Semantic analysis method and device for vehicle-mounted voice, electronic equipment and storage medium
By comparing parameters of online and offline semantic parsing and iteratively maintaining them in modules, the problem of inconsistency between online and offline semantic parsing results was solved, improving user experience and reducing development costs.
Patent Information
- Application Number
- CN202511500016.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-20
- Publication Date
- 2026-01-09
AI Technical Summary
The difference between online and offline semantics leads to inconsistent parsing results for the same semantic meaning, affecting user experience and increasing development costs.
After converting user-input speech into text, online and offline parsing operations are performed. The tuple parameters and specific semantic parameters are compared to ensure consistency of results. Semantic consistency is also guaranteed through modular iterative maintenance.
It achieves consistency between online and offline semantic parsing results, improves user experience, and reduces development costs.
Smart Images
Figure CN121306141A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle technology, and in particular to a semantic parsing method, device, electronic device, and storage medium for in-vehicle voice. Background Technology
[0002] In the wave of intelligent vehicle development, in-vehicle voice interaction systems have become a key technology for improving driving experience and safety.
[0003] Semantic understanding is a crucial aspect of using in-vehicle voice interaction systems. To better ensure a good user experience, the semantic understanding function is divided into online and offline semantic parsing. The advantage of online semantic parsing is its real-time updates, allowing it to quickly adapt to new functions and features; the advantage of offline semantic parsing is its fast response speed, ensuring the availability of the voice interaction system even in environments with poor network conditions.
[0004] However, with version iterations and the enrichment of semantics, the differences between online and offline semantics are becoming increasingly significant. This leads to two different parsing results for the same semantic statement, impacting user experience. Furthermore, due to rapid semantic updates and a large number of semantic statements, checking each one individually would greatly increase development costs. Summary of the Invention
[0005] The purpose of this invention is to provide a semantic parsing method, device, electronic device, and storage medium for in-vehicle voice, which can at least realize online and offline semantic parsing and difference comparison of in-vehicle voice, and then perform adaptive modular iterative maintenance on the vehicle based on the comparison results, ensuring that the user obtains the same results regardless of whether online or offline semantic parsing is used, thereby improving user satisfaction with voice interaction and reducing development costs.
[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a semantic parsing method for in-vehicle voice, comprising at least:
[0007] Responding to user voice input, converting user-inputted speech into user-inputted text;
[0008] Perform online parsing on the user input text to obtain online parsing semantics, and perform offline parsing on the user input text to obtain offline parsing semantics;
[0009] The tuple parameters of the online parsing semantics are compared with the tuple parameters of the offline parsing semantics to obtain a first comparison result;
[0010] The specific semantic parameters of the online parsing semantics are compared with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result;
[0011] Based on the first comparison result and the second comparison result, perform modular iterative maintenance on the vehicle to at least ensure online and offline semantic consistency.
[0012] Optionally, after comparing the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result, the method further includes:
[0013] Based on the first comparison result and the second comparison result, anomaly semantics are determined, and based on the anomaly semantics and preset business logic, final semantics are determined so as to distribute the final semantics to at least one functional application to realize the user's expected operation.
[0014] Optionally, the multivariate parameter set includes at least domain parameters, behavior parameters, purpose parameters, and type parameters.
[0015] Optionally, the modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the iterative maintenance of the ride comfort function module, and the iterative maintenance of the infotainment function module.
[0016] Based on the same concept, in a second aspect, the present invention also provides a semantic parsing apparatus for in-vehicle voice, for performing the semantic parsing method for in-vehicle voice as described in any of the first aspects;
[0017] The semantic parsing device for in-vehicle voice communication includes at least:
[0018] The speech conversion module is used to respond to the user's voice input and convert the user's voice input into the user's text input.
[0019] The online parsing module is used to perform online parsing operations on the user input text to obtain online parsing semantics, and to perform offline parsing operations on the user input text to obtain offline parsing semantics;
[0020] The first comparison module is used to compare the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result;
[0021] The second comparison module is used to compare the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result.
[0022] The iterative maintenance module is used to perform sub-module iterative maintenance on the vehicle based on the first comparison result and the second comparison result, so as to at least ensure online and offline semantic consistency.
[0023] Optionally, it also includes:
[0024] The semantic execution module is used to determine abnormal semantics based on the first comparison result and the second comparison result, and to determine the final semantics based on the abnormal semantics and preset business logic, so as to distribute the final semantics to at least one functional application and realize the user's expected operation.
[0025] Optionally, the multivariate parameter set includes at least domain parameters, behavior parameters, purpose parameters, and type parameters.
[0026] Optionally, the modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the iterative maintenance of the ride comfort function module, and the iterative maintenance of the infotainment function module.
