Vehicle-mounted voice complaint processing method, device and equipment, and storage medium

Abstract

The application discloses a vehicle-mounted voice complaint processing method and device, equipment and storage medium, the method receives the complaint instruction, uses the generative model of integrated vehicle knowledge base to analyze the complaint instruction, generates the analysis result containing the disposal strategy; according to the disposal strategy in the analysis result, drive business execution, and the flow state generated in the execution process is synchronized to the after-sales management system in real time; the flow state in the after-sales management system is monitored, when detecting that the flow state changes the preset closed loop condition, feedback notification is generated and is pushed to the user through the vehicle-mounted terminal, the accuracy of complaint response and the automation processing efficiency can be significantly improved, the lag and deviation caused by manual intervention are avoided, the traceability and transparency of processing progress are ensured, the whole-process closed loop management from complaint acceptance to result feedback is realized, and the user vehicle experience and after-sales service quality are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent connected vehicle technology, and in particular to an in-vehicle voice customer complaint processing method, device, equipment, and storage medium. Background Technology

[0002] Currently, in-vehicle voice interaction has become widespread, serving as a core standard feature of smart cockpits and a mainstream entry point for human-vehicle interaction.

[0003] With the rapid iteration of intelligent vehicle technology, in-vehicle voice assistants are continuously upgrading their capabilities in command recognition, function coverage, and adaptation to complex scenarios, which can meet multi-dimensional car usage needs such as navigation and audio-visual, body control, and air conditioning adjustment.

[0004] At present, mainstream car manufacturers have successively installed voice-based customer complaint handling modules, but existing solutions mostly rely on users to actively submit problems via voice or text.

[0005] Because users have varying levels of expressive ability, their problem descriptions are often fragmented, vague, and fail to get to the heart of the matter. After complaint information is uploaded to the cloud, it still needs to be manually reviewed, classified, and archived one by one, making the process cumbersome and inefficient, and making it difficult to achieve rapid closed-loop management of problems. Summary of the Invention

[0006] The main objective of this invention is to provide a method, apparatus, device, and storage medium for handling in-vehicle voice customer complaints, aiming to solve the technical problems in the prior art where in-vehicle customer complaint handling relies on manual labor, resulting in low efficiency; lack of vehicle knowledge base support, leading to inaccurate responses; and the inability to synchronize processing status in real time, making it difficult to achieve closed-loop feedback.

[0007] In a first aspect, the present invention provides a method for handling in-vehicle voice customer complaints, the method comprising the following steps: Upon receiving a customer complaint instruction, the system uses a generative model with an integrated vehicle knowledge base to parse the complaint instruction and generate analysis results containing handling strategies. The business execution is driven by the handling strategies in the analysis results, and the flow status generated during the execution process is synchronized to the after-sales management system in real time. The system monitors the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user via the vehicle terminal.

[0008] Optionally, the step of receiving a customer complaint instruction and parsing the instruction using a generative model integrated with a vehicle knowledge base to generate an analysis result containing a handling strategy includes: Receive customer complaint instructions in natural language form issued by users through the vehicle terminal; A customized generative artificial intelligence model is used to perform semantic understanding and key information extraction on the customer complaint instructions; Based on the extracted key information and the vehicle's official manual knowledge base, knowledge matching and reasoning are performed to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidelines, configuration instructions, or work order creation suggestions.

[0009] Optionally, the extraction of key information is used for knowledge matching and reasoning with the vehicle's official manual knowledge base to generate structured analysis results, including: The extracted key information is converted into a vector representation, and target knowledge fragments with a similarity higher than a preset similarity threshold are retrieved from the vehicle's official manual knowledge base. The key information and the target knowledge fragment are assembled into prompt words, and the prompt words are input into the customized generative artificial intelligence model; The customized generative artificial intelligence model performs reasoning based on the prompt words to generate structured analysis results.

[0010] Optionally, the step of driving business execution based on the handling strategy in the analysis results and synchronizing the flow status generated during the execution process to the after-sales management system in real time includes: Parse the strategy type of the handling strategy in the analysis results, and call the business interface that matches the strategy type to execute the corresponding business operation; Monitor node changes during the execution of the business operation and generate a flow status that represents the progress of the business; The circulation status is pushed to the after-sales management system in real time via the data interface for updates.

[0011] Optionally, monitoring node changes during the execution of the business operation and generating a flow status representing the business progress includes: Real-time capture of state change events triggered during the execution of the business operations; Extract the current business stage identifier and change time information from the status change event; Based on the current business stage identifier and the change time information, a flow status representing the business progress is encapsulated and generated.

[0012] Optionally, the monitoring of the circulation status in the after-sales management system, when a preset closed-loop condition for a change in the circulation status is detected, generates a feedback notification and pushes it to the user via the vehicle terminal, including: The system periodically polls or monitors changes in the circulation status of the after-sales management system through a preset data interface. When the flow state is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered; A feedback notification containing specific processing results is constructed based on the notification generation mechanism, and the feedback notification is pushed to the user through the vehicle terminal.

