Voice interaction processing method and device based on diagnostic equipment, terminal equipment and storage medium

Abstract

The application is suitable for the field of automobile maintenance diagnosis and intelligent voice interaction technology, and provides a voice interaction processing method and device based on a diagnosis equipment, a terminal equipment and a storage medium, which comprises the following steps: receiving a voice control instruction of a user operating a diagnosis equipment; matching a corresponding target preset model according to the voice control instruction; sending the voice control instruction to the target preset model, so that the target preset model performs semantic conversion on the voice control instruction to obtain structured instruction data output by the target preset model; and executing the structured instruction data to respond to the operation demand of the user. The above method can improve the response speed and accuracy of intelligent voice AI.

Description

Technical Field

[0001] This application belongs to the field of automotive repair diagnosis and intelligent voice interaction technology, and in particular relates to a voice interaction processing method, device, terminal equipment and storage medium based on diagnostic equipment. Background Technology

[0002] In automotive repair scenarios, technicians need to use diagnostic equipment (primarily an Android tablet, which communicates with the vehicle via a VCI diagnostic box to obtain information such as vehicle year, engine system, and fault data) to conduct repair and diagnostic work. To avoid the inconvenience of manually operating the equipment when hands are covered in oil, intelligent voice artificial intelligence (AI) has become an important interaction method for technicians to operate diagnostic equipment. Its core requirement is to quickly and accurately trigger various functions of the diagnostic equipment through voice commands (such as opening diagnostic reports, starting specific diagnostic systems, and performing intelligent diagnostics), assisting technicians in efficiently completing repair tasks.

[0003] In pursuit of accuracy in diagnostic data, related technologies generally employ models with large memory requirements, resulting in longer processing times, slower voice response, and reduced repair efficiency. On the other hand, replacing the model with one with a smaller memory model to improve speed leads to lower recognition accuracy, making it easy to misinterpret the technician's voice commands, which in turn causes the diagnostic equipment to perform incorrect operations, affecting the normal progress of repair work. Summary of the Invention

[0004] This application provides a voice interaction processing method, apparatus, terminal device, and storage medium based on diagnostic equipment, which can improve the response speed and accuracy of intelligent voice AI.

[0005] In a first aspect, embodiments of this application provide a voice interaction processing method based on a diagnostic device, applied to the diagnostic device; the diagnostic device includes different preset models; each preset model has different processing speeds and processing accuracy; including: Receive voice control commands from users to operate diagnostic equipment; Match the corresponding target preset model according to the voice control command; The voice control command is sent to the target preset model so that the target preset model can perform semantic conversion on the voice control command and obtain the structured command data output by the target preset model; Execute structured instruction data in response to user control requests.

[0006] In this embodiment, the diagnostic device first receives natural language control commands (voice control commands) issued by the user through an audio acquisition module. Then, based on the function type of the command, it matches and adapts to a pre-defined target model. The command is then sent to the model, which uses a corresponding algorithm to perform semantic conversion from natural language to structured command data. Finally, the device parses and executes the structured data to fulfill the user's control needs. This method uses precise "command-model" matching to allow simple commands to be processed quickly using lightweight models (low memory, low computational load), avoiding resource redundancy. Simultaneously, structured command data can be directly executed by the system, shortening the execution chain. Furthermore, high-complexity commands are matched with high-performance models (more sophisticated semantic recognition algorithms), and different models are specifically adapted to corresponding command scenarios, reducing generalization errors and making the semantic understanding of different types of commands more accurate.

[0007] In one possible implementation of the first aspect, matching a corresponding target preset model based on the voice control command includes: Determine the type of voice control command; Match the corresponding target preset model based on the instruction type.

[0008] In this embodiment, by first determining the instruction type and then matching the corresponding model, simple instructions are matched with lightweight models to ensure response speed, while complex instructions are matched with high-performance models to improve semantic understanding accuracy. This achieves "scenario-specific adaptation," which avoids resource waste and can specifically improve the processing effect of different types of instructions.

[0009] In one possible implementation of the first aspect, matching the corresponding target preset model according to the instruction type includes: In the case of instruction type 1, detect the first available memory of the diagnostic device; If the available memory is greater than the first preset model, the first preset model is determined as the target preset model; wherein, the first preset model is the preset model with the lowest memory requirement among multiple preset models.

[0010] In this embodiment, the first preset model with the lowest memory requirement is matched for the first type of basic instruction, and it is only enabled when the available memory is sufficient. This ensures the fast response of basic instructions and minimizes memory resource consumption, thus avoiding interference with other core functions of the diagnostic device.

[0011] In one possible implementation of the first aspect, the method includes: In the case of instruction type 2, detect the second available memory of the diagnostic device; If the second available memory is greater than the second preset model, the second preset model is determined as the target preset model; wherein, the memory requirement of the second preset model is higher than that of the first preset model, and the performance of the second preset model is higher than that of the first preset model; the functions of the second type are higher than those of the first type.

[0012] In this embodiment, for the second type of instruction with more complex functions, a second preset model with better performance and higher memory requirements is matched, and the model's operating conditions are ensured through memory detection. This achieves precise adaptation between the processing performance of complex instructions and device resources, which not only ensures the processing quality of the second type of instructions, but also forms a layered design with the lightweight processing of the first type of instructions, thereby improving the adaptability and reliability of the overall voice control system.

[0013] In one possible implementation of the first aspect, the method includes: If the second available memory is less than the second preset model but greater than the third preset model, the third preset model is determined as the target preset model; wherein, the memory requirement of the third preset model is greater than that of the first preset model but less than that of the second preset model; and the performance of the third preset model is higher than that of the first preset model but less than that of the second preset model.

[0014] In this embodiment, a third preset model with intermediate performance is added as an alternative for the second type of instruction. When the device memory is insufficient to support the second preset model, the system automatically switches to this model. This avoids the problem of complex instructions being unable to be processed due to insufficient resources, and also achieves flexible adaptation between performance and memory resources, ensuring the stability and continuity of the processing of the second type of instruction.

