Voice interaction method, server and computer readable storage medium

Abstract

The application discloses a voice interaction method, comprising: receiving a voice request forwarded by a vehicle; processing the voice request, determining a target vehicle-mounted system applet responding to the voice request and a target function page in the target vehicle-mounted system applet for realizing the voice request, and displaying the target function page through the vehicle to complete voice interaction. In the application, a user interacts with the vehicle through voice, and a server can determine a target function page in a target vehicle-mounted system applet that can realize a function demand. The vehicle can directly display the target function page to complete the voice interaction process. The voice interaction method can identify the function demand of the voice request of the user, start running the corresponding vehicle-mounted system applet and directly reach the page realizing the target function, so that the function demand can be realized without too much manual operation of the user, the convenience of voice interaction and the use efficiency of the vehicle-mounted system applet are improved, and the user experience is improved.

Description

Technical Field

[0001] This application relates to the field of vehicle voice technology, and in particular to a voice interaction method, a server, and a computer-readable storage medium. Background Technology

[0002] Currently, in-vehicle voice technology allows users to interact within the vehicle cabin via voice, such as controlling vehicle components or interacting with components in the in-vehicle system's user interface. However, in scenarios where users interact with in-vehicle system mini-programs via voice, only voice control of opening or closing the mini-program is supported. If users need to achieve specific functions within the mini-program, they often need to perform multiple manual operations, affecting the smoothness and convenience of voice interaction and resulting in a poor user experience. Summary of the Invention

[0003] This application provides a voice interaction method, a server, and a computer-readable storage medium.

[0004] The voice interaction method of this application includes: Receive voice requests forwarded by vehicles; The system processes the voice request, determines the target in-vehicle system applet that responds to the voice request, and the target function page in the target in-vehicle system applet that is used to implement the voice request, so as to complete the voice interaction by displaying the target function page through the vehicle.

[0005] Thus, in this application, when a user needs to implement a specific function within the mini-program, they can interact with the vehicle via voice. After processing the received voice request, the server can determine the target in-vehicle system mini-program that can fulfill the functional requirement, as well as the target function page that enables the function. The vehicle can directly display the target function page to complete the voice interaction process. The voice interaction method of this application can recognize the user's voice request for a specific function, run the corresponding in-vehicle system mini-program, and directly access the page that achieves the target function. This eliminates the need for excessive manual operation by the user, improving the convenience of voice interaction and the efficiency of using the in-vehicle system mini-program, thereby enhancing the user experience.

[0006] The method further includes: Based on the intent information obtained from processing the voice request, it is confirmed that the voice request will be implemented through the in-vehicle system applet.

[0007] In this way, based on the intent information of the voice request, it can be determined whether the voice request needs to be implemented by the in-vehicle system mini-program. Users can then control the in-vehicle system mini-program through voice interaction, improving the convenience of in-vehicle voice interaction and the usage rate of the in-vehicle system mini-program.

[0008] The step of determining, based on processing the voice request, a target in-vehicle system mini-program that responds to the voice request and a target function page within the target in-vehicle system mini-program for implementing the voice request, and then displaying the target function page in the vehicle to complete the voice interaction, includes: Based on the slot information obtained from processing the voice request and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the target in-vehicle system mini-program and the target function page are determined.

[0009] In this way, the target in-vehicle system mini-program and the target function page can be determined by combining the custom vocabulary of the aforementioned in-vehicle system mini-program and the matching relationship of the mini-program process nodes with the extracted voice request slot information.

[0010] The step of determining the target in-vehicle system mini-program and the target function page based on the slot information obtained from processing the voice request, the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, and the matching relationship of mini-program flow nodes includes: The target in-vehicle mini-program is determined based on the slot information; Based on the matching relationship of the mini-program process nodes of the target in-vehicle mini-program, the slot information is matched to obtain multiple corresponding process nodes, wherein each corresponding process node corresponds to the location identifier of a function page.

[0011] Thus, the target in-vehicle mini-program is determined based on the slot information of the user's voice request, and the slot information is matched with the process nodes according to the matching relationship of the mini-program's process nodes. The slot information corresponds to multiple process nodes to determine the target process node.

[0012] The slot information includes user-defined elements. The process of determining the target in-vehicle system mini-program and the target function page based on the slot information obtained from processing the voice request, a pre-built custom vocabulary of predetermined categories of in-vehicle system mini-programs, and the matching relationship of mini-program flow nodes includes: If the user-defined element exists in the custom vocabulary, multiple corresponding process nodes are concatenated and combined to determine the target process node.