[0013] Optionally, the step of constructing a feedback notification containing specific processing results according to the notification generation mechanism, and pushing the feedback notification to the user through the vehicle terminal, includes: The notification generation mechanism is used to parse the status type of the flow state and match the notification content template corresponding to the status type. Obtain the specific processing result associated with the flow status, fill the specific processing result into the notification content template, and generate a feedback notification; The message push channel of the vehicle terminal is determined, and the feedback notification is sent to the vehicle terminal through the message push channel.

[0014] Secondly, to achieve the above objectives, the present invention also proposes an in-vehicle voice customer complaint processing device, the in-vehicle voice customer complaint processing device comprising: The instruction parsing module is used to receive customer complaint instructions, parse the customer complaint instructions using a generative model with an integrated vehicle knowledge base, and generate analysis results containing handling strategies. The business execution module is used to drive business execution based on the handling strategies in the analysis results, and to synchronize the flow status generated during the execution process to the after-sales management system in real time. The monitoring and feedback module is used to monitor the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user through the vehicle terminal.

[0015] Thirdly, to achieve the above objectives, the present invention also proposes an in-vehicle voice customer complaint processing device, the in-vehicle voice customer complaint processing device comprising: a memory, a processor, and an in-vehicle voice customer complaint processing program stored in the memory and executable on the processor, the in-vehicle voice customer complaint processing program being configured to implement the steps of the in-vehicle voice customer complaint processing method described above.

[0016] Fourthly, to achieve the above objectives, the present invention also proposes a storage medium storing an in-vehicle voice customer complaint processing program, wherein when the in-vehicle voice customer complaint processing program is executed by a processor, it implements the steps of the in-vehicle voice customer complaint processing method described above.

[0017] The in-vehicle voice customer complaint handling method proposed in this invention receives customer complaint instructions, parses the instructions using a generative model integrated with a vehicle knowledge base, and generates analysis results containing handling strategies. Based on the handling strategies in the analysis results, business execution is driven, and the flow status generated during execution is synchronized to the after-sales management system in real time. The flow status in the after-sales management system is monitored, and when a change in the flow status is detected under preset closed-loop conditions, a feedback notification is generated and pushed to the user via the in-vehicle terminal. This method, through intelligent parsing and strategy generation of customer complaint instructions using a generative model integrated with a vehicle knowledge base, significantly improves the accuracy of customer complaint response and the efficiency of automated processing, avoiding delays and deviations caused by manual intervention. Simultaneously, through real-time synchronization of business execution status with the after-sales management system and monitoring of closed-loop conditions, the traceability and transparency of processing progress are ensured, achieving closed-loop management of the entire process from complaint acceptance to result feedback, effectively improving the user's driving experience and after-sales service quality. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the vehicle-mounted voice customer complaint processing method of the present invention. Figure 3 This is a flowchart illustrating the second embodiment of the vehicle-mounted voice customer complaint processing method of the present invention; Figure 4 This is a flowchart illustrating the third embodiment of the vehicle-mounted voice customer complaint processing method of the present invention; Figure 5 This is a flowchart illustrating the fourth embodiment of the vehicle-mounted voice customer complaint processing method of the present invention; Figure 6 This is an example diagram of the customer complaint feedback dialogue in the vehicle-mounted voice customer complaint processing method of the present invention; Figure 7 This is a functional block diagram of the first embodiment of the vehicle-mounted voice customer complaint processing device of the present invention.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0021] The solution of this invention mainly involves: receiving customer complaint instructions, parsing the instructions using a generative model integrated with a vehicle knowledge base, and generating analysis results containing handling strategies; driving business execution based on the handling strategies in the analysis results, and synchronizing the flow status generated during execution to the after-sales management system in real time; monitoring the flow status in the after-sales management system, and generating a feedback notification when a change in the flow status is detected under preset closed-loop conditions, and pushing it to the user through the vehicle terminal. This intelligently parses and generates strategies for customer complaint instructions using a generative model integrated with a vehicle knowledge base, significantly improving the accuracy of customer complaint responses and the efficiency of automated processing, avoiding delays and deviations caused by manual intervention. Simultaneously, through real-time synchronization of business execution status with the after-sales management system and monitoring of closed-loop conditions, the traceability and transparency of processing progress are ensured, achieving closed-loop management of the entire process from complaint acceptance to result feedback. This effectively improves the user's vehicle experience and after-sales service quality, solving the technical problems of inefficient vehicle-based customer complaint handling due to reliance on manual labor, inaccurate responses due to lack of vehicle knowledge base support, and difficulty in achieving closed-loop feedback due to the inability to synchronize processing status in real time in existing technologies.

[0022] Reference Figure 1 , Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0023] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0024] Those skilled in the art will understand that Figure 1 The device structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0025] like Figure 1As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and an in-vehicle voice customer complaint processing program.

[0026] The device of the present invention calls the in-vehicle voice customer complaint processing program stored in the memory 1005 through the processor 1001 and performs the following operations: Upon receiving a customer complaint instruction, the system uses a generative model with an integrated vehicle knowledge base to parse the complaint instruction and generate analysis results containing handling strategies. The business execution is driven by the handling strategies in the analysis results, and the flow status generated during the execution process is synchronized to the after-sales management system in real time. The system monitors the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user via the vehicle terminal.