[0015] In one possible implementation of the first aspect, executing structured instruction data in response to user control requests includes: If the instruction type is type 1, then structured instruction data is executed in response to the user's control requirements; If the instruction type is type 2, then the vehicle data required for the structured instruction data is obtained; Vehicle data is processed to respond to user control needs.

[0016] In this embodiment, a layered execution logic is adopted for different types of instructions. Basic instructions are executed directly to ensure response efficiency; diagnostic instructions first acquire vehicle data and then process it to ensure the accuracy and professionalism of instruction execution, thus achieving optimal adaptation of instruction execution in different scenarios.

[0017] In one possible implementation of the first aspect, the method further includes: Obtain user feedback data regarding the operation and diagnostic equipment; The preset model is optimized based on the feedback data to obtain the optimized preset model; Based on the optimized preset model, the next voice control command is semantically transformed to obtain the structured command data corresponding to the next voice control command. The diagnostic equipment is controlled based on the next structured instruction data.

[0018] In this embodiment, a complete closed loop of "feedback collection - model optimization - iterative application" is constructed. The preset model is continuously optimized through user feedback and implemented in subsequent command processing, realizing dynamic iterative improvement of the performance of the voice control system and effectively enhancing the response speed and accuracy of intelligent voice AI.

[0019] Secondly, embodiments of this application provide a voice interaction processing device based on a diagnostic device, comprising: The instruction receiving module is used to receive voice control instructions from users to operate the diagnostic equipment; The model matching module is used to match the corresponding target preset model according to the voice control command; The semantic conversion module is used to send voice control commands to the target preset model so that the target preset model can perform semantic conversion on the voice control commands and obtain the structured command data output by the target preset model. The device control module is used to execute structured instruction data in response to the user's control needs.

[0020] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a voice interaction processing device based on a diagnostic device as described in any of the first aspects above.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the voice interaction processing method based on a diagnostic device as described in any of the first aspects above.

[0022] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the voice interaction processing method based on a diagnostic device as described in any of the first aspects above.

[0023] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating the voice interaction processing method based on diagnostic equipment provided in an embodiment of this application; Figure 2 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 1 ; Figure 3 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 2 ; Figure 4 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 3 ; Figure 5 This is a schematic diagram of the process for responding to control requests provided in an embodiment of this application; Figure 6 This is a schematic diagram of the model optimization process provided in the embodiments of this application; Figure 7 This is a summary structural block diagram of the voice interaction processing method based on diagnostic equipment provided in the embodiments of this application; Figure 8 This is a structural block diagram of the voice interaction processing device based on diagnostic equipment provided in the embodiments of this application; Figure 9 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0026] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0027] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0028] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0029] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0030] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0032] In automotive repair scenarios, technicians need to use diagnostic equipment (primarily an Android tablet, which communicates with the vehicle via a VCI diagnostic box to obtain information such as vehicle year, engine system, and fault data) to conduct repair and diagnostic work. To avoid the inconvenience of manually operating the equipment when hands are covered in oil, intelligent voice artificial intelligence (AI) has become an important interactive method for technicians to operate diagnostic equipment. Its core requirement is to quickly and accurately trigger various functions of the diagnostic equipment through voice commands (such as opening diagnostic reports, starting specific diagnostic systems, and performing intelligent diagnostics), assisting technicians in efficiently completing repair tasks.

[0033] In pursuit of accuracy in diagnostic data, related technologies generally employ models with large memory requirements, resulting in longer processing times, slower voice response, and reduced repair efficiency. On the other hand, replacing the model with one with a smaller memory model to improve speed leads to lower recognition accuracy, making it easy to misinterpret the technician's voice commands, which in turn causes the diagnostic equipment to perform incorrect operations, affecting the normal progress of repair work.

[0034] To address the aforementioned technical issues, this application provides a voice interaction processing method, apparatus, terminal device, and storage medium based on diagnostic equipment. By pre-setting models with different memory requirements (1~2GB to 8~12GB) and speed and accuracy gradients (A / B / C / D, etc.), and combining diagnostic function classifications (basic / diagnostic / intelligent diagnostic), the model is dynamically matched based on the available memory of the diagnostic equipment and the function type corresponding to the technician's voice commands. Through iterative model training and continuous optimization, the overall improvement of intelligent voice AI adaptability, response speed, and accuracy is achieved, solving the problem of the difficulty in achieving both in the prior art.

[0035] See Figure 1 This is a flowchart illustrating a voice interaction processing method based on a diagnostic device provided in an embodiment of this application. The method is applied to a diagnostic device; the diagnostic device includes different preset models; each preset model has different processing speeds and accuracy; as an example and not a limitation, the method may include the following steps: S101 receives voice control commands from the user to operate the diagnostic equipment.

[0036] In this embodiment, the diagnostic device can be an Android tablet, used as a visualization device for the automotive repair process. It typically communicates with the VCI (Vehicle Communication Interface) via Bluetooth. The VCI, also known as a diagnostic box, is a device used to communicate with vehicle data. The software system of this diagnostic device pre-integrates multiple intelligent voice processing models of different specifications, and the memory usage requirements of each model vary significantly, adapting to different device operating states and command processing scenarios.

[0037] The diagnostic equipment uses a built-in microphone acquisition module in its hardware terminal to monitor and capture natural language speech signals (voice control commands) issued by users (such as automotive mechanics) in real time. The core purpose of these speech signals is to control the diagnostic equipment to perform specific functions. The acquisition module first converts the speech signal in the form of sound waves into electrical signals, and then converts it into digital audio signals that the diagnostic equipment system can recognize through a digital-to-analog converter circuit. This completes the reception and preliminary preprocessing of the user's voice control commands, providing the raw data foundation for subsequent steps such as model matching, command recognition, and escaping.

[0038] S102 matches the corresponding target preset model according to the voice control command.

[0039] In this embodiment of the application, after receiving a user's voice control command, the diagnostic device will select and match the most suitable model for processing the command from multiple preset models built into the device with different memory requirements, based on the specific attributes of the command (such as function type, complexity, etc.). The selected model is the target preset model.