[0013] In this way, when the user-defined elements contained in the voice request slot information exist in the custom dictionary, the corresponding process nodes of each slot information can be linked together and identified as the target process node, which makes it easier to identify the target function page and improves the user experience.

[0014] The step of determining the target in-vehicle system mini-program and the target function page based on the slot information obtained from processing the voice request, the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, and the matching relationship of mini-program flow nodes includes: Determine the location identifier of the target function page corresponding to the target process node; Assign corresponding slot information to each corresponding process node to obtain the information of the target process node; The target function page is determined based on the information of the target process node and the location identifier.

[0015] In this way, the information of the target process node can be obtained based on the matching results, and the target function page can be determined so that the target function page can be displayed through the vehicle to achieve the target function and complete the voice interaction.

[0016] The step of determining the target in-vehicle system mini-program and the target function page based on the slot information obtained from processing the voice request, the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, and the matching relationship of mini-program flow nodes includes: If the user-defined element does not exist in the custom vocabulary, the preceding process node of the process node corresponding to the user-defined element will be determined as the target process node.

[0017] In this way, when the user-defined element contained in the voice request slot information does not exist in the custom dictionary, the previous process node can be returned and identified as the target process node, instead of directly ending the voice interaction process. This makes it easier for users to make other choices in the custom dictionary and improves the user experience.

[0018] The predefined categories of in-vehicle system mini-programs include those for catering, travel, music, and / or payment.

[0019] In this way, users can directly access the target function pages of various types of in-vehicle system mini-programs, making the user experience more diversified.

[0020] The step of determining, based on processing the voice request, a target in-vehicle system mini-program that responds to the voice request and a target function page within the target in-vehicle system mini-program for implementing the voice request, and then displaying the target function page in the vehicle to complete the voice interaction, includes: Process the voice request to obtain intent information and slot information; The slot information is normalized. Based on the intent information and the normalized slot information, the target vehicle system applet and the target function page are determined.

[0021] In this way, in scenarios where users interact with non-shopping in-vehicle system mini-programs via voice, the intent information obtained from voice requests can be processed, combined with normalized slot information, to determine the target in-vehicle system mini-program and directly open the target function page, making the voice interaction process more time-saving and convenient.

[0022] The server of this application includes a processor and a memory, wherein the memory stores a computer program, which, when executed by the processor, implements the above-described method.

[0023] The computer-readable storage medium of this application stores a computer program that, when executed by one or more processors, implements the above-described method.

[0024] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0025] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, wherein: Figure 1 This is one of the flowcharts illustrating the voice interaction method of this application; Figure 2 This is the second flowchart of the voice interaction method in this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of this application, and should not be construed as limiting the embodiments of this application.

[0027] Please see Figure 1 and Figure 2 This application provides a voice interaction method, including: 01: Receive voice requests forwarded by vehicles; 02: Process voice requests, determine the target in-vehicle system mini-program that responds to the voice request, and the target function page in the target in-vehicle system mini-program that is used to implement the voice request, so as to complete the voice interaction by displaying the target function page through the vehicle.

[0028] This application also provides a server, which includes a memory and a processor. The voice processing method of this application can be implemented by the server of this application. Specifically, the memory stores a computer program, and the processor is used to receive voice requests forwarded by the vehicle, process the voice requests, determine the target in-vehicle system applet that responds to the voice request, and the target function page in the target in-vehicle system applet used to implement the voice request, so as to complete the voice interaction by displaying the target function page through the vehicle.

[0029] With the development and popularization of vehicle electronic technology, vehicle display devices, such as central control displays, have become much richer in terms of interactive functions, and more and more services are being integrated into the vehicle's infotainment system in the form of mini-programs. The introduction of mini-program platforms has met the diverse needs of drivers and passengers during their journeys. However, voice operation only allows users to open mini-programs by voice; subsequent operations within the mini-program still require manual completion, resulting in poor convenience and a subpar user experience. Related technologies, to facilitate user operation, can use voice to directly access mini-program pages or page-style elements. For example, when a user's voice request is "I want to order product 'a' from restaurant mini-program A," only voice control of opening or closing the mini-program is supported; it cannot directly complete product selection, delivery address setting, and direct redirection to the payment confirmation page, resulting in poor convenience and a subpar user experience. Therefore, in situations where user needs involve more complex steps, the corresponding mini-programs cannot be efficiently utilized to meet those needs.