[0027] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: Receive customer complaint instructions in natural language form issued by users through the vehicle terminal; A customized generative artificial intelligence model is used to perform semantic understanding and key information extraction on the customer complaint instructions; Based on the extracted key information and the vehicle's official manual knowledge base, knowledge matching and reasoning are performed to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidelines, configuration instructions, or work order creation suggestions.

[0028] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: The extracted key information is converted into a vector representation, and target knowledge fragments with a similarity higher than a preset similarity threshold are retrieved from the vehicle's official manual knowledge base. The key information and the target knowledge fragment are assembled into prompt words, and the prompt words are input into the customized generative artificial intelligence model; The customized generative artificial intelligence model uses the prompt words to perform reasoning and generate structured analysis results.

[0029] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: Parse the strategy type of the handling strategy in the analysis results, and call the business interface that matches the strategy type to execute the corresponding business operation; Monitor node changes during the execution of the business operation and generate a flow status that represents the progress of the business; The circulation status is pushed to the after-sales management system in real time via the data interface for updates.

[0030] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: Real-time capture of state change events triggered during the execution of the business operations; Extract the current business stage identifier and change time information from the status change event; Based on the current business stage identifier and the change time information, a flow status representing the business progress is encapsulated and generated.

[0031] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: The system periodically polls or monitors changes in the circulation status of the after-sales management system through a preset data interface. When the flow state is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered; A feedback notification containing specific processing results is constructed based on the notification generation mechanism, and the feedback notification is pushed to the user through the vehicle terminal.

[0032] The device of the present invention, through processor 1001 calling the in-vehicle voice customer complaint processing program stored in memory 1005, also performs the following operations: The notification generation mechanism is used to parse the status type of the flow state and match the notification content template corresponding to the status type. Obtain the specific processing result associated with the flow status, fill the specific processing result into the notification content template, and generate a feedback notification; The message push channel of the vehicle terminal is determined, and the feedback notification is sent to the vehicle terminal through the message push channel.

[0033] This embodiment, through the above-described scheme, receives customer complaint instructions, parses the instructions using a generative model integrated with a vehicle knowledge base, and generates analysis results containing handling strategies. Based on the handling strategies in the analysis results, business execution is driven, and the flow status generated during execution is synchronized to the after-sales management system in real time. The flow status in the after-sales management system is monitored, and when a change in the flow status is detected under preset closed-loop conditions, a feedback notification is generated and pushed to the user via the in-vehicle terminal. This intelligent parsing and strategy generation of customer complaint instructions through a generative model integrated with a vehicle knowledge base significantly improves the accuracy of customer complaint response and the efficiency of automated processing, avoiding delays and deviations caused by manual intervention. Simultaneously, through real-time synchronization of business execution status with the after-sales management system and monitoring of closed-loop conditions, the traceability and transparency of processing progress are ensured, achieving closed-loop management of the entire process from complaint acceptance to result feedback, effectively improving the user's vehicle experience and the quality of after-sales service.

[0034] Based on the above hardware structure, an embodiment of the vehicle-mounted voice customer complaint processing method of the present invention is proposed.

[0035] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the vehicle-mounted voice customer complaint processing method of the present invention.

[0036] In the first embodiment, the in-vehicle voice customer complaint processing method includes the following steps: Step S10: Receive customer complaint instructions, parse the customer complaint instructions using a generative model with an integrated vehicle knowledge base, and generate analysis results containing handling strategies.

[0037] It should be noted that after receiving a customer complaint instruction, the received instruction can be parsed, and the vehicle knowledge base can be used as the basis for model generation to obtain analysis results containing handling strategies. This makes the process of generating handling strategies constrained by the vehicle knowledge base, thus realizing the generation of handling strategies from customer complaint instructions.

[0038] Step S20: Drive business execution based on the handling strategy in the analysis results, and synchronize the flow status generated during the execution process to the after-sales management system in real time.

[0039] It should be understood that by transforming the handling strategies in the analysis results into executable business actions and capturing status change data during the execution of the actions, a real-time data link between the business execution process and the after-sales management system is established, enabling the after-sales management system to obtain the business flow progress in real time and achieving synchronization and consistency between strategy execution and status tracking.

[0040] Step S30: Monitor the circulation status in the after-sales management system. When a preset closed-loop condition for a change in circulation status is detected, generate a feedback notification and push it to the user through the vehicle terminal.

[0041] Understandably, by tracking the business progress data of the after-sales management system and identifying whether it meets the closed-loop completion requirements, a synchronous link between the backend processing status and the frontend user perception is established, so that the notification generation is bound to the actual processing progress. Users can learn about the final conclusion of the customer complaint handling through the vehicle interface without actively querying, thus realizing the closed-loop notification of the service process.