[0040] The diagnostic equipment's software system pre-integrates multiple intelligent processing models with speech recognition and speech decryption functions. Each model employs different speech signal processing algorithms and semantic analysis logic, resulting in significant differences in memory usage, response speed, and recognition accuracy. Specific parameter configurations are as follows: Model A employs a lightweight frame-matching speech recognition algorithm and keyword-mapping speech escaping method, with memory requirements controlled within the range of 1-2GB. It boasts the fastest response time among all models and is suitable for simple instruction processing where accuracy requirements are not high, achieving a general level of accuracy. Model B utilizes an optimized dynamic time warping speech recognition algorithm and phrase-level semantic escaping method, with memory requirements ranging from 2-4GB. It maintains a relatively fast response time and achieves good accuracy, making it suitable for balanced processing of moderately complex instructions. Model C employs a deep neural network speech recognition algorithm and sentence-level semantic escaping method, with memory requirements ranging from 4-8GB. It has a moderate response time and excellent accuracy, making it suitable for complex instruction processing with high precision requirements. Model D combines a convolutional neural network and recurrent neural network speech recognition algorithm with context-related semantic escaping method, resulting in extremely high memory requirements, reaching 8-12GB. It has the slowest response time among all models but achieves the highest accuracy, making it suitable for core instruction processing with stringent accuracy requirements.

[0041] In one embodiment, see Figure 2 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 1 ,like Figure 2 As shown, step S102 includes: S201, determine the command type of the voice control command.

[0042] In this embodiment, after receiving a user's voice control command and completing lightweight preprocessing, the diagnostic device uses a built-in lightweight algorithm and a preset keyword-function type mapping table to quickly and initially determine the core function of the command, classifying it into one of the following: basic function command, diagnostic function command, or intelligent diagnostic function command. This provides a core basis for subsequent model matching based on the device's memory status.

[0043] For example, the preliminary instruction type determination module of the diagnostic device receives a digital audio signal (voice control instruction) after lightweight preprocessing and the core keyword fragments extracted from it. Then, it retrieves the preset keyword-function type mapping table built into the device. This mapping table stores the core keyword sets corresponding to three categories: basic function instructions, diagnostic function instructions, and intelligent diagnostic function instructions. The module uses a lightweight keyword matching algorithm to quickly compare the extracted keyword fragments with the keyword sets in the mapping table. If keywords such as "open," "return," "print," "home," and "report" are matched, it is determined to be a basic function instruction. If keywords such as "engine," "transmission," "ABS," "fault code," and "data stream" are matched, it is determined to be a diagnostic function instruction. If keywords such as "start," "rapid diagnosis," "analyze fault," and "generate solution" are matched, it is determined to be an intelligent diagnostic function instruction. Finally, the module completes the preliminary determination of the instruction type of the voice control instruction and generates determination result data containing instruction type tags, which is then transmitted to the model matching module or the diagnostic business processing module.

[0044] S202, Match the corresponding target preset model according to the instruction type.

[0045] In this embodiment of the application, after the diagnostic device completes the type determination of the voice control command, it will combine the preset model matching rule library, take the command type as the core basis and the real-time available memory of the device as the constraint, and select and determine the unique model that best matches the current command processing requirements from multiple preset models with different memory requirements. This model is the corresponding target preset model.

[0046] In the above method, by first determining the instruction type and then matching the corresponding model, simple instructions are matched with lightweight models to ensure response speed, while complex instructions are matched with high-performance models to improve semantic understanding accuracy. This achieves "scenario-specific adaptation", which avoids resource waste and can specifically improve the processing effect of different types of instructions.

[0047] In one embodiment, see Figure 3 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 2 ,like Figure 3 As shown, step S202 includes: S301, when the instruction type is type 1, detects the first available memory of the diagnostic device.

[0048] In this embodiment of the application, when the diagnostic device initially determines the received voice control command as a first type (such as a basic function type command), it will specifically detect the current real-time available memory of the diagnostic device. The memory data obtained from this detection is defined as the first available memory, providing a memory status basis for subsequent model matching for basic function type commands.

[0049] For example, when the instruction type preliminary determination module of the diagnostic device outputs the instruction type label as the first type (basic function type), the built-in system resource monitoring module of the device is immediately triggered to execute a dedicated memory detection process. This module reads the free memory space data in the device's memory controller in real time by calling the underlying system interface of the diagnostic device, filters out unusable parts such as system reserved memory and temporary cache memory, accurately obtains the actual memory size that the current diagnostic device can use to load the preset model, and defines the detection result as the first available memory. At the same time, it generates standardized detection data containing the first type instruction label and the specific value of the first available memory, and transmits it synchronously to the model matching module to provide a complete judgment basis for subsequent target preset model matching for basic function type instructions.

[0050] S302, if the available memory is greater than the first preset model, the first preset model is determined as the target preset model; wherein, the first preset model is the preset model with the lowest memory requirement among multiple preset models.

[0051] In this embodiment of the application, when the first available memory (the real-time available memory corresponding to the basic function type instruction) detected by the diagnostic device is greater than the memory requirement of the first preset model, the first preset model is directly determined as the target preset model for processing the instruction; wherein, the first preset model is explicitly defined as the model with the lowest memory usage requirement among the multiple preset models A, B, C, and D built into the device (i.e., model A, with a memory requirement of 1~2GB).

[0052] Specifically, when the model matching module of the diagnostic device receives standardized detection data containing the first type of instruction label and the specific value of the first available memory, it immediately retrieves the parameter information of the built-in first preset model. The first preset model is model A, which has the lowest memory usage among the four gradient preset models A, B, C, and D pre-stored by the device. Its memory requirement threshold is 1~2GB. The module compares the real-time detected specific value of the first available memory with the maximum memory requirement threshold (2GB) of the first preset model through the built-in numerical comparison algorithm. If it is determined that the first available memory is greater than the memory requirement of the first preset model, it means that the current memory resources of the device are sufficient to support the stable operation of the model and meet the core requirement of the first type of instruction for fast processing. At this time, the module will directly execute the model matching operation, determine the first preset model (model A) as the target preset model for processing the basic function type instruction, and generate standardized matching result data containing the first preset model identifier, instruction type label, matching success status, and memory comparison result. This data is synchronously transmitted to the device system to trigger the start of the target preset model, preparing for subsequent speech recognition and speech translation operations.