[0030] This application supports direct access to mini-program elements based on voice requests. For example... Figure 2 As shown in one example of this application, a user submits the message "I want to order product a from restaurant mini-program A". The server can then redirect the mini-program directly to the final order submission page, where the user only needs to confirm payment information. Both "restaurant mini-program A" and "product a" can be changed according to the user's needs.

[0031] The in-vehicle system mini-program involved in this application is a mini-program that runs on the in-vehicle system according to the user's personal needs, without requiring pre-installation or adaptation in the vehicle. The target in-vehicle system mini-program is one that can respond to the user's voice requests and fulfill the voice request functionality requirements. The target function page is the page within the target in-vehicle system mini-program that has the capability to fulfill the voice request functionality requirements. After receiving such a voice request, the server can determine the corresponding target in-vehicle system mini-program for the voice request and directly access the target function page used to fulfill the voice request. Ultimately, the target function page can be displayed through the vehicle to complete the voice interaction process.

[0032] Furthermore, in the above embodiments, the actual content of the voice request can be limited by location. For example, if a user issues a voice request saying "Order item a from restaurant mini-program A to my company," and the location information of the "company" set by the user exists in the historical order information, after selecting "restaurant mini-program A" and "item a," the delivery address can be automatically filled in with the address of the "company" set by the user. The mini-program can then directly lead to the final order submission page, where the user only needs to confirm the payment information. The voice interaction method of this application allows users to achieve the specific functions of the in-vehicle system mini-program without performing too many manual operations.

[0033] In summary, in this application, when a user needs to implement a specific function within a mini-program, they interact with the vehicle via voice. After processing the received voice request, the server can determine the target in-vehicle system mini-program that can fulfill the functional requirements, as well as the target function page that enables the function. The vehicle can directly display the target function page to complete the voice interaction process. This application's voice interaction method can recognize the user's voice request for a specific function, start running the corresponding in-vehicle system mini-program, and directly access the page that achieves the target function. This eliminates the need for excessive manual operation by the user, improving the convenience of voice interaction and the efficiency of using the in-vehicle system mini-program, thus enhancing the user experience.

[0034] Voice interaction methods also include: Based on the intent information obtained from processing the voice request, it is confirmed that the voice request will be implemented through the in-vehicle system applet.

[0035] The processor is used to confirm whether the voice request is fulfilled through the in-vehicle system applet based on the intent information obtained from processing the voice request.

[0036] Specifically, please refer to Figure 2 This allows for the analysis of the functions of all mini-programs connected to the in-vehicle system, identifying the business domain in which the user's voice request intent information resides. For example, in the above example, if the intent information of the voice request includes "shopping cart," it is determined that the voice request needs to be implemented through an in-vehicle system mini-program, and the server can then run the corresponding in-vehicle system mini-program in the in-vehicle system user interface.

[0037] In this way, based on the intent information of the voice request, it can be determined whether the voice request needs to be implemented by the in-vehicle system mini-program. Users can then control the in-vehicle system mini-program through voice interaction, improving the convenience of in-vehicle voice interaction and the usage rate of the in-vehicle system mini-program.

[0038] The reservation category of in-vehicle system mini-programs includes catering, travel, music and / or payment in-vehicle system mini-programs.

[0039] Specifically, in-vehicle system mini-programs include categories such as food and beverage, travel, music, and payment. A single in-vehicle system may contain one or more of these mini-programs. The implementation of various functions within each mini-program in the in-vehicle system may involve the display of multiple pages. There is no limit to the number of each type of mini-program or the specific mini-programs used.

[0040] In this way, users can directly access the target function pages of various types of in-vehicle system mini-programs, making the user experience more diversified.

[0041] Step 02 includes: Based on the slot information obtained from processing voice requests and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the target in-vehicle system mini-program and the target function page are determined.

[0042] The processor is used to determine the target in-vehicle system mini-program and the target function page based on the slot information obtained from processing voice requests, the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, and the matching relationship of mini-program flow nodes.

[0043] A custom vocabulary is a list of various elements stored within an in-vehicle system mini-program by the information provider, serving as a reference for voice interaction. As shown in Table 1, the predefined categories of in-vehicle system mini-programs include catering, travel, music, and payment mini-programs. The server can access mini-programs and their corresponding custom vocabulary created by external data platforms.