[0042] This embodiment, through the above-described scheme, receives customer complaint instructions, parses the instructions using a generative model integrated with a vehicle knowledge base, and generates analysis results containing handling strategies. Based on the handling strategies in the analysis results, business execution is driven, and the flow status generated during execution is synchronized to the after-sales management system in real time. The flow status in the after-sales management system is monitored, and when a change in the flow status is detected under preset closed-loop conditions, a feedback notification is generated and pushed to the user via the in-vehicle terminal. This intelligent parsing and strategy generation of customer complaint instructions through a generative model integrated with a vehicle knowledge base significantly improves the accuracy of customer complaint response and the efficiency of automated processing, avoiding delays and deviations caused by manual intervention. Simultaneously, through real-time synchronization of business execution status with the after-sales management system and monitoring of closed-loop conditions, the traceability and transparency of processing progress are ensured, achieving closed-loop management of the entire process from complaint acceptance to result feedback, effectively improving the user's vehicle experience and the quality of after-sales service.

[0043] Furthermore, Figure 3 This is a flowchart illustrating the second embodiment of the in-vehicle voice customer complaint processing method of the present invention, as shown below. Figure 3 As shown, based on the first embodiment, a second embodiment of the in-vehicle voice customer complaint processing method of the present invention is proposed. In this embodiment, step S10 specifically includes the following steps: Step S11: Receive customer complaint instructions in natural language form sent by the user through the vehicle terminal.

[0044] It should be noted that by capturing customer complaint instructions issued by users in natural language form through the vehicle terminal, an interactive link is established between the user's subjective demands and the system processing entry point.

[0045] Step S12: Use a customized generative artificial intelligence model to perform semantic understanding and key information extraction on the customer complaint instructions.

[0046] Understandably, by using a customized generative artificial intelligence model to perform semantic understanding and key information extraction on customer complaint instructions, a mapping relationship between natural language instructions and core information elements is established, enabling the model to parse the semantics of instructions and extract key content.

[0047] Step S13: Based on the extracted key information and the vehicle's official manual knowledge base, perform knowledge matching and reasoning to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidelines, configuration instructions, or work order creation suggestions.

[0048] It should be understood that by matching and reasoning the extracted key information with the vehicle's official manual knowledge base, a verification and association mechanism between key information elements and authoritative knowledge base is established. This ensures that the generation of handling strategies is constrained by the content of the official manual to guarantee that the response meets official standards and that the analysis results are structured to facilitate automated execution. This achieves the transformation from key information to operation guidelines, configuration instructions, or work order creation suggestions.

[0049] Furthermore, step S13 specifically includes the following steps: The extracted key information is converted into a vector representation, and target knowledge fragments with a similarity higher than a preset similarity threshold are retrieved from the vehicle's official manual knowledge base. The key information and the target knowledge fragment are assembled into prompt words, and the prompt words are input into the customized generative artificial intelligence model; The customized generative artificial intelligence model uses the prompt words to perform reasoning and generate structured analysis results.

[0050] It should be noted that by converting key information into vector representations and retrieving similar target knowledge fragments from the vehicle's official manual knowledge base, a semantic association between the query information and the authoritative knowledge base is established. This allows the assembled prompt words to incorporate target knowledge fragments from the external knowledge base as contextual basis. Subsequently, the model performs reasoning based on the prompt words containing contextual basis, generating structured analysis results and realizing model reasoning generation based on knowledge base retrieval.

[0051] This embodiment, through the above-described scheme, receives customer complaint instructions in natural language form issued by users through in-vehicle terminals; utilizes a customized generative artificial intelligence model to perform semantic understanding and key information extraction on the customer complaint instructions; performs knowledge matching and reasoning based on the extracted key information and the vehicle's official manual knowledge base to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidance, configuration instructions, or work order creation suggestions; it can reduce the threshold for users to file complaints through natural language interaction, ensure the accuracy and authority of handling strategies by using a generative model integrated with the official knowledge base, avoid the subjective bias of manual analysis, and output structured operation guidance or work order suggestions, providing standardized input for the automated execution of subsequent business processes, significantly improving the intelligence level and response efficiency of in-vehicle customer complaint handling.

[0052] Furthermore, Figure 4 This is a flowchart illustrating the third embodiment of the in-vehicle voice customer complaint processing method of the present invention, as shown below. Figure 4 As shown, based on the first embodiment, a third embodiment of the in-vehicle voice customer complaint processing method of the present invention is proposed. In this embodiment, step S20 specifically includes the following steps: Step S21: Parse the strategy type of the handling strategy in the analysis results, and call the business interface that matches the strategy type to execute the corresponding business operation.

[0053] It should be noted that by parsing the handling strategies in the analysis results and establishing a call mapping relationship between strategy types and business interfaces, the system can automatically select the matching business interface for call according to the strategy type, thereby realizing the automated conversion and execution of handling strategies into specific business operations.

[0054] Step S22: Monitor the node changes during the execution of the business operation and generate a flow status that represents the progress of the business.

[0055] Understandably, by monitoring the changes in nodes during the execution of business operations, a correlation was established between business execution actions and progress status data, enabling the generated flow status to represent business progress and realizing the generation of status during business execution.

[0056] Furthermore, step S22 specifically includes the following steps: Real-time capture of state change events triggered during the execution of the business operations; Extract the current business stage identifier and change time information from the status change event; Based on the current business stage identifier and the change time information, a flow status representing the business progress is encapsulated and generated.