[0053] In the above method, the first preset model with the lowest memory requirement is matched for the first type of basic instructions, and it is only activated when the available memory is sufficient. This ensures the fast response of basic instructions and minimizes memory resource consumption, thus avoiding interference with other core functions of the diagnostic device.

[0054] In one embodiment, see Figure 4 This is a flowchart illustrating the matching of a preset model provided in the embodiments of this application. Figure 3 ,like Figure 4 As shown, step S202 includes: S401, when the instruction type is type 2, detects the second available memory of the diagnostic device.

[0055] In this embodiment, when the diagnostic device initially determines that the received voice control command is a second type of command, it will trigger a corresponding dedicated memory detection process to accurately detect the device's current real-time available memory and define the detection result as the second available memory. The second type of command is a diagnostic type, such as a diagnostic function type command or an intelligent diagnostic function command. Both types of commands need to be matched with the corresponding preset model (B / C / D model) based on the detection result of the second available memory.

[0056] For example, in the case of the second type of diagnostic function type instruction, the diagnostic function type instruction is used to control the diagnostic equipment and perform voice commands for functions such as reading data of a single system and querying basic faults. Its complexity is higher than that of basic function instructions but lower than that of intelligent diagnostic function instructions. For example, when the technician issues the voice command "read engine fault codes", the instruction type determination module of the diagnostic equipment will initially determine it as a diagnostic function type instruction in the second type based on the keywords "engine" and "fault codes". Then, the system resource monitoring module is triggered to detect the current real-time available memory of the equipment and define the result as the second available memory, providing a basis for subsequent matching of model B (2~4GB memory) or model C (4~8GB memory).

[0057] Another example is the case where the second type is an intelligent diagnostic function type command. Intelligent diagnostic functions are used to control diagnostic equipment and perform advanced functions such as multi-system joint analysis, fault cause reasoning, and repair plan generation. Its complexity is the highest among all command types, and the requirements for model recognition accuracy are also the highest. For example, when a technician issues the voice command "Start rapid diagnosis and analyze the cause of insufficient vehicle power", the command type determination module of the diagnostic equipment will initially determine it as an intelligent diagnostic function type command in the second type based on the keywords "start rapid diagnosis" and "analyze the cause". Then, the system resource monitoring module is triggered to detect the current real-time available memory of the equipment and define the result as the second available memory, providing a basis for subsequent matching of C model (4~8GB memory) or D model (8~12GB memory).

[0058] S402, if the second available memory is greater than the second preset model, the second preset model is determined as the target preset model; wherein, the memory requirement of the second preset model is higher than that of the first preset model, and the performance of the second preset model is higher than that of the first preset model; the functions of the second type are higher than those of the first type.

[0059] In this embodiment, when the second available memory detected by the diagnostic device is greater than the memory requirement of the second preset model, the second preset model is directly determined as the target preset model for processing the instruction; there are three implicit key logical constraints: the memory requirement of the second preset model is higher than that of the first preset model (Model A); the voice processing performance (combined performance of speed and accuracy) of the second preset model is better than that of the first preset model; and the functional complexity and processing requirements of the second type of instruction are higher than those of the first type of instruction (basic function type).

[0060] For example, the technician issues a voice command, "Read the real-time data stream of the transmission." This command is identified as a diagnostic function command in the second type. The diagnostic equipment triggers a dedicated testing process, determining that the second available memory is 5GB. Here, the second preset model is selected as Model C (memory requirement 4-8GB, medium speed, excellent accuracy). Its matching rules are as follows: The second available memory (5GB) > the memory requirement of the second preset model (Model C) (4~8GB, meeting the minimum threshold of 4GB); the memory requirement of the second preset model (Model C) (4~8GB) > the first preset model (Model A, 1~2GB); the performance of the second preset model (Model C) (excellent accuracy) > the first preset model (Model A, average accuracy); the function of the second type of instruction (reading the transmission data stream) > the first type of instruction (such as opening the report). According to the rules, Model C is determined as the target preset model to handle this diagnostic function instruction, ensuring that the accuracy of data stream reading meets the maintenance requirements.

[0061] In another example, when a technician issues the voice control command "Initiate rapid diagnostics to analyze the cause of insufficient vehicle power," this command is identified as a smart diagnostic function command in the second type. The diagnostic device triggers a dedicated testing process, obtaining a second available memory of 10GB. Here, the second preset model is selected as the D model (memory requirement 8-12GB, slow speed, highest accuracy). The matching rule is: The second available memory (10GB) > the memory requirement of the second preset model (Model D) (8~12GB, meeting the minimum threshold of 8GB); the memory requirement of the second preset model (Model D) (8~12GB) > the first preset model (Model A, 1~2GB); the performance of the second preset model (Model D) (highest accuracy) > the first preset model (Model A, moderate accuracy); the function of the second type of instruction (analyzing the cause of insufficient power) > the first type of instruction (such as returning to the homepage). According to the rules, Model D is determined as the target preset model to handle this intelligent diagnostic function instruction, ensuring the accuracy of multi-system joint analysis and providing a reliable basis for maintenance.

[0062] In the above method, for the second type of instructions with more complex functions, a second preset model with better performance and higher memory requirements is matched, and memory detection is used to ensure the model's operating conditions. This achieves precise adaptation between the processing performance of complex instructions and device resources, which not only ensures the processing quality of the second type of instructions, but also forms a layered design with the lightweight processing of the first type of instructions, thereby improving the adaptability and reliability of the overall voice control system.

[0063] In one implementation, the method also includes: If the second available memory is less than the second preset model but greater than the third preset model, the third preset model is determined as the target preset model; wherein, the memory requirement of the third preset model is greater than that of the first preset model but less than that of the second preset model; and the performance of the third preset model is higher than that of the first preset model but less than that of the second preset model.

[0064] In this embodiment of the application, when the second available memory detected by the diagnostic device cannot meet the memory requirements of the second preset model (i.e., the second available memory is less than the second preset model), but can meet the memory requirements of the third preset model (i.e., the second available memory is greater than the third preset model), the third preset model will be determined as the target preset model for processing the instruction.

