In-vehicle manual-based dialogue method, device, electronic device, and readable storage medium

Abstract

The present application relates to the field of smart cockpit technology, and provides a method, device, electronic device, and readable storage medium for dialogue based on an on-board manual. The method includes: receiving a query message sent by a user, and obtaining the current driving status of the vehicle, which includes a driving status and a stopped status; vectorizing the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message; determining the complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title; determining a dialogue mode based on the complexity level of the current query message, the current driving status, and the manual text title, and replying with the manual text content corresponding to the manual text title as the dialogue content according to the dialogue mode. The on-board manual dialogue method provided by the present application can more conveniently understand and solve problems encountered when using a vehicle, and improves the user experience of using an electronic manual.

Description

Technical Field

[0001] The present application relates to the field of smart cockpit technology, and in particular to a method, device, electronic device and readable storage medium based on an in-vehicle manual dialogue. Background Art

[0002] In the prior art, to help car owners properly use and maintain their vehicles, onboard user manuals (i.e., electronic manuals) have become essential tools for users. To better facilitate the use of electronic manuals, images, text, or videos are added to help users understand and resolve issues that arise during vehicle use. As electronic manuals become increasingly rich and complex with each new version, reading navigation and search functions are incorporated into electronic manuals to facilitate user navigation and search.

[0003] However, as the content of electronic manuals is becoming increasingly rich, even if there are reading navigation and search functions in the electronic manuals, if users want to find the content they need when using the electronic manuals, the in-vehicle system still needs to choose a way to communicate with the user based on factors such as the complexity of the content and the user's reading scenario. Otherwise, it will be difficult and cumbersome for users to understand the content of the electronic manual. Summary of the Invention

[0004] In view of this, the embodiments of the present application provide a vehicle manual dialogue method, device, electronic device and readable storage medium to solve the problem in the prior art that it is difficult and cumbersome for users to understand the content of the electronic manual.

[0005] A first aspect of an embodiment of the present application provides a method for dialog based on an in-vehicle manual, comprising:

[0006] Receive query messages sent by users and obtain the current vehicle driving status, including driving status and stopped status;

[0007] Vectorize the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message;

[0008] Determining a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title, where the complexity levels include a first complexity level, a second complexity level, and a third complexity level. The complexity levels are ranked from easy to difficult in the order of first complexity level, second complexity level, and third complexity level.

[0009] Based on the complexity level of the current query message, the current driving status and the manual text title, a dialogue mode is determined, and the manual text content corresponding to the manual text title is replied as the dialogue content according to the dialogue mode.

[0010] A second aspect of an embodiment of the present application provides a vehicle manual-based dialogue device, comprising:

[0011] A receiving module is configured to receive a query message sent by a user and obtain the current driving status of the vehicle, which includes a driving state and a stopped state;

[0012] A processing module is configured to vectorize the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message;

[0013] a determination module configured to determine a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title, the complexity levels including a first complexity level, a second complexity level, and a third complexity level, the complexity levels being in the order of first complexity level, second complexity level, and third complexity level from easy to difficult;

[0014] The dialogue module is configured to determine a dialogue mode based on the complexity level of the current query message, the current driving status and the manual text title, and reply with the manual text content corresponding to the manual text title as the dialogue content according to the dialogue mode.

[0015] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0016] According to a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above method are implemented.

[0017] The beneficial effects of the embodiments of the present application compared with the prior art are: based on the query message sent by the user and the current driving status of the vehicle, the text content corresponding to the query message is vectorized, and the complexity level corresponding to the current vectorized query text is determined based on the vectorized query text and the manual text title. According to the complexity level, driving status and manual text title, the conversation content and conversation method for replying to the query message sent by the user can be determined, thereby providing a more intelligent and interactive electronic manual, which can determine the conversation method according to the driving status and conversation content, allowing users to more conveniently understand and solve problems encountered when using the vehicle, thereby improving the user experience of using the electronic manual. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 This is a schematic diagram of an application scenario of an embodiment of the present application;

[0020] Figure 2 This is a flowchart of a vehicle manual dialogue method provided by an embodiment of the present application;

[0021] Figure 3 This is a flow chart of a method for communicating with a vehicle controller to obtain parameters, provided in an embodiment of the present application;

[0022] Figure 4 This is a schematic diagram of a vehicle manual dialogue device provided by an embodiment of the present application;

[0023] Figure 5 Schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0025] A method and device for dialogue based on an in-vehicle manual according to an embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0026] Figure 1 1 is a schematic diagram of an application scenario of an embodiment of the present application. The application scenario may include a first terminal device 101, a second terminal device 102, a server 103, and a network 104.