[0057] It should be noted that by capturing state change events triggered during the execution of business operations in real time, a real-time acquisition mechanism between the business execution process and event data has been established. This enables the extraction of the current business stage identifier and change time information as the data basis for state generation. Based on this data basis, the flow status representing the business progress is encapsulated and generated, realizing the event-driven generation and real-time reflection of the business progress status.

[0058] Step S23: Push the circulation status to the after-sales management system in real time via the data interface for updating.

[0059] It should be understood that by using a data interface to push the circulation status to the after-sales management system in real time, a transmission link can be established between the circulation status data and the after-sales management system, enabling the after-sales management system to perform update operations based on the received circulation status.

[0060] This embodiment, through the above-described scheme, parses the strategy type of the handling strategy in the analysis results, calls the business interface matching the strategy type to execute the corresponding business operation; monitors the node changes during the execution of the business operation, generates a flow status representing the business progress; and pushes the flow status to the after-sales management system in real time for updates through the data interface. This enables the automated conversion and execution of handling strategies into business operations, avoiding efficiency losses and errors caused by manual operation. At the same time, it establishes a real-time data link between the business execution process and the after-sales management system, ensuring the consistency and timeliness of business progress data across multiple systems. It achieves visualized tracking of the entire business processing flow, laying a data foundation for subsequently pushing accurate closed-loop feedback to users.

[0061] Furthermore, Figure 5 This is a flowchart illustrating the fourth embodiment of the in-vehicle voice customer complaint processing method of the present invention, as shown below. Figure 5 As shown, based on the first embodiment, a fourth embodiment of the in-vehicle voice customer complaint processing method of the present invention is proposed. In this embodiment, step S30 specifically includes the following steps: Step S31: Periodically poll or monitor the changes in the circulation status in the after-sales management system through a preset data interface.

[0062] It should be noted that by using a pre-defined data interface to build a data transmission channel with the after-sales management system, and by using a periodic polling or real-time monitoring mechanism to actively obtain updated information on the flow status, the system can promptly perceive changes in the business progress of the after-sales management system.

[0063] Step S32: When the flow state is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered.

[0064] Understandably, by judging whether the flow status has reached the preset closed-loop conditions, a triggering association between the status change event and the notification generation mechanism is established, so that the notification generation mechanism is activated only when it is confirmed that the business processing has reached the closed-loop requirements.

[0065] Step S33: Construct a feedback notification containing specific processing results according to the notification generation mechanism, and push the feedback notification to the user through the vehicle terminal.

[0066] It should be understood that by constructing feedback notifications based on specific processing results according to the notification generation mechanism, a connection is established between processing result data and notification content, enabling feedback notifications to carry specific processing results and push feedback notifications to users through the vehicle terminal as a transmission channel.

[0067] Furthermore, step S33 specifically includes the following steps: The notification generation mechanism is used to parse the status type of the flow state and match the notification content template corresponding to the status type. Obtain the specific processing result associated with the flow status, fill the specific processing result into the notification content template, and generate a feedback notification; The message push channel of the vehicle terminal is determined, and the feedback notification is sent to the vehicle terminal through the message push channel.

[0068] It should be noted that by parsing the status type of the flow state and matching the corresponding notification content template, a correlation mechanism between the business status and the notification presentation format is established. This enables the specific processing results to be standardized and filled according to the template to generate feedback notifications. By determining the message push channel, the feedback notifications are sent to the vehicle terminal, thus achieving the formatted generation of notification content and the accurate matching of the transmission channel.

[0069] In practice, this embodiment can be implemented based on a dedicated customer complaint handling model, which can be developed by the vehicle manufacturer based on an open-source base model or independently trained and built.

[0070] The model has a built-in knowledge base of official vehicle manuals for all models, focusing on in-vehicle scenarios and specifically designed to receive, analyze, and handle various user complaints about vehicle use.

[0071] When a user wakes up "I want to complain" via voice or triggers the service by clicking the dedicated customer complaint entry in the vehicle, the customer complaint big data model automatically starts and carries out intelligent processing, which is divided into three closed-loop scenarios: 1. Complaints related to functionality For user feedback regarding functional abnormalities and operational questions, such as the inability to use seat ventilation or massage functions, etc.

[0072] The model will combine the knowledge from the built-in vehicle manual to push illustrated operation guides to users; if the inability to use a function is due to differences in vehicle configuration, it will directly and accurately explain to the user.

[0073] If the problem persists after the user follows the instructions, the model will automatically create an after-sales work order and connect to the Just In Time Issue Resolution and Tracking (JIRA) system and other automotive issue management systems such as ZenTao to complete the synchronous entry of the work order.

[0074] Once the problem has been rectified and repaired, the vehicle's infotainment system will proactively push a notification of the results, thus achieving a closed-loop process for the entire problem.

[0075] 2. Complaints related to discrepancies between expected and unexpected experiences For user-described, vague, and fragmented complaints, such as feedback on poor navigation usage, this approach addresses user experience issues.

[0076] Such problems cannot be directly transformed into effective optimization requirements. The large model will use multiple rounds of intelligent dialogue to deeply guide communication and gradually uncover the user's real pain points and specific problem details.