[0027] Based on the same concept, in a third aspect, the present invention also provides an electronic device, including a memory and a processor, the memory storing a computer program executable on the processor, wherein the processor, when executing the program, implements the steps of the semantic parsing method for in-vehicle voice as described in any of the first aspects.
[0028] Based on the same concept, in a fourth aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the semantic parsing methods for in-vehicle voice in the first aspect.
[0029] The technical solution provided in this embodiment of the invention firstly converts the user's voice input into user-input text in response to the user's voice input; further, it performs online parsing on the user-input text to obtain online parsing semantics, and performs offline parsing on the user-input text to obtain offline parsing semantics; further, it compares the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result; further, it compares the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result; finally, it performs modular iterative maintenance on the vehicle based on the first comparison result and the second comparison result to at least ensure consistency between online and offline semantics.
[0030] Therefore, the embodiments of the present invention can at least realize online and offline semantic parsing and difference comparison of in-vehicle voice, and then perform adaptive modular iterative maintenance on the vehicle based on the comparison results, so as to ensure that the user obtains the same result regardless of whether online or offline semantic parsing is used, which helps to improve the user's satisfaction with voice interaction and reduce development costs. Attached Figure Description
[0031] Figure 1 This is a flowchart of a semantic parsing method for in-vehicle voice provided in an embodiment of the present invention;
[0032] Figure 2This is a flowchart of another semantic parsing method for in-vehicle voice provided in an embodiment of the present invention;
[0033] Figure 3 This is a schematic diagram of the structure of a semantic parsing device for in-vehicle voice provided in an embodiment of the present invention;
[0034] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0035] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0036] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.
[0037] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.
[0038] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.
[0039] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”
[0040] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.
[0041] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.
[0042] Figure 1 This is a flowchart of a semantic parsing method for in-vehicle voice according to an embodiment of the present invention. This embodiment is applicable to at least any iterative testing and maintenance scenario of in-vehicle voice systems in any vehicle. The semantic parsing method for in-vehicle voice systems can be, but is not limited to, executed by the semantic parsing device for in-vehicle voice systems in this embodiment of the present invention. This execution entity can be implemented in software and / or hardware. Figure 1 As shown, the semantic parsing method for in-vehicle voice communication includes at least the following steps:
[0043] S1. Responding to the user's voice input, convert the user's voice input into the user's text input.
[0044] When converting user-input speech into user-input text, any existing speech-to-text (STT) technology can be used.
[0045] S2. Perform online parsing on the user input text to obtain online parsing semantics, and perform offline parsing on the user input text to obtain offline parsing semantics.
[0046] Online parsing can involve sending user-input text (e.g., "I want to listen to Zhou XX's Sunny Day") to the cloud via a network data interface, where the cloud processes the text into data objects needed by the vehicle-side application (e.g., containing quadruples, detailed semantics, and other required parameters). This is online semantic parsing. Conversely, offline parsing can involve recognizing the original semantics of user-input text through a local app (without relying on the network), and then parsing the original semantics into data objects needed by the vehicle-side application. This is offline semantic parsing.
[0047] S3. Compare the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain the first comparison result.
[0048] The multi-tuple parameter can refer to the aforementioned quadruple; in a specific implementation, the multi-tuple parameter may optionally include at least domain parameters, behavior parameters, purpose parameters, and type parameters.
[0049] In another specific implementation, step S3 essentially involves comparing the four-tuples of online and offline semantic parsing: Domain (i.e., the aforementioned domain parameter), Action (i.e., the aforementioned behavior parameter), Intention (i.e., the aforementioned purpose parameter), and Type (i.e., the aforementioned type parameter). If any parameter in the four-tuples of online and offline semantic parsing is different, the first comparison result is determined to be different, because the four-tuple has the function of uniquely determining the semantic distribution application and the specific application's execution operation.
[0050] S4. Compare the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain the second comparison result.
[0051] The specific semantic parameter can refer to `Semantic`. Specifically, `Semantic` is a JSON string that may contain one or more layers of information. During step S4, each field in the JSON string needs to be compared. For multi-level structures, a depth-first search is required to compare the fields. Fields that do not affect the final judgment result also need to be filtered. After traversing the online results (i.e., the aforementioned online parsing semantics) and offline results (i.e., the aforementioned offline parsing semantics), if the valid information in the two results is the same, then the comparison result is determined to be consistent (i.e., the second comparison result is determined to be consistent).