[0027] The first terminal device 101 can be hardware or software. When the first terminal device 101 is hardware, it can be various electronic devices with a display screen and supporting communication with the server 103, including but not limited to smartphones, tablet computers, laptop computers, and desktop computers; when the first terminal device 101 is software, it can be installed in the electronic devices described above. The first terminal device 101 can be implemented as multiple software or software modules, or as a single software or software module, and this embodiment of the application does not limit this. Furthermore, various applications can be installed on the first terminal device 101, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, etc.

[0028] The second terminal device 102 can be hardware or software. When the second terminal device 102 is hardware, it can be various electronic devices with a display screen and supporting communication with the server 103, including but not limited to an onboard computer and a vehicle controller; when the second terminal device 102 is software, it can be installed in the electronic device described above. The second terminal device 102 can be implemented as multiple software or software modules, or as a single software or software module, and this embodiment of the application does not limit this. Furthermore, various applications can be installed on the second terminal device 102, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, vehicle control applications, etc.

[0029] Server 103 can be a server that provides various services, for example, a backend server that receives requests sent by terminal devices that establish communication connections with it. The backend server can receive and analyze the requests sent by the terminal devices, and generate processing results. Server 103 can be a single server, a server cluster consisting of multiple servers, or a cloud computing service center, and the embodiments of the present application are not limited thereto.

[0030] It should be noted that the server 103 can be either hardware or software. When the server 103 is hardware, it can be various electronic devices that provide various services to the first terminal device 101 and the second terminal device 102. When the server 103 is software, it can be multiple software or software modules that provide various services to the first terminal device 101 and the second terminal device 102, or it can be a single software or software module that provides various services to the first terminal device 101 and the second terminal device 102, and this embodiment of the application does not limit this.

[0031] The network 104 may be a wired network connected by coaxial cables, twisted pairs, and optical fibers, or a wireless network that interconnects various communication devices without wiring, such as Bluetooth, NFC, infrared, etc., which is not limited in the embodiments of the present application.

[0032] It should be noted that the specific types, quantities and combinations of the first terminal device 101, the second terminal device 102, the server 103 and the network 104 can be adjusted according to the actual needs of the application scenario, and the embodiments of the present application do not limit this.

[0033] Figure 2 This is a flow chart of a method for dialog based on a vehicle manual provided by an embodiment of the present application. Figure 2 As shown, the vehicle manual-based dialogue method includes the following steps:

[0034] S201, receiving a query message sent by a user and obtaining the current driving status of the vehicle;

[0035] S202, vectorizing the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message;

[0036] S203, determining a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title;

[0037] S204, based on the complexity level of the current query message, the current driving status and the manual text title, determine the dialogue mode, and reply with the manual text content corresponding to the manual text title as the dialogue content according to the dialogue mode.

[0038] The driving state includes a driving state and a stopped state, the complexity levels include a first complexity level, a second complexity level and a third complexity level, and the complexity levels are arranged in the order of first complexity level, second complexity level and third complexity level from easy to difficult.

[0039] Figure 2 The car manual dialogue method can be Figure 1 The in-vehicle manual can be executed on any terminal device and can come from the server.

[0040] In some embodiments, a query message sent by a user is received and the driving status of the vehicle is obtained in real time. The query message sent by the user can be a voice query message or a text query message, and the driving status of the vehicle includes a driving state and a stopped state.

[0041] The text content corresponding to the current voice query message or text query message is vectorized, and the vectorized text content is called the vectorized query text. The complexity level corresponding to the current vectorized query text is determined based on the vectorized query text and the manual text title.