[0077] Once the user clearly defines their specific needs and the problem scenario, a standardized work order is automatically generated and synchronized to the automaker's problem management system. After the problem is optimized, the processing results are pushed to the vehicle's system, forming a complete closed loop.

[0078] 3. Complaints that cannot be resolved online for an extended period of time. If user-reported issues cannot be resolved through short-term methods such as temporary debugging or over-the-air (OTA) updates, the product or vehicle project team can fill in a user-specific response in the issue management system, clearly indicating the issue optimization cycle, the final version, or the specific reasons why rectification is not possible due to objective limitations such as regulations or safety.

[0079] The final response will be proactively pushed to the user by the vehicle's infotainment system, ensuring the user's right to know, improving the user's perception of vehicle services, and ensuring that every user request receives a response and feedback.

[0080] See Figure 6 , Figure 6 This is an example diagram of the customer complaint feedback dialogue in the in-vehicle voice customer complaint processing method of the present invention, as shown below. Figure 6 As shown, this dialogue example demonstrates how a user submits a natural language complaint about a malfunctioning navigation module. The system utilizes a generative model integrated with a vehicle knowledge base to perform semantic understanding and extract key information from the command. It identifies specific issues such as untimely road condition updates and generates a handling strategy that includes entering the complaint into a work order and coordinating optimization with the R&D team. This drives business execution to enter the problem into the work order system and pushes a feedback notification containing problem confirmation and optimization commitments to the user via the in-vehicle terminal. This establishes an automated mapping link between the user's natural language request, the backend work order processing, and the frontend feedback, enabling accurate responses and closed-loop notifications in the complaint handling process without human intervention.

[0081] Understandably, this embodiment can automate the entire customer complaint handling process, significantly reducing labor costs. Based on a customized in-vehicle customer complaint model, it enables automatic receipt, intelligent identification, accurate classification, and automatic creation of work orders for user complaints, replacing tedious traditional manual input, screening, and archiving operations, significantly improving customer complaint handling efficiency and reducing after-sales manpower investment. It accurately understands user demands and resolves pain points caused by ambiguous expressions. The model has a built-in knowledge base of the vehicle's official manual, combined with multi-turn dialogue capabilities, and can intelligently respond to different types of complaints such as operational questions, functional abnormalities, and dissatisfaction with the experience. It proactively guides the exploration of ambiguous complaints, accurately extracting real pain points, avoiding information omissions or deviations, and improving the effectiveness of complaint collection. It connects with the internal systems of car manufacturers to achieve rapid problem closure. It deeply integrates with car manufacturer problem management systems such as JIRA and ZenTao, allowing complaint work orders to be synchronized with one click and automatically transferred without manual intermediaries. After the problem is resolved, the result notification is proactively pushed through the vehicle's system, forming a "complaint-processing-feedback-notification" process. A complete closed loop shortens the problem-solving cycle; significantly improves user experience and enhances service perception; functional issues can be addressed with real-time illustrated tutorials for quick self-service solutions; for issues that cannot be resolved immediately, the solution timeline, OTA version, or reason for non-resolvability are clearly stated, ensuring user requests are responded to, progressed, and resolved, greatly improving user satisfaction and brand trust; specifically adapted for in-vehicle scenarios, combining practicality and versatility; the model is specifically trained for in-vehicle customer complaint scenarios, supporting multiple access methods such as voice wake-up and in-vehicle system entry triggering, adapting to driving environments; it can be further developed based on open-source models or independently trained, adapting to multiple vehicle models and configurations, facilitating large-scale deployment and iterative upgrades for automakers; it supports continuous product optimization, generating data feedback; the system automatically accumulates complaint data and issue types, forming a customer complaint big data dashboard, helping automakers accurately locate common product issues and high-frequency defects, providing data support for in-vehicle system OTA upgrades, function optimization, and product design improvements, enabling continuous product and service iteration.

[0082] This embodiment, through the above-described scheme, periodically polls or monitors changes in the workflow status of the after-sales management system via a preset data interface. When a change in the workflow status to a preset closed-loop condition is detected, a notification generation mechanism is triggered. A feedback notification containing specific processing results is constructed based on this mechanism and pushed to the user via the vehicle terminal. This enables proactive tracking and real-time awareness of the workflow status of the after-sales management system, ensuring that changes in backend processing progress are not missed. Simultaneously, triggering notification generation based on preset closed-loop conditions ensures consistency between feedback timing and business completion status, avoiding interference from invalid notifications. Furthermore, the notification containing specific processing results is accurately pushed via the vehicle terminal, allowing users to obtain the final processing conclusion without actively querying, thus achieving transparent tracking and closed-loop management of the customer complaint service process.

[0083] Accordingly, the present invention further provides an in-vehicle voice customer complaint processing device.

[0084] Reference Figure 7 , Figure 7 This is a functional block diagram of the first embodiment of the vehicle-mounted voice customer complaint processing device of the present invention.

[0085] In a first embodiment of the in-vehicle voice customer complaint processing device of the present invention, the in-vehicle voice customer complaint processing device includes: The instruction parsing module 10 is used to receive customer complaint instructions, parse the customer complaint instructions using a generative model with an integrated vehicle knowledge base, and generate analysis results containing handling strategies.