[0065] For example, when a technician issues the voice command "Read the engine's real-time data stream," and this is determined to be a diagnostic function command in the second type, to match the performance requirements of this command, the second preset model is pre-set as Model C (4~8GB, excellent accuracy, optimal choice), the third preset model is Model B (2~4GB, good accuracy, downgraded choice), and the first preset model is Model A (1~2GB, does not meet the functional requirements of this command). When the diagnostic equipment detects that the second available memory is 3GB, rule matching judgment is performed: The second available memory (3GB) is less than the minimum memory requirement of the second preset model (Model C) (4GB), thus not meeting the optimal matching condition; the second available memory (3GB) is greater than the minimum memory requirement of the third preset model (Model B) (2GB), meeting the downgrade matching condition; the memory requirement of the third preset model (Model B) (2~4GB) is greater than that of the first preset model (Model A, 1~2GB) and less than that of the second preset model (Model C, 4~8GB); the performance (good accuracy) of the third preset model (Model B) is greater than that of the first preset model (Model A, average accuracy) and less than that of the second preset model (Model C, excellent accuracy). According to the rules, the third preset model (Model B) is determined as the target preset model. Under conditions of insufficient memory resources, it ensures the normal processing of diagnostic function instructions and has better performance than the first preset model.

[0066] In another instance, the technician issues the voice command "Initiate rapid diagnostics to analyze the cause of insufficient vehicle power." This is identified as a smart diagnostic function command in the second type. To match the high-performance requirements of this command, the second preset model is pre-set as Model D (8~12GB, highest accuracy, optimal choice), the third preset model is Model C (4~8GB, excellent accuracy, downgraded choice), and the first preset model is Model A (1~2GB, does not meet the functional requirements of this command). The diagnostic equipment detects 6GB of available memory. Rule matching judgment: The second available memory (6GB) is less than the minimum memory requirement of the second preset model (Model D) (8GB), thus failing to meet the optimal matching condition; the second available memory (6GB) is greater than the minimum memory requirement of the third preset model (Model C) (4GB), thus meeting the downgrade matching condition; the memory requirement of the third preset model (Model C) (4~8GB) is greater than that of the first preset model (Model A, 1~2GB) and less than that of the second preset model (Model D, 8~12GB); the performance (excellent accuracy) of the third preset model (Model C) is greater than that of the first preset model (Model A, average accuracy) and less than that of the second preset model (Model D, highest accuracy). According to the rules, the third preset model (Model C) is determined as the target preset model. Even when memory resources cannot support the optimal model, reliable processing of intelligent diagnostic function instructions is achieved, and its performance is far superior to that of the first preset model.

[0067] In the above method, a third preset model with intermediate performance is added as an alternative for the second type of instructions. When the device memory is insufficient to support the second preset model, the system automatically switches to this model. This not only avoids the problem that complex instructions cannot be processed due to insufficient resources, but also achieves flexible adaptation between performance and memory resources, ensuring the stability and continuity of the processing of the second type of instructions.

[0068] S103, the voice control command is sent to the target preset model so that the target preset model performs semantic conversion on the voice control command and obtains the structured command data output by the target preset model.

[0069] In this application, after determining the target preset model for processing voice control commands, the diagnostic device sends the voice control command (preprocessed digital audio signal) to the target preset model so that the target preset model performs a dedicated semantic conversion operation on the voice control command. Finally, the target preset model outputs structured command data that the device system can directly recognize and execute, providing a basis for triggering and executing subsequent diagnostic functions.

[0070] For example, after the model matching module of the diagnostic device completes the determination of the target preset model and generates a start signal, the device's instruction transmission module immediately receives the start signal and the digital audio signal of the voice control instruction after lightweight preprocessing. Through the device's built-in model communication interface, the voice control instruction is sent to the activated target preset model. The target preset model can be any one of the first preset model (model A), the second preset model (model B / C / D), or the third preset model (model B / C).

[0071] Upon receiving a voice control command, the target preset model immediately invokes its proprietary speech recognition algorithm and speech escaping method to perform a complete semantic conversion process on the command. Specifically, Model A performs lightweight frame matching recognition and keyword mapping escaping, Model B performs optimized dynamic time warping recognition and phrase-level escaping, Model C performs deep neural network recognition and sentence-level escaping, and Model D performs recognition and context-related escaping using a combination of convolutional and recurrent neural networks. Finally, the target preset model outputs standardized structured command data containing command function type labels, specific operation codes, and execution priorities, which is then fed back to the system execution module of the diagnostic device through the model communication interface.

[0072] S104 executes structured instruction data in response to user control requests.

[0073] In this embodiment, after the system execution module of the diagnostic device receives the structured instruction data output by the target preset model, it will automatically trigger and execute the corresponding diagnostic device function according to the operation code, function type and execution priority contained in the data, so as to accurately and efficiently respond to the user's (technician's) voice control needs and complete the entire voice control process loop.

[0074] In the above method, the diagnostic device first receives natural language control commands (voice control commands) issued by the user through the audio acquisition module. Then, based on the function type of the command, it matches and adapts a target preset model. Subsequently, the command is sent to the model, which uses a corresponding algorithm to complete the semantic conversion from natural language to structured command data. Finally, the device parses and executes the structured data to fulfill the user's control needs. This method uses precise "command-model" matching to allow simple commands to be processed quickly using lightweight models (low memory, low computational load), avoiding resource redundancy. At the same time, structured command data can be directly executed by the system, shortening the execution chain. Moreover, high-complexity commands are matched with high-performance models (more sophisticated semantic recognition algorithms), and different models are specifically adapted to corresponding command scenarios, reducing generalization errors and making the semantic understanding of different types of commands more accurate.

[0075] In one embodiment, see Figure 5 This is a flowchart illustrating the response to control requests provided in an embodiment of this application, such as... Figure 5 As shown, step S104 includes: S501, when the instruction type is type 1, executes structured instruction data to respond to the user's control requirements.