[0042] The electronic manual is also referred to as a manual in the embodiments of this application. The manual includes a manual title and manual content, and there is a one-to-one correspondence between the manual title and the manual content. Complexity levels include a first complexity level, a second complexity level, and a third complexity level. The first complexity level is the lowest complexity, the second complexity level is moderate complexity, and the third complexity level is the highest complexity.

[0043] Based on the complexity level corresponding to the current query message, the current vehicle's driving status and the manual text title, the vehicle system determines the dialogue mode for communicating with the user based on the manual text, and responds with the manual text content corresponding to the manual text title as the dialogue content according to the determined dialogue mode.

[0044] According to the technical solution provided in the embodiment of the present application, the conversation content and conversation method for replying to the query message sent by the user can be determined, thereby providing a more intelligent and interactive electronic manual. The conversation method can be determined according to the driving status and conversation content, so that users can more conveniently understand and solve problems encountered when using the vehicle, thereby improving the user experience of using the electronic manual.

[0045] In some embodiments, before determining the complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title, the process includes:

[0046] Each corpus in the preprocessing corpus is a manual text title or a basic vehicle information. The basic information includes all the indicator lights and indicator parameters of the vehicle.

[0047] The pre-processed manual text title and basic vehicle information are used as input signals and fed into the BERT model to obtain the vectors corresponding to each corpus.

[0048] The corpus also includes the manual text content, which also includes the basic information of the vehicle and the corresponding troubleshooting methods;

[0049] The preprocessing includes: marking the text content of each manual including pictures, texts and videos, associating the manual text title and the manual text content, associating the basic information of the vehicle and the troubleshooting method of the basic information of the vehicle.

[0050] Before determining the complexity level corresponding to the current vectorized query text according to the current vectorized query text and the manual text title, each corpus in the corpus is preprocessed.

[0051] Among them, a corpus is a manual text title, which also includes basic information about the vehicle, including all indicator lights and indication parameters of the vehicle. The corpus also includes fault problems, such as "changing tires".

[0052] During the preprocessing phase, the manual text and troubleshooting solutions in the corpus, including graphic or video instructions, are individually tagged. This allows the user to directly access the corresponding manual text page or graphic or video content when determining the conversational content. This tagging method allows for the user to specify the content format corresponding to each field. The corresponding content is automatically identified when the manual text, troubleshooting problem, or indicator light troubleshooting solution is input into the vehicle system.

[0053] At the same time, in the preprocessing stage, the manual text title and the manual text content are associated one by one, and the basic information and the troubleshooting method of the basic information are associated one by one, so as to perform vectorized operations on the manual text title, basic information and fault problems. The corresponding manual text content or troubleshooting method can be queried based on the manual text title, basic information, or fault problem, without the need to perform vectorized operations on the manual text content and troubleshooting method, thus saving computing costs.

[0054] The pre-processed manual text title and basic vehicle information are fed into the pre-trained Bidirectional Encoder Representations from Transformers (BERT) model, which serves as the pre-trained language model. The vectorization operation is as follows: the pre-processed manual text title and basic vehicle information are fed into the text vectorization component of the BERT model, and the BERT-processed vectors are output end-to-end, resulting in vectors corresponding to each corpus.

[0055] After performing vectorization operations on each corpus in the corpus, a full vector library A and a tag vector library B are obtained, where B is a subset of A.

[0056] In an exemplary embodiment of the present application, taking "The tire is flat, how to change the spare tire" as an example, the storage format of the vector library is exemplarily explained, and the text is input into the BERT model to perform a vectorization operation on the text to determine the vector corresponding to the text, which is stored in the vector library as [2.1319, -2.1413, -1.6260, -0.8638, 3.3173, 0.1796, -4.4853, 1.1793, -4.4853, -0.9740, -3.1780, 0.1846, -1.5481].

[0057] Based on the BERT model, the model parameters are fine-tuned and the model is trained with training data to adapt the BERT model to the scenario of in-vehicle conversational interaction through voice or text query messages. During the fine-tuning and training process, the structure and parameters of the BERT model need to be selected and adjusted according to the requirements of the task.