[0086] The business execution module 20 is used to drive business execution based on the handling strategy in the analysis results, and to synchronize the flow status generated during the execution process to the after-sales management system in real time.

[0087] The monitoring feedback module 30 is used to monitor the circulation status in the after-sales management system. When a preset closed-loop condition for a change in the circulation status is detected, a feedback notification is generated and pushed to the user through the vehicle terminal.

[0088] The instruction parsing module 10 is also used to receive customer complaint instructions in natural language form issued by the user through the vehicle terminal; to perform semantic understanding and key information extraction on the customer complaint instructions using a customized generative artificial intelligence model; to perform knowledge matching and reasoning based on the extracted key information and the vehicle's official manual knowledge base, and to generate structured analysis results; wherein, the handling strategy in the analysis results includes at least one of operation guidance, configuration instructions or work order creation suggestions.

[0089] The instruction parsing module 10 is further configured to convert the extracted key information into a vector representation, and retrieve target knowledge fragments in the vehicle official manual knowledge base whose similarity to the vector representation is higher than a preset similarity threshold; assemble the key information and the target knowledge fragments into prompt words, input the prompt words into the customized generative artificial intelligence model; and generate structured analysis results by reasoning based on the prompt words through the customized generative artificial intelligence model.

[0090] The business execution module 20 is also used to parse the strategy type of the handling strategy in the analysis results, call the business interface matching the strategy type to execute the corresponding business operation; monitor the node changes during the execution of the business operation, generate a flow status representing the business progress; and push the flow status to the after-sales management system in real time for updating through the data interface.

[0091] The business execution module 20 is also used to capture state change events triggered during the execution of the business operation in real time; extract the current business stage identifier and change time information from the state change events; and encapsulate and generate a flow status representing the business progress based on the current business stage identifier and the change time information.

[0092] The monitoring feedback module 30 is also used to periodically poll or monitor changes in the circulation status in the after-sales management system through a preset data interface; when the circulation status is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered; a feedback notification containing specific processing results is constructed according to the notification generation mechanism, and the feedback notification is pushed to the user through the vehicle terminal.

[0093] The monitoring feedback module 30 is further configured to parse the status type of the flow state according to the notification generation mechanism, match the notification content template corresponding to the status type; obtain the specific processing result associated with the flow state, fill the specific processing result into the notification content template, and generate a feedback notification; determine the message push channel of the vehicle terminal, and send the feedback notification to the vehicle terminal through the message push channel.

[0094] The steps for implementing each functional module of the vehicle-mounted voice customer complaint processing device can be referred to in the various embodiments of the vehicle-mounted voice customer complaint processing method of the present invention, and will not be repeated here.

[0095] Furthermore, this embodiment of the invention also proposes a storage medium storing an in-vehicle voice customer complaint processing program. When the in-vehicle voice customer complaint processing program is executed by a processor, it performs the following operations: Upon receiving a customer complaint instruction, the system uses a generative model with an integrated vehicle knowledge base to parse the complaint instruction and generate analysis results containing handling strategies. The business execution is driven by the handling strategies in the analysis results, and the flow status generated during the execution process is synchronized to the after-sales management system in real time. The system monitors the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user via the vehicle terminal.

[0096] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: Receive customer complaint instructions in natural language form issued by users through the vehicle terminal; A customized generative artificial intelligence model is used to perform semantic understanding and key information extraction on the customer complaint instructions; Based on the extracted key information and the vehicle's official manual knowledge base, knowledge matching and reasoning are performed to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidelines, configuration instructions, or work order creation suggestions.

[0097] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: The extracted key information is converted into a vector representation, and target knowledge fragments with a similarity higher than a preset similarity threshold are retrieved from the vehicle's official manual knowledge base. The key information and the target knowledge fragment are assembled into prompt words, and the prompt words are input into the customized generative artificial intelligence model; The customized generative artificial intelligence model performs reasoning based on the prompt words to generate structured analysis results.

[0098] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: Parse the strategy type of the handling strategy in the analysis results, and call the business interface that matches the strategy type to execute the corresponding business operation; Monitor node changes during the execution of the business operation and generate a flow status that represents the progress of the business; The circulation status is pushed to the after-sales management system in real time via the data interface for updates.

[0099] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: Real-time capture of state change events triggered during the execution of the business operations; Extract the current business stage identifier and change time information from the status change event; Based on the current business stage identifier and the change time information, a flow status representing the business progress is encapsulated and generated.

[0100] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: The system periodically polls or monitors changes in the circulation status of the after-sales management system through a preset data interface. When the flow state is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered; A feedback notification containing specific processing results is constructed based on the notification generation mechanism, and the feedback notification is pushed to the user through the vehicle terminal.

[0101] Furthermore, when the in-vehicle voice customer complaint processing program is executed by the processor, it also performs the following operations: The notification generation mechanism is used to parse the status type of the flow state and match the notification content template corresponding to the status type. Obtain the specific processing result associated with the flow status, fill the specific processing result into the notification content template, and generate a feedback notification; The message push channel of the vehicle terminal is determined, and the feedback notification is sent to the vehicle terminal through the message push channel.