[0076] In this embodiment, when the diagnostic device initially determines the received voice control command as the first type (basic function type), and obtains the corresponding structured command data through model matching and semantic conversion, it will directly trigger the system execution module to execute the structured command data, thereby quickly and efficiently responding to the user's (technician's) control needs for the basic functions of the diagnostic device.

[0077] For example, when the instruction type determination module of the diagnostic device determines the type label of the voice control instruction as the first type (basic function type), and determines the first preset model (A model) as the target preset model through model matching, and then the A model performs semantic conversion on the instruction and outputs the corresponding standardized structured instruction data, the system execution module of the device will immediately identify the first type instruction exclusive identifier in the structured instruction data, and directly parse the data without waiting for the execution of other priority instructions.

[0078] The system extracts the basic function operation codes and related parameters contained therein, and then calls the corresponding function modules in the device's built-in basic function execution library to perform basic function operations such as opening diagnostic reports, returning to the operation homepage, printing test data, and switching display interfaces. The entire execution process relies on the fast processing characteristics of the A model, and has the characteristics of fast execution speed and low system resource consumption. Finally, through the collaborative work of hardware such as the diagnostic device's display screen and printer, the corresponding basic function operations are completed, accurately responding to the user's voice control needs for the basic functions of the diagnostic device.

[0079] S502, when the instruction type is type 2, obtains the vehicle data required for the structured instruction data.

[0080] In this embodiment, the diagnostic device initially determines the received voice control command as the second type, and after obtaining the corresponding structured command data through model matching and semantic conversion, it will first trigger a dedicated data acquisition process. The core implementation method is that the diagnostic device establishes a wireless connection with the vehicle communication interface (VCI) through the built-in Bluetooth communication module, and collects the vehicle-related data necessary for executing the structured command data from the vehicle's bus system through the VCI, providing core data support for the accurate execution of subsequent commands.

[0081] Specifically, when the diagnostic device's instruction type determination module identifies the voice control instruction type label as the second type, and determines the second or third preset model as the target preset model through model matching, and then the corresponding target preset model performs semantic conversion on the instruction to output the corresponding standardized structured instruction data, the device's system execution module will immediately identify the second type instruction exclusive identifier in the structured instruction data, and then trigger the built-in vehicle data acquisition module. This module first starts the diagnostic device's Bluetooth communication module to establish a stable wireless communication connection with the pre-paired vehicle communication interface (VCI). Then, according to the vehicle data requirement parameters carried in the structured instruction data, it generates standardized data acquisition instructions and sends them to the VCI through the Bluetooth communication link. The VCI converts the acquisition instructions into signals that can be recognized by the vehicle bus system, and selectively acquires the corresponding data from the vehicle's ECU, sensors, and other hardware. Among them, diagnostic function instructions correspond to acquiring basic vehicle data such as real-time data streams of a single system, fault codes, and system operating status.

[0082] Intelligent diagnostic function commands acquire in-depth vehicle data such as multi-system joint operation data, historical fault records, component factory parameters, and real-time sensor data. The VCI transmits the collected vehicle data back to the diagnostic equipment via Bluetooth communication link. The diagnostic equipment's data acquisition module performs format standardization, integrity verification, and outlier filtering on the transmitted data to ensure the accuracy and availability of the data. Finally, the verified vehicle data is associated and stored with the structured command data, and a trigger signal for data acquisition completion is generated, providing sufficient and reliable data support for the execution of subsequent second-type commands.

[0083] The S503 processes vehicle data to respond to the user's control needs.

[0084] In this embodiment, after the diagnostic device obtains the vehicle data required to execute structured instructions via Bluetooth+VCI, it processes the vehicle data accordingly (basic analysis / deep analysis) based on the specific subtype of the instruction (diagnostic function / intelligent diagnostic function), and finally outputs the processing result, thereby accurately responding to the user's control needs.

[0085] For example, in the case of the second type of diagnostic function, such as the voice control command "Read the engine's real-time fault codes", the diagnostic device communicates with the VCI via Bluetooth to obtain the original fault code data (such as P0301), fault occurrence time, and number of fault occurrences from the engine ECU. The diagnostic device then performs basic analysis and standardization on the obtained original vehicle data: converting the original fault code P0301 into a user-understandable Chinese description "cylinder 1 misfire"; associating the fault occurrence time and number of occurrences to generate basic fault statistics; and formatting the analyzed data according to the diagnostic device's standard fault report format. After processing, the diagnostic device displays the results in a clear list format on the display screen. Engine real-time fault code detection results Fault code: P0301 Fault description: Cylinder 1 misfire Last occurred: 2026-01-08 10:25 Total number of occurrences: 5 In another exemplary case, under the second type of intelligent diagnostic function, such as the voice control command "Start rapid diagnostics to analyze the cause of insufficient vehicle power," the diagnostic device communicates with the VCI via Bluetooth and obtains multi-dimensional vehicle data from multiple related system ECUs, including real-time engine data streams (intake pressure, fuel injection quantity, ignition advance angle), transmission gear shift data, fuel pump pressure data, air flow meter data, and historical fault records. The diagnostic device performs in-depth joint analysis and reasoning processing on the acquired multi-dimensional vehicle data: First, it cross-validates the data from each system to filter out abnormal data (such as intake pressure below the standard value or insufficient fuel pump pressure); second, based on the built-in intelligent diagnostic algorithm, it establishes a causal relationship between abnormal data and insufficient power faults; finally, combined with historical fault records, it eliminates occasional fault factors, determines the core fault cause, and generates corresponding repair suggestions.

[0086] After processing, the diagnostic device displays the results on the screen in the form of an intelligent diagnostic report, and also supports voice broadcast of the key conclusions: Vehicle underpower intelligent diagnostic report I. Causes of the core failure Fuel system malfunction: The fuel pump pressure is 2.0 Bar, which is lower than the standard value of 3.0 Bar, resulting in insufficient fuel supply to the engine.

[0087] Intake system auxiliary malfunction: The air flow meter data fluctuates greatly, which may be due to a loose sensor.

[0088] II. Repair Recommendation Priority First, check and replace the fuel pump to restore fuel pressure.