[0052] For example, `Semantic` is a dynamic additional field containing information to specify a concrete command, such as "turn on the air conditioning." The `Semantic` can include location information to specify which location in the vehicle's air conditioning will be turned on. It's understandable that comparing `Semantic` values isn't simply a matter of comparing parameter consistency. In practice, separate comparisons are needed based on different business logic and parameters. Some parameters may differ, but this doesn't necessarily affect the final execution result. For instance, the `Semantic` field for an online parsing semantic might be "semantic='{"position":"UNKNOWN"}'", while the corresponding offline parsing semantic's `Semantic` field is "semantic='{"position":"FRONT_LEFT"}'". Although the `Semantic` values aren't completely identical, the final executed command is to turn on the driver's side air conditioning. Therefore, in automated comparisons of `Semantic` results, they are considered identical. In other words, in automated comparisons, the position information needs to be processed separately, and both are considered identical in the above scenario.
[0053] S5. Perform modular iterative maintenance on the vehicle based on the first comparison result and the second comparison result to at least ensure online and offline semantic consistency.
[0054] The vehicle modules that can be iteratively maintained can be of various types. In another specific implementation, optionally, the modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the ride comfort function module, and the infotainment function module.
[0055] The technical solution provided in this embodiment firstly converts the user's voice input into user-input text in response to the user's voice input; further, it performs online parsing on the user-input text to obtain online parsing semantics, and performs offline parsing on the user-input text to obtain offline parsing semantics; further, it compares the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result; further, it compares the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result; finally, it performs modular iterative maintenance on the vehicle based on the first comparison result and the second comparison result to at least ensure consistency between online and offline semantics.
[0056] Therefore, this embodiment can at least realize online and offline semantic parsing and difference comparison of in-vehicle voice, and then perform adaptive modular iterative maintenance on the vehicle based on the comparison results, ensuring that the user gets the same result regardless of whether online or offline semantic parsing is used, which helps to improve user satisfaction with voice interaction and reduce development costs.
[0057] It should be noted that, in another specific implementation, optionally, after comparing the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result, the method further includes:
[0058] The abnormal semantics are determined based on the first comparison result and the second comparison result, and the final semantics are determined based on the abnormal semantics and the preset business logic, so as to distribute the final semantics to at least one functional application and realize the user's expected operation.
[0059] Based on this Figure 2 This is a flowchart of another semantic parsing method for in-vehicle voice provided in an embodiment of the present invention, see [link / reference]. Figure 2 The semantic parsing method for in-vehicle voice communication includes at least the following steps:
[0060] S1. Responding to the user's voice input, convert the user's voice input into the user's text input.
[0061] S2. Perform online parsing on the user input text to obtain online parsing semantics, and perform offline parsing on the user input text to obtain offline parsing semantics.
[0062] S3. Compare the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain the first comparison result.
[0063] S4. Compare the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain the second comparison result.
[0064] S6. Determine the abnormal semantics based on the first comparison result and the second comparison result, and determine the final semantics based on the abnormal semantics and the preset business logic, so as to distribute the final semantics to at least one functional application and realize the user's expected operation.
[0065] The preset business logic can correspond to functional applications. When the first comparison result and / or the second comparison result are inconsistent or different, the different semantics (i.e., the aforementioned abnormal semantics) can be sorted out based on the comparison results, and then each item can be reviewed according to the specific business, and finally, based on the business logic, it can be determined whether to prioritize online or offline semantic parsing.
[0066] S5. Perform modular iterative maintenance on the vehicle based on the first comparison result and the second comparison result to at least ensure online and offline semantic consistency.
[0067] In another specific implementation, the technical solution of this embodiment can also be applied to the vehicle testing process, and the specific steps are as follows:
[0068] 1. Prepare a test dataset with two semantic entries in the dataset list: "Open the online music recommendation page" and "I want to listen to Zhou XX's Sunny Day".
[0069] 2. The semantics in the dataset were parsed sequentially, both online and offline. The parsing results are as follows:
[0070] (1) Open the online music recommendation page:
[0071] Online semantics:
[0072] {domain='cmd', intention='instructing', action='execute', query='Open the online music recommendation page', type='open_net_music', tts='OK', semantic='{"label":"_KU_WO_RECOMMEND_"}', reply={"open_net_music":[]}}.
[0073] Offline semantics:
[0074] {domain='nlu_reject', intention='nlu_reject', action='execute',queryId='', query='Open the online music recommendation page', type='nlu_reject', tts='Sorry, the network is not good, this function is not supported at the moment', displayGuide=null, semantic=''}.
[0075] (2) I want to hear Zhou XX's "Sunny Day":
[0076] Online semantics:
[0077] {domain='music', intention='searching', action='execute', query='I want to listen to Zhou XX's Sunny Day', type='song', tts='Okay', displayGuide=null, semantic='{"singer":["Zhou XX"],"song":"Sunny Day"}', needCpService=false, dispatchInfo=null,reply={"song":[]}}.
[0078] Offline semantics:
[0079] {domain='music', intention='searching', action='execute', queryId='',query='I want to listen to "Sunny Day" by Zhou XX', type='song', tts='There is no such song as "Sunny Day" for you. How about trying another song?', displayGuide=['