[0102] Those skilled in the art will understand that all or part of the steps in the methods described above can be implemented by a program instructing related hardware. The program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium is a computer-readable storage medium, including: USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.

[0103] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0104] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0105] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for handling in-vehicle voice customer complaints, characterized in that, The in-vehicle voice customer complaint handling method includes: Upon receiving a customer complaint instruction, the system uses a generative model with an integrated vehicle knowledge base to parse the complaint instruction and generate analysis results containing handling strategies. The business execution is driven by the handling strategies in the analysis results, and the flow status generated during the execution process is synchronized to the after-sales management system in real time. The system monitors the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user via the vehicle terminal.

2. The in-vehicle voice customer complaint processing method as described in claim 1, characterized in that, The process of receiving a customer complaint instruction involves parsing the instruction using a generative model integrated with a vehicle knowledge base, generating analysis results that include handling strategies, including: Receive customer complaint instructions in natural language form issued by users through the vehicle terminal; A customized generative artificial intelligence model is used to perform semantic understanding and key information extraction on the customer complaint instructions; Based on the extracted key information and the vehicle's official manual knowledge base, knowledge matching and reasoning are performed to generate structured analysis results; wherein, the handling strategies in the analysis results include at least one of operation guidelines, configuration instructions, or work order creation suggestions.

3. The in-vehicle voice customer complaint processing method as described in claim 2, characterized in that, The extracted key information is matched and reasoned with the vehicle's official manual knowledge base to generate structured analysis results, including: The extracted key information is converted into a vector representation, and target knowledge fragments with a similarity higher than a preset similarity threshold are retrieved from the vehicle's official manual knowledge base. The key information and the target knowledge fragment are assembled into prompt words, and the prompt words are input into the customized generative artificial intelligence model; The customized generative artificial intelligence model uses the prompt words to perform reasoning and generate structured analysis results.

4. The in-vehicle voice customer complaint processing method as described in claim 1, characterized in that, The process of driving business execution based on the handling strategies derived from the analysis results, and synchronizing the workflow status generated during execution to the after-sales management system in real time, includes: Parse the strategy type of the handling strategy in the analysis results, and call the business interface that matches the strategy type to execute the corresponding business operation; Monitor node changes during the execution of the business operation and generate a flow status that represents the progress of the business; The circulation status is pushed to the after-sales management system in real time via the data interface for updates.

5. The in-vehicle voice customer complaint processing method as described in claim 4, characterized in that, The monitoring of node changes during the execution of the business operation and the generation of a flow status representing the business progress include: Real-time capture of state change events triggered during the execution of the business operations; Extract the current business stage identifier and change time information from the status change event; Based on the current business stage identifier and the change time information, a flow status representing the business progress is encapsulated and generated.

6. The in-vehicle voice customer complaint processing method as described in claim 1, characterized in that, The system monitors the circulation status in the after-sales management system. When a change in the circulation status is detected under preset closed-loop conditions, a feedback notification is generated and pushed to the user via the vehicle terminal, including: The system periodically polls or monitors changes in the circulation status of the after-sales management system through a preset data interface. When the flow state is detected to change to a preset closed-loop condition, a notification generation mechanism is triggered; A feedback notification containing specific processing results is constructed based on the notification generation mechanism, and the feedback notification is pushed to the user through the vehicle terminal.

7. The in-vehicle voice customer complaint processing method as described in claim 6, characterized in that, The step of constructing a feedback notification containing specific processing results according to the notification generation mechanism and pushing the feedback notification to the user through the vehicle terminal includes: The notification generation mechanism is used to parse the status type of the flow state and match the notification content template corresponding to the status type. Obtain the specific processing result associated with the flow status, fill the specific processing result into the notification content template, and generate a feedback notification; The message push channel of the vehicle terminal is determined, and the feedback notification is sent to the vehicle terminal through the message push channel.

8. A vehicle-mounted voice customer complaint processing device, characterized in that, The in-vehicle voice customer complaint processing device includes: The instruction parsing module is used to receive customer complaint instructions, parse the customer complaint instructions using a generative model with an integrated vehicle knowledge base, and generate analysis results containing handling strategies. The business execution module is used to drive business execution based on the handling strategies in the analysis results, and to synchronize the flow status generated during the execution process to the after-sales management system in real time. The monitoring and feedback module is used to monitor the circulation status in the after-sales management system. When a change in the circulation status is detected under a preset closed-loop condition, a feedback notification is generated and pushed to the user through the vehicle terminal.

9. A vehicle-mounted voice customer complaint processing device, characterized in that, The in-vehicle voice customer complaint processing device includes: a memory, a processor, and an in-vehicle voice customer complaint processing program stored in the memory and executable on the processor, wherein the in-vehicle voice customer complaint processing program is configured to implement the steps of the in-vehicle voice customer complaint processing method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores an in-vehicle voice customer complaint processing program, which, when executed by a processor, implements the steps of the in-vehicle voice customer complaint processing method as described in any one of claims 1 to 7.

Images