[0089] Check the installation of the air flow meter, and calibrate or replace it if necessary.

[0090] III. Related Data Anomalies (with a list of key abnormal data for the engine and fuel system and a comparison with standard values).

[0091] In the above method, a layered execution logic is adopted for different types of instructions. Basic instructions are executed directly to ensure response efficiency; diagnostic instructions first acquire vehicle data and then process it to ensure the accuracy and professionalism of instruction execution, thus achieving optimal adaptation of instruction execution in different scenarios.

[0092] In one embodiment, see Figure 6 This is a schematic diagram of the model optimization process provided in the embodiments of this application, such as... Figure 6 As shown, it includes: S601, acquires user feedback data regarding the operation and diagnostic equipment.

[0093] In this embodiment, after the diagnostic device responds to the user's voice control request (performing basic functions or processing vehicle data), it triggers a dedicated feedback data collection process to actively or passively acquire the user's feedback data throughout the entire process of this voice control of the diagnostic device. This feedback data will be used for subsequent optimization of preset model performance, improvement of voice recognition accuracy, and iteration of command execution logic, thereby continuously improving the user experience of voice control.

[0094] In addition to collecting basic user feedback data, we will also collect multi-dimensional customer feedback information, including newly added corpora, optimization suggestions for diagnostic business processing modules, diagnostic function requirements, and suggestions for improving the dynamic adjustment mechanism. We will use this multi-dimensional feedback information as core training data to systematically train the original preset model, thereby achieving comprehensive optimization of the model in terms of speech recognition, business adaptation, and functional response, making it more in line with the actual diagnostic business needs.

[0095] S602, optimize the preset model based on the feedback data to obtain the optimized preset model.

[0096] In this embodiment, the model optimization module uses the collected multi-dimensional customer feedback data as the core input. Through targeted model training and parameter tuning, it optimizes the original four groups of gradient preset models (A / B / C / D) to obtain an optimized preset model that improves speech recognition accuracy, diagnostic business adaptability, functional response accuracy, and dynamic adjustment flexibility. This step realizes a complete technical closed loop from customer feedback collection to model performance iteration, ensuring that the preset model continuously meets the needs of actual diagnostic business.

[0097] S603 performs semantic conversion on the next voice control command based on the optimized preset model to obtain the structured command data corresponding to the next voice control command.

[0098] In this embodiment of the application, after the diagnostic device completes the optimization of the preset model and obtains the optimized preset model, it will apply the optimized model to the processing flow of the next voice control command. Specifically, after completing the command type determination and model matching for the next voice control command, the corresponding optimized preset model is called to perform semantic conversion on the voice control command, and finally obtains the more accurate structured command data corresponding to the next voice control command, so as to realize the real-time implementation of the optimization effect.

[0099] S604 controls the diagnostic device based on the next structured instruction data.

[0100] In this embodiment, after the system execution module of the diagnostic device receives the next structured instruction data output by the optimized preset model, it will trigger the corresponding execution logic layer by layer according to the instruction type, operation code and function parameters contained in the data, and accurately control the diagnostic device to complete the corresponding function. This step is the final manifestation of the model optimization effect. Compared with the original model, the accuracy of its operation, response efficiency and business adaptability are significantly improved. At the same time, it strictly follows the layered logic of directly executing the first type of instruction and first obtaining vehicle data and then processing the second type of instruction.

[0101] The above method constructs a complete closed loop of "feedback collection - model optimization - iterative application". By continuously optimizing the preset model through user feedback and implementing it in subsequent command processing, the performance of the voice control system is dynamically and iteratively improved, effectively enhancing the response speed and accuracy of intelligent voice AI.

[0102] See Figure 7 This is a summary structural block diagram of the voice interaction processing method based on diagnostic equipment provided in the embodiments of this application. The specific steps are as follows: 1. The process starts from the "Start" node and first receives the user's voice control commands.

[0103] 2. Instruction preprocessing The received instructions are "instruction-classified" to determine their type; Based on the instruction type, perform preliminary preparations for "matching the preset model".

[0104] 3. Instruction type layering determination Check the command type in sequence: First, determine if it is a basic function instruction; If not, then determine whether it is a diagnostic function instruction; If it still doesn't, then determine whether it's a smart diagnostic function instruction.

[0105] 4. Classification + Memory Matching Model Based on the instruction type and the available memory detection results, the target model is determined: Basic function instructions: Check if the available memory is greater than the preset model A; if so, select model A. Diagnostic function commands: First check if the available memory is greater than the C preset model; if so, select the C model. If not, then check if it is greater than the B preset model; if so, select the B model. Intelligent diagnostic function commands: First, check if the available memory is greater than the preset model D. If yes, select model D. If no, check if it is greater than the preset model C. If yes, select model C.

[0106] 5. Generate structured instructions The selected model (A / B / C / D) is used to perform semantic conversion on the voice commands to obtain structured command data. Finally, the structured command data is executed to fulfill the user's voice control needs.

[0107] It should be understood that the sequence number of each step in the above embodiments does not imply 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 the embodiments of this application.

[0108] Corresponding to the voice interaction processing method based on diagnostic equipment in the above embodiment, Figure 8 This is a structural block diagram of a voice interaction processing device based on a diagnostic device provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0109] Reference Figure 8 The device includes: The instruction receiving module 81 is used to receive voice control instructions from the user to operate the diagnostic equipment; Model matching module 82 is used to match the corresponding target preset model according to the voice control command; The semantic conversion module 83 is used to send voice control commands to the target preset model so that the target preset model can perform semantic conversion on the voice control commands and obtain the structured command data output by the target preset model. The device control module 84 is used to execute structured instruction data in response to the user's control needs.

[0110] Optionally, the model matching module 82 is also used for: Determine the type of voice control command; Match the corresponding target preset model based on the instruction type.

[0111] Optionally, the model matching module 82 is also used for: When the instruction type is type 1, detect the first available memory of the diagnostic device; If the available memory is greater than the first preset model, the first preset model is determined as the target preset model; wherein, the first preset model is the preset model with the lowest memory requirement among multiple preset models.