[0044] Table 1 Specifically, the slot information of voice requests can be correlated with the custom vocabulary of the in-vehicle system mini-program accessed by the server to obtain the matching relationship of mini-program flow nodes. Please refer to Table 2. Taking the catering mini-program A as an example, the matching relationship of mini-program flow nodes during the operation of the mini-program is shown in the table.

[0045] Table 2 Based on the custom vocabulary of the aforementioned in-vehicle system mini-program and the matching relationship of the mini-program's process nodes, combined with the extracted voice request slot information, the target in-vehicle system mini-program and the target function page can be determined.

[0046] In this way, the target in-vehicle system mini-program and the target function page can be determined by combining the custom vocabulary of the aforementioned in-vehicle system mini-program and the matching relationship of the mini-program process nodes with the extracted voice request slot information.

[0047] Based on the slot information obtained from processing voice requests and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the steps to determine the target in-vehicle system mini-program and the target function page include: Determine the target in-vehicle mini-program based on the slot information; Based on the matching relationship of the mini-program process nodes of the target in-vehicle mini-program, the slot information is matched and processed to obtain multiple corresponding process nodes.

[0048] The processor is used to determine the target in-vehicle mini-program based on the slot information, and to perform matching processing on the slot information based on the matching relationship of the mini-program process nodes of the target in-vehicle mini-program to obtain multiple corresponding process nodes.

[0049] Each corresponding process node corresponds to a location identifier on a functional page. Specifically, the server can process the user's voice request to obtain the slot information of the voice request. For example, when the user's voice request is "Order one item a from Restaurant Mini Program A to the company", the extracted slot information is "Restaurant Mini Program A", "Item a", "One Item", "Company". Thus, the target in-vehicle system mini program can be determined as "Restaurant Mini Program A".

[0050] Subsequently, based on the matching relationship of the process nodes corresponding to the target mini-program, the information of each slot can be matched, i.e., name = "Restaurant Mini-Program A", search = "product a", quantity = "1", and address = "company". The matched elements directly lead to the set process nodes, i.e., the slot information "Restaurant Mini-Program A" corresponds to "Node 1"; "product a" corresponds to "Node 3"; and "company" corresponds to "Node 6".

[0051] In this way, the target in-vehicle mini-program can be determined based on the slot information of the user's voice request, and the slot information can be matched with the process nodes according to the matching relationship of the mini-program process nodes. The slot information corresponds to multiple process nodes to determine the target process node.

[0052] Based on the slot information obtained from processing voice requests and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the steps to determine the target in-vehicle system mini-program and the target function page include: When a user-defined element exists in a custom dictionary, multiple corresponding process nodes are chained together to determine the target process node.

[0053] The processor is used to connect and combine multiple corresponding process nodes in a custom dictionary to determine the target process node when the user-defined element exists in the custom dictionary.

[0054] Specifically, please refer to Figure 2User-defined elements include various elements defined by the user according to their preferences or needs. When the slot information of a user's voice request contains a user-defined element, and the user-defined element exists in the custom vocabulary, after matching the slot information containing the user-defined element with the process nodes, multiple corresponding process nodes can be combined in a chain, and finally the target process node can be determined according to a certain process order.

[0055] For example, a user makes a voice request, "Order one serving of product 'a' from restaurant mini-program A to the company." Here, "restaurant mini-program A," "product 'a'," "one serving," and "company" are all custom elements in a custom vocabulary. Each custom element can be matched to the resulting process node. That is, the slot information "restaurant mini-program A" corresponds to "node one"; "product 'a'" and "one serving" correspond to "node three"; and "company" corresponds to "node six." By connecting these nodes according to the mini-program's process sequence, the target process node can be determined as node seven, namely the "confirm payment information" node.

[0056] In this way, when the user-defined elements contained in the voice request slot information exist in the custom dictionary, the corresponding process nodes of each slot information can be linked together and identified as the target process node, which makes it easier to identify the target function page and improves the user experience.

[0057] Based on the slot information obtained from processing voice requests and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the steps to determine the target in-vehicle system mini-program and the target function page include: Determine the location identifier of the target functional page corresponding to the target process node; Assign corresponding slot information to each corresponding process node to obtain the information of the target process node; Based on the information and location identifiers of the target process nodes, determine the target function page.