[0058] In some embodiments, a loss function can be used to fine-tune the BERT model to update the parameters of the BERT model and improve the performance of the BERT model, wherein the loss function can be a maximum likelihood estimation function (MaximumLikelihoodEstimation) or an adversarial generative network function (Adversarial Training). The BERT model can also be fine-tuned using a self-supervised learning method to improve the contextual understanding ability of the BERT model, such as using the conversation history context for mask prediction. In this application, other methods can also be used to fine-tune the BERT model, or a combination of multiple fine-tuning methods can be used, while using high-quality training data as training samples, so that the BERT model can be more suitable for scenarios where conversations are interacted in the car through voice query messages or text query messages.

[0059] According to the technical solution provided in the embodiment of the present application, the manual text title and the manual text content can be associated in the preprocessing stage, and the basic information of the vehicle and the troubleshooting method of the basic information of the vehicle can be associated. The preprocessed manual text title and the basic information of the vehicle are input into the BERT model to obtain the vectors corresponding to each corpus, so that the similarity between the vectors corresponding to each corpus and the vectorized query text can be calculated, the conversation content can be determined, and the calculation efficiency and accuracy can be improved.

[0060] In some embodiments, determining the complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title includes:

[0061] Determine the similarity between the current vectorized query text and all manual text titles, and determine the manual text title with the highest similarity as the manual text title corresponding to the current vectorized query text;

[0062] The complexity level of the current vectorized query text is determined based on the corresponding manual text title and the corresponding preset complexity level.

[0063] Calculate the similarity between the vectorized query text corresponding to the current query message and all manual text titles. This similarity can be calculated by calculating the cosine similarity between the above vectors, i.e., using the formula cosine_sim = (C·B) / (||C||*||B||), where B represents the labeled complex corpus vector, C represents the vectorized query text, C·B represents the dot product between vector C and vector B, and ||B|| and ||C|| represent the norms of vectors B and C, i.e., the lengths of the vectors. Other similarity calculation methods can also be used to calculate the similarity between the vectorized query text and the manual text titles.

[0064] The manual text title with the highest similarity is determined as the manual text title corresponding to the current vectorized query text. Based on the manual text title and the corresponding preset complexity level, the complexity level of the current vectorized query text is determined, and then the method of communication between the in-vehicle system and the user is determined.

[0065] According to the technical solution provided in the embodiment of the present application, the complexity level of the current vectorized query text can be determined based on the preset complexity level of the manual text title, thereby determining the way of communication between the vehicle-mounted system and the user, making it easier for the user to understand the manual text intuitively, clearly and conveniently.

[0066] In some embodiments, the dialogue mode is determined based on the complexity level of the current query message, the current driving status, and the manual text title, including:

[0067] If the current query message is at the first complexity level and the current driving state is a moving state or a stopped state, the conversation mode is determined to be a voice conversation mode;

[0068] If the current query message is at the second complexity level and the current driving state is a moving state or a stopped state, the conversation mode is determined to be a voice conversation mode or a multimedia conversation mode, and the multimedia conversation mode includes a conversation mode presented through graphics, text, or video;

[0069] If the current query message is at the third complexity level and the current driving state is driving, the dialogue mode is determined to be a voice dialogue for key content and a multimedia dialogue for the dialogue content, and the key content includes key content pre-marked in the manual text content;

[0070] If the current query message is at the third complexity level and the current driving state is a stopped state, the dialogue mode is determined to be a voice dialogue mode and a multimedia dialogue mode.

[0071] If the query is determined to be Level 1 complexity and the vehicle is currently in motion or stopped, the manual text will be displayed via voice dialogue. Level 1 complexity provides relatively simple responses or operations. For example, if a user sends a query asking "How much battery is left?", the onboard system will recognize the query and, based on the manual text title, determine that the question falls under Level 1 complexity. The system will then display the current vehicle energy remaining via voice dialogue, where remaining energy includes one or more of the remaining battery charge and remaining fuel.

[0072] If it is determined that the current query message is at the second complexity level, the current driving state is driving state or stopped state, and the dialogue mode is determined to be voice dialogue mode and multimedia dialogue mode, wherein the multimedia dialogue mode includes a dialogue mode presented through graphics or video.

