optimized scene description with reference to the sample scene description transformation example; Inputting the sample optimized scene description into the text graph model to obtain a sample scene schematic diagram; Based on user feedback data of the sample scene schematic diagram, selecting an optimized sample scene description modification example from the sample optimized scene description and the sample scene description; The optimized sample scene description modification example is determined as the sample scene description modification example, and the step of generating a third prompt word based on the sample scene description and the sample scene description modification example is returned to execute until the loop stop condition is met, and the optimized sample scene description modification example is determined as the scene description modification example.

8. The scene generation method according to any one of claims 1 to 7, characterized in that: The method further comprises: Based on the target scenario, a target product requirement document is generated.

9. An electronic device, characterized in that: including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the scene generation method according to any one of claims 1 to 8 by running the program in the memory.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the scene generation method according to any one of claims 1 to 8 is implemented.

Images