[0112] Optionally, the model matching module 82 is also used for: In the case of instruction type 2, detect the second available memory of the diagnostic device; If the second available memory is greater than the second preset model, the second preset model is determined as the target preset model; wherein, the memory requirement of the second preset model is higher than that of the first preset model, and the performance of the second preset model is higher than that of the first preset model; the functions of the second type are higher than those of the first type.

[0113] Optionally, the model matching module 82 is also used for: If the second available memory is less than the second preset model but greater than the third preset model, the third preset model is determined as the target preset model; wherein, the memory requirement of the third preset model is greater than that of the first preset model but less than that of the second preset model; and the performance of the third preset model is higher than that of the first preset model but less than that of the second preset model.

[0114] Optionally, the device control module 84 is also used for: If the instruction type is type 1, then structured instruction data is executed in response to the user's control requirements; If the instruction type is type 2, then the vehicle data required for the structured instruction data is obtained; Vehicle data is processed to respond to user control needs.

[0115] The voice interaction processing device 8 based on diagnostic equipment also includes a model optimization module 85, used for: Obtain user feedback data regarding the operation and diagnostic equipment; The preset model is optimized based on the feedback data to obtain the optimized preset model; Based on the optimized preset model, the next voice control command is semantically transformed to obtain the structured command data corresponding to the next voice control command. The diagnostic equipment is controlled based on the next structured instruction data.

[0116] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0117] in addition, Figure 8 The voice interaction processing device based on the diagnostic equipment shown can be a software unit, a hardware unit, or a combination of software and hardware built into the existing terminal device. It can also be integrated into the terminal device as an independent component, or exist as a standalone terminal device.

[0118] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0119] Figure 9 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. For example... Figure 9 As shown, the terminal device 9 of this embodiment includes: at least one processor 90 ( Figure 9 (Only one is shown in the image) a processor, a memory 91, and a computer program 92 stored in the memory 91 and capable of running on at least one processor 90. When the processor 90 executes the computer program 92, it implements the steps in any of the above embodiments of the voice interaction processing method based on diagnostic devices.

[0120] The terminal device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 9 This is merely an example of terminal device 9 and does not constitute a limitation on terminal device 9. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0121] The processor 90 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0122] In some embodiments, memory 91 may be an internal storage unit of terminal device 9, such as a hard disk or memory of terminal device 9. In other embodiments, memory 91 may be an external storage device of terminal device 9, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD) card, flash card, etc., equipped on terminal device 9. Furthermore, memory 91 may include both internal storage units and external storage devices of terminal device 9. Memory 91 is used to store operating system, application programs, boot loader, data, and other programs, such as program code of computer programs. Memory 91 may also be used to temporarily store data that has been output or will be output.

[0123] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the steps in the above-described method embodiments.

[0124] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0125] If the integrated unit is implemented as 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, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0126] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0127] Those skilled in the art will recognize that the units and algorithm steps of the various examples 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 implemented in hardware or software 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 beyond the scope of this application.

[0128] In the embodiments provided in this application, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A voice interaction processing method based on diagnostic equipment, characterized in that, Applied to diagnostic equipment; the diagnostic equipment includes different preset models; each preset model has different processing speed and processing accuracy; the method includes: Receives voice control commands from the user to operate the diagnostic device; Match the corresponding target preset model according to the voice control command; The voice control command is sent to the target preset model, so that the target preset model performs semantic conversion on the voice control command to obtain the structured command data output by the target preset model; The structured instruction data is executed in response to the user's control requirements.

2. The voice interaction processing method based on diagnostic equipment as described in claim 1, characterized in that, The step of matching the corresponding target preset model according to the voice control command includes: Determine the command type of the voice control command; Match the corresponding target preset model according to the instruction type.

3. The voice interaction processing method based on diagnostic equipment as described in claim 2, characterized in that, The step of matching the corresponding target preset model according to the instruction type includes: If the instruction type is the first type, detect the first available memory of the diagnostic device; If the first available memory is greater than the first preset model, the first preset model is determined as the target preset model; wherein, the first preset model is the preset model with the lowest memory requirement among the multiple preset models.

4. The voice interaction processing method based on diagnostic equipment as described in claim 3, characterized in that, The method includes: In the case where the instruction type is the second type, the second available memory of the diagnostic device is detected; If the second available memory is greater than the second preset model, the second preset model is determined as the target preset model; wherein, the memory requirement of the second preset model is higher than that of the first preset model, and the performance of the second preset model is higher than that of the first preset model; the functionality of the second type is higher than that of the first type.

5. The voice interaction processing method based on diagnostic equipment as described in claim 4, characterized in that, The method includes: If the second available memory is less than the second preset model but greater than the third preset model, the third preset model is determined as the target preset model; wherein, the memory requirement of the third preset model is greater than that of the first preset model and less than that of the second preset model; and the performance of the third preset model is higher than that of the first preset model and less than that of the second preset model.

6. The voice interaction processing method based on diagnostic equipment as described in claim 5, characterized in that, The execution of the structured instruction data in response to the user's control requirements includes: If the instruction type is the first type, then the structured instruction data is executed to respond to the user's control requirements; If the instruction type is the second type, then the vehicle data required to obtain the structured instruction data is obtained; The vehicle data is processed to respond to the user's control needs.

7. The voice interaction processing method based on diagnostic equipment as described in claim 1, characterized in that, The method further includes: Obtain user feedback data regarding the operation of the diagnostic device; The preset model is optimized based on the feedback data to obtain the optimized preset model; Based on the optimized preset model, the next voice control command is semantically transformed to obtain the structured command data corresponding to the next voice control command. The diagnostic device is controlled according to the next structured instruction data.

8. A voice interaction processing device based on diagnostic equipment, characterized in that, include: The instruction receiving module is used to receive voice control instructions from the user to operate the diagnostic device; The model matching module is used to match the corresponding target preset model according to the voice control command; A semantic conversion module is used to send the voice control command to the target preset model, so that the target preset model performs semantic conversion on the voice control command to obtain the structured command data output by the target preset model; The device control module is used to execute the structured instruction data in response to the user's control requirements.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images