[0073] In an exemplary embodiment of the present application, an example of an inquiry message such as "My tire is flat, how do I change the spare tire" is used for explanation. When the vehicle-mounted system receives the inquiry message and determines the corresponding manual text title and manual text content in the corpus, and determines that the manual text content corresponding to the inquiry message includes picture or video display content, the electronic manual is opened on the target display screen and jumps to the page number corresponding to the question, the corresponding picture or video is displayed on the target display screen, and the corresponding text content is broadcast through voice dialogue.

[0074] Among them, if there is content corresponding to pictures, text or video at the same time, the pictures, text or video can be set to be displayed on the target display screen at the same time, or the priority of video playback can be set to be higher than the priority of picture and text playback, or interactive controls can be provided to allow users to choose the playback method themselves.

[0075] If the current query level is the third complexity level and the current vehicle driving status is driving status, the dialogue mode is determined to play the key content pre-marked in the manual text content through voice dialogue, and at the same time display all graphics or video content on the target display screen.

[0076] In another exemplary embodiment of the present disclosure, taking the case where the vehicle is in driving state and the query information is "the tire has fallen off" as an example, the vehicle-mounted system detects that the current driving state of the vehicle is driving state, and receives the query information of "the tire has fallen off", and the complexity level of the query information is the third complexity level, then the key content is played in the form of voice dialogue, and the corresponding video content or graphic content is played on the display screen at the same time, so that the user can respond quickly and accurately to the problem of the third complexity level, and minimize the impact of the query message of the third complexity level on the user.

[0077] If the current query message is at the third complexity level and the vehicle is in a stopped state, all conversation contents are communicated with the user in a voice conversation mode and a multimedia conversation mode.

[0078] According to the technical solution provided in the embodiment of the present application, the dialogue mode can be determined according to the complexity level of the query message, the current driving status and the manual text title. When encountering a query message of the third complexity level, a fast, accurate and convenient dialogue mode can be provided to solve the corresponding problems for users and improve the convenience and usage experience of users in using the electronic manual.

[0079] In some embodiments, the vehicle manual-based dialogue method further includes:

[0080] When the conversation content corresponding to the query message is the actual value corresponding to the basic information of the current vehicle;

[0081] Obtain the current actual value based on the Controller Area Network bus, input the current actual value into the BERT model, and obtain the vector result corresponding to the current basic information;

[0082] Determine the manual text content corresponding to the current query message based on the vector result and the manual text title;

[0083] The steps of determining the dialogue mode are performed based on the complexity level of the current query message, the current driving status and the manual text title, and a reply is made according to the dialogue mode.

[0084] In some cases, the query message sent by the user does not contain clear parameters. It is necessary to obtain the current vehicle's real-time parameter information from the vehicle controller system and input it into the BERT model to obtain the corresponding manual file content. For example, the following two questions are: Question 1: What do the icons on the dashboard mean? Question 2: How long until I need maintenance? In this case, real-time data information from the vehicle controller is obtained through the Controller Area Network (CAN) bus, such as the fault monitoring results obtained from the data-driven vehicle history record (VHR) system, or the current dashboard fault icon flash code or current total mileage is obtained as a parameter through the electronic control unit (ECU). The BERT model obtains the corresponding results from the manual text content. When a fault occurs, the solution to the fault is also included in the dialogue content with the user.

[0085] Figure 3 This is a flow chart of a method for communicating with a vehicle controller to obtain parameters provided by an embodiment of the present application. Figure 3As shown in the figure, the data information of the current ECU (including the instrument panel ECU) or VHR is used as the actual value and input into the BERT model through the CAN bus to obtain the vector result corresponding to the current basic information. Based on the vector result and the vector corresponding to each corpus in the corpus, the manual text content corresponding to the current query message is determined. After determining the corresponding complexity level, the method of communicating with the user is determined.

[0086] According to the technical solution provided in the embodiment of the present application, it is possible to obtain real-time relevant data information in combination with the vehicle controller, and after determining the complexity level of the query message, conduct a dialogue with the user in a dialogue method corresponding to the complexity level, thereby improving the flexibility of the dialogue.

[0087] In some embodiments, the query message includes a voice query message and a text query message, and vectorizing the text content corresponding to the query message to obtain the corresponding vectorized query text includes:

[0088] Input the text content into the BERT model to obtain the corresponding vectorized query text;

[0089] Before inputting the text content into the BERT model to obtain the corresponding vectorized query text, the following steps are also included:

[0090] If the query message is a voice query message, the query message is converted into text content based on a natural language processing model.

[0091] After receiving a query message sent by the user, it is determined whether the query message is a voice query message or a text query message. If it is a voice query message, the query message is converted into text content through the natural language processing model. The text content is used as the input signal and input into the fine-tuned and trained BERT model to perform vectorization operations on the query message text to obtain the vector C mentioned above.

[0092] According to the technical solution provided in the embodiment of the present application, it is possible to ensure that the BERT model that performs vectorization on the text content corresponding to the query message and the BERT model that preprocesses the corpus are the same model, thereby ensuring that the same technology and standards are used when vectorizing various types of text, so that the results obtained in calculating the vectorized query text and manual text title are more accurate, thereby improving the accuracy of the conversation.

[0093] In some embodiments, after replying the manual text content corresponding to the manual text title as the conversation content in a conversational manner, the method further includes:

[0094] Sending a feedback message, which is used to send feedback data information of the current conversation to the target display screen;

[0095] Receive the return information of the feedback message and adjust the conversation content according to the return information.

[0096] The corpus involved in the embodiment of the present application is a dynamic corpus. The manufacturer can update the corpus contained in the corpus, or collect query information sent by users through acquisition to compare with the manual text titles and manual text contents in the existing corpus, update the existing corpus, and provide users with a better user experience.

[0097] After a conversation with the user, a feedback message is sent to the user, which is displayed on the target display screen. The system also receives return information corresponding to the feedback message provided by the user based on the conversation, and adjusts the conversation content replied by the in-vehicle system or adjusts the BERT model according to the return information.

[0098] At the same time, the user's emotions can also be judged, and the user's positive and negative emotions can be used as a feedback mechanism for the conversation to improve the content of the conversation between the in-vehicle system and the user.

[0099] According to the technical solution provided in the embodiment of the present application, users can initiate and obtain real-time replies to the in-vehicle system anytime and anywhere, which improves user satisfaction and usage experience. At the same time, the system has a high level of semantic understanding and generation capabilities, and adjusts the conversation content or model based on user feedback messages, thereby providing more accurate and natural replies.

[0100] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0101] All of the above optional technical solutions can be combined in any way to form optional embodiments of the present application, and will not be described in detail here.

[0102] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0103] Figure 4 This is a schematic diagram of a vehicle manual dialogue device provided by an embodiment of the present application. Figure 4 As shown, the vehicle manual-based dialogue device includes: a receiving module 401 , a processing module 402 , a determining module 403 , and a dialogue module 404 .

[0104] The receiving module 401 is configured to receive a query message sent by a user and obtain the current driving status of the vehicle, which includes the driving state and the stopped state;

[0105] The processing module 402 is configured to vectorize the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message;

[0106] Determining module 403 is configured to determine a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title, where the complexity levels include a first complexity level, a second complexity level, and a third complexity level, and the complexity levels are ranked from easy to difficult in the order of first complexity level, second complexity level, and third complexity level;

[0107] The dialogue module 404 is configured to determine a dialogue mode based on the complexity level of the current query message, the current driving status and the manual text title, and reply with the manual text content corresponding to the manual text title as the dialogue content according to the dialogue mode.

[0108] In some embodiments, the determination module 403 is configured to, before determining the complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title,:

[0109] Each corpus in the preprocessing corpus is a manual text title, which also includes the basic information of the vehicle, including all the indicator lights and indicator parameters of the vehicle;

[0110] The pre-processed manual text titles are used as input signals and fed into the BERT model to obtain the vectors corresponding to each corpus;

[0111] The corpus also includes the manual text content, which also includes the basic information of the vehicle and the corresponding troubleshooting methods;

[0112] The preprocessing includes: marking the text content of each manual including pictures, texts and videos, associating the manual text title and the manual text content, associating the basic information of the vehicle and the fault solution corresponding to the basic information of the vehicle.