[0058] The processor is used to determine the location identifier of the target function page corresponding to the target process node, and assign the corresponding slot information to each corresponding process node to obtain the information of the target process node, and determine the target function page based on the information of the target process node and the location identifier.

[0059] Specifically, taking a restaurant mini-program A as an example, please refer to Table 2. Each node has a corresponding location identifier, which can be a Uniform Resource Locator (URL). The type of location identifier is not limited here. For example, in the voice request "Order product a from restaurant mini-program A to the company," the location identifier of the target function page corresponding to the target process node is obtained, which is "Confirm payment." Executing the location identifier corresponding to the target function page can present the user with a page containing the voice request slot information.

[0060] After the slot information and process node correspondence are matched, the server can assign slot information to each node, i.e., Open="Restaurant Mini Program A", Search="Product a", Specifications="default", Number="1", Add address="Company". After the assignment, the information of the node is: Search="Product a"; Specifications="default"; Number="1"; Add address="Company"; Confirm Payment, and the target process node is "Node Seven".

[0061] Finally, based on the information of the target process node, namely node seven, and the location identifier, namely "Confirmpayment," the target functional page can be determined as the "Confirm Payment Information" page. This allows the in-vehicle system to automatically complete steps such as opening the page, searching for the product, selecting product specifications, adding to the cart, confirming the order, and adding the delivery address without requiring manual user intervention. The system then directly leads to the user's payment confirmation page, where the user can directly confirm their payment information.

[0062] In this way, the information of the target process node can be obtained based on the matching results, and the target function page can be determined so that the target function page can be displayed through the vehicle to achieve the target function and complete the voice interaction.

[0063] Based on the slot information obtained from processing the voice request and the pre-built custom vocabulary of in-vehicle system mini-programs of predetermined categories, as well as the matching relationship of mini-program flow nodes, the target in-vehicle system mini-program and the target function page are determined to include: If a user-defined element does not exist in the custom dictionary, the preceding process node of the process node corresponding to the user-defined element will be determined as the target process node.

[0064] The processor is used to determine the preceding process node of the process node corresponding to the user-defined element as the target process node when the user-defined element does not exist in the custom dictionary.

[0065] Specifically, each user's voice request corresponds to a specific intent, and each specific intent corresponds to a unique location identifier. This means the server can only support direct access to the page and cannot directly manipulate elements within the page. To achieve "custom" element access, elements in a custom vocabulary must be defined in advance. Since each voice request corresponds to a new intent, for example, two voice requests, "Order product a from restaurant mini-program A to the office" and "Order product b from restaurant mini-program A to the office," differ only in the ordered product. However, because different products need to be stored in the shopping cart, their corresponding location identifiers are also different. Therefore, defining a minimal location identifier is akin to exhaustive searching. In the example above, for the voice request "Order product a from restaurant mini-program A to the office," if restaurant mini-program has discontinued product a, then the location identifier corresponding to product a will no longer exist in the custom vocabulary. If the voice assistant cannot execute the user's voice request containing "product a" and instead responds with "I don't understand" or similar, without returning to the previous page to allow the user to select other products, it not only reduces the efficiency of the voice interaction process but also results in a poor user experience.

[0066] In this application, when a user-defined element does not exist in the custom dictionary, only the custom dictionary needs to be queried. For example, when "product A" is removed from the merchant's listings and does not exist in the dictionary, there is no need to directly access the target function page. Instead, the previous process node that does not contain "product A" is identified as the target process node. That is, when the voice request is "Order product A from catering mini-program A to the office," "product A" is no longer in the dictionary. At this time, the mini-program cannot directly access the page to confirm payment information. Instead, it can send a voice reply to the user such as "product A is no longer available" and return to the product selection page so that the user can select other products.

[0067] In this way, when the user-defined element contained in the voice request slot information does not exist in the custom dictionary, the previous process node can be returned and identified as the target process node, instead of directly ending the voice interaction process. This makes it easier for users to make other choices in the custom dictionary and improves the user experience.

[0068] Step 02 also includes: Process voice requests to obtain intent information and slot information; Normalize the slot information; Based on the intent information and the normalized slot information, the target vehicle system mini-program and target function page are determined.

[0069] The processor is used to process voice requests to obtain intent information and slot information, normalize the slot information, and determine the target vehicle system applet and target function page based on the intent information and the normalized slot information.