[0113] In some embodiments, the determination module 403 is configured to determine the complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title, for:

[0114] Determine the similarity between the current vectorized query text and all manual text titles, and determine the manual text title with the highest similarity as the manual text title corresponding to the current vectorized query text;

[0115] The complexity level of the current vectorized query text is determined based on the corresponding manual text title and the corresponding preset complexity level.

[0116] In some embodiments, the dialogue module 404 is configured to determine a dialogue mode based on the complexity level of the current query message, the current driving status, and the manual text title, for:

[0117] If the current query message is at the first complexity level and the current driving state is a moving state or a stopped state, the conversation mode is determined to be a voice conversation mode;

[0118] If the current query message is at the second complexity level and the current driving state is a moving state or a stopped state, the conversation mode is determined to be a voice conversation mode or a multimedia conversation mode, and the multimedia conversation mode includes a conversation mode presented through graphics, text, or video;

[0119] If the current query message is at the third complexity level and the current driving state is driving, the dialogue mode is determined to be a voice dialogue for key content and a multimedia dialogue for the dialogue content, and the key content includes key content pre-marked in the manual text content;

[0120] If the current query message is at the third complexity level and the current driving state is a stopped state, the dialogue mode is determined to be a voice dialogue mode and a multimedia dialogue mode.

[0121] In some embodiments, the vehicle manual-based dialogue device is further configured to:

[0122] When the conversation content corresponding to the query message is the actual value corresponding to the basic information of the current vehicle;

[0123] Obtain the current actual value based on the Controller Area Network bus, input the current actual value into the BERT model, and obtain the vector result corresponding to the current basic information;

[0124] Determine the manual text content corresponding to the current query message based on the vector result and the manual text title;

[0125] The steps of determining the dialogue mode are performed based on the complexity level of the current query message, the current driving status and the manual text title, and a reply is made according to the dialogue mode.

[0126] In some embodiments, the processing module 402 is configured to vectorize text content corresponding to the query message, including a voice query message and a text query message, to obtain a corresponding vectorized query text, for:

[0127] Input the text content into the BERT model to obtain the corresponding vectorized query text;

[0128] Before inputting the text content into the BERT model to obtain the corresponding vectorized query text, the following steps are also included:

[0129] If the query message is a voice query message, the query message is converted into text content based on a natural language processing model.

[0130] In some embodiments, the dialogue module 404 is configured to, after replying in a dialogue manner using the manual text content corresponding to the manual text title as dialogue content, further be used to:

[0131] Sending a feedback message, which is used to send feedback data information of the current conversation to the target display screen;

[0132] Receive the return information of the feedback message and adjust the conversation content according to the return information.

[0133] Figure 5 Schematic diagram of the electronic device 5 provided in the embodiment of the present application. Figure 5 As shown, the electronic device 5 of this embodiment includes: a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and executable by the processor 501. When the processor 501 executes the computer program 503, the steps of the above-mentioned method embodiments are implemented. Alternatively, when the processor 501 executes the computer program 503, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0134] The electronic device 5 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 5 may include but is not limited to a processor 501 and a memory 502. Those skilled in the art will appreciate that Figure 5 This is merely an example of the electronic device 5 and does not limit the electronic device 5 . The electronic device 5 may include more or fewer components than shown in the figure, or different components.

[0135] The processor 501 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0136] The memory 502 can be an internal storage unit of the electronic device 5, such as a hard disk or memory of the electronic device 5. The memory 502 can also be an external storage device of the electronic device 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. The memory 502 can also include both an internal storage unit of the electronic device 5 and an external storage device. The memory 502 is used to store computer programs and other programs and data required by the electronic device.

[0137] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by 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 above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0138] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium (such as a computer-readable storage medium). Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. Computer-readable storage media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0139] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A vehicle manual-based dialogue method, characterized in that: include: Receive a query message sent by a user and obtain the current driving status of the vehicle, which includes the driving state and the stopped state; Vectorizing the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message; determining a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title, the complexity levels including a first complexity level, a second complexity level, and a third complexity level, the complexity levels being in an ascending order of complexity as follows: the first complexity level, the second complexity level, and the third complexity level; Based on the complexity level of the current query message, the current driving status and the manual text title, a dialogue mode is determined, and the manual text content corresponding to the manual text title is replied as the dialogue content according to the dialogue mode.

2. The method according to claim 1, characterized in that Before determining the complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title, the method includes: Preprocessing each corpus in the corpus, wherein one corpus is a title of the manual text, the manual text title also includes basic information of the vehicle, and the basic information includes all indicator lights and indication parameters of the vehicle; Input the pre-processed manual text title as an input signal to the BERT model to obtain a vector corresponding to each corpus; The corpus also includes the manual text content, and the manual text content also includes the fault solution corresponding to the basic information of the vehicle; The preprocessing includes: marking the manual text content including pictures, text and videos, associating the manual text title and the manual text content, and associating the basic information of the vehicle and the fault resolution method corresponding to the basic information of the vehicle.

3. The method according to claim 2, characterized in that Determining a complexity level corresponding to the current vectorized query text based on the current vectorized query text and the manual text title includes: Determining the similarity between the current vectorized query text and all the manual text titles, and determining the manual text title with the highest similarity as the manual text title corresponding to the current vectorized query text; Based on the corresponding manual text title and the corresponding preset complexity level, the complexity level of the current vectorized query text is determined.

4. The method according to claim 1, wherein Determining a dialogue mode based on the complexity level of the current query message, the current driving status, and the manual text title includes: If the current query message is of the first complexity level and the current driving state is the driving state or the stopped state, determining that the dialogue mode is a voice dialogue mode; If the current query message is at the second complexity level and the current driving state is the driving state or the stopped state, determining that the conversation mode is the voice conversation mode and the multimedia conversation mode, wherein the multimedia conversation mode includes a conversation mode presented through graphics, text, or video; If the current query message is at the third complexity level and the current driving state is the driving state, determining the dialogue mode to be a dialogue on key content using the voice dialogue mode and a dialogue on the dialogue content using the multimedia dialogue mode, wherein the key content includes key content pre-marked in the manual text content; If the current query message is at the third complexity level and the current driving state is the stopped state, the dialogue mode is determined to be the voice dialogue mode and the multimedia dialogue mode.

5. The method according to claim 1, wherein Also includes: When the conversation content corresponding to the query message is the actual value corresponding to the basic information of the current vehicle; Acquire the current actual value based on the controller area network bus, input the current actual value into the BERT model, and obtain a vector result corresponding to the current basic information; Determining the manual text content corresponding to the current query message according to the vector result and the manual text title; Execute the step of determining the dialogue mode based on the complexity level of the current query message, the current driving status and the manual text title, and reply according to the dialogue mode.

6. The method according to claim 1, characterized in that The query message includes a voice query message and a text query message, and vectorizing the text content corresponding to the query message to obtain a corresponding vectorized query text includes: Input the text content into the BERT model to obtain the corresponding vectorized query text; Before inputting the text content into the BERT model to obtain the corresponding vectorized query text, the following steps are also included: If the query message is the voice query message, the query message is converted into the text content based on a natural language processing model.

7. The method according to any one of claims 1 to 6, characterized in that After replying the manual text content corresponding to the manual text title as the dialogue content in the dialogue mode, the method further includes: Sending a feedback message, wherein the feedback message is used to send data information of the current conversation to the target display screen; Receive return information of the feedback message, and adjust the conversation content according to the return information.

8. A vehicle manual dialogue device, characterized in that: include: A receiving module is configured to receive a query message sent by a user and obtain the current driving status of the vehicle, wherein the driving status includes a driving state and a stopped state; A processing module is configured to vectorize the text content corresponding to the current query message to obtain a vectorized query text corresponding to the text content corresponding to the current query message; a determination module configured to determine a complexity level corresponding to the current vectorized query text based on the vectorized query text and the manual text title, the complexity levels including a first complexity level, a second complexity level, and a third complexity level, the complexity levels being in an ascending order of complexity: the first complexity level, the second complexity level, and the third complexity level; The dialogue module is configured to determine the dialogue mode based on the complexity level of the current query message, the current driving status and the manual text title, and reply with the manual text content corresponding to the manual text title as the dialogue content according to the dialogue mode.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images