[0070] Specifically, in scenarios involving voice interaction between users and non-shopping in-vehicle system mini-programs, there's no need to display multiple pages sequentially; the voice interaction process can be completed in a simpler way. First, the user's voice request is processed through natural language understanding to obtain the intent information and slot information of the voice request. Then, the slot information is normalized to finally determine the target in-vehicle system mini-program and the target function page, and the in-vehicle system mini-program and corresponding page are opened directly.

[0071] The process of users presenting their personal health status identification information requires them to manually click to enter and open the corresponding secondary page, which is relatively time-consuming. In one example of this application, when a user issues a voice request "Open Health Mini Program A for self-check," the intent information in the voice request can be identified through natural language understanding as "Open Health Mini Program A"; the slot information is "Health Mini Program A" and "Self-check," which, after normalization, yields the slot information as "Health Mini Program A" and "Self-check my personal health status identification information." Finally, by combining the intent information and the normalized slot information, the target in-vehicle system mini program and the target function page can be determined, and the corresponding target function page, namely the health status identification information page, can be directly opened, completing the voice interaction process.

[0072] In this way, in scenarios where users interact with non-shopping in-vehicle system mini-programs via voice, the intent information obtained from voice requests can be processed, combined with normalized slot information, to determine the target in-vehicle system mini-program and directly open the target function page, making the voice interaction process more time-saving and convenient.

[0073] The computer-readable storage medium of this application stores a computer program that, when executed by one or more processors, implements the above-described method.

[0074] In the description of this specification, the terms "above," "specifically," "similarly," "furthermore," etc., refer to specific features, structures, materials, or characteristics described in connection with embodiments or examples that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0075] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of executable request code comprising one or more steps for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0076] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A voice interaction method, characterized in that, include: Receive voice requests forwarded by vehicles; The voice request is processed, and the target in-vehicle system applet that responds to the voice request and the target function page in the target in-vehicle system applet that is used to implement the voice request are determined, so that the voice interaction is completed by displaying the target function page through the vehicle. The step of processing the voice request, determining the target in-vehicle system applet that responds to the voice request, and the target function page in the target in-vehicle system applet used to implement the voice request, so as to complete the voice interaction by displaying the target function page through the vehicle, includes: The target vehicle system applet is determined based on the slot information obtained from processing the voice request; Based on the matching relationship of the mini-program process nodes of the target vehicle system mini-program, the slot information is matched to obtain multiple corresponding process nodes. The matching relationship of the mini-program process nodes is used to define multiple process nodes arranged in sequence. Each corresponding process node corresponds to the location identifier of a function page, and each process node is related to the slot information that needs to be filled in the corresponding function page. When user-defined elements exist in the custom vocabulary of a pre-built vehicle system mini-program of a predetermined category, the corresponding process nodes are connected and combined in series according to the arrangement order of the multiple process nodes to form a node path from the starting process node to the ending process node. According to the matching relationship of the process nodes corresponding to the target vehicle system mini-program, each slot information is matched and the matched element directly reaches the set process node to determine the target process node. The slot information includes user-defined elements, and the custom vocabulary is a vocabulary containing multiple elements stored in the vehicle system mini-program by the information provider of the vehicle system mini-program. If the user-defined element does not exist in the custom vocabulary, the preceding process node of the process node corresponding to the user-defined element will be determined as the target process node. Determine the location identifier of the target function page corresponding to the target process node; Assign corresponding slot information to each corresponding process node to obtain the information of the target process node; The target function page is determined based on the information of the target process node and the location identifier.

2. The voice interaction method according to claim 1, characterized in that, The method further includes: Based on the intent information obtained from processing the voice request, it is confirmed that the voice request will be implemented through the in-vehicle system applet.

3. The voice interaction method according to claim 1, characterized in that, The predefined categories of in-vehicle system mini-programs include those for catering, travel, music, and / or payment.

4. The voice interaction method according to claim 1, characterized in that, The process of processing the voice request, determining the target in-vehicle system applet that responds to the voice request, and the target function page in the target in-vehicle system applet used to implement the voice request, so as to complete the voice interaction by displaying the target function page in the vehicle, includes: Process the voice request to obtain intent information and slot information; The slot information is normalized. Based on the intent information and the normalized slot information, the target vehicle system applet and the target function page are determined.

5. A server, characterized in that, The server includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the method according to any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by one or more processors, implements the method as described in any one of claims 1-4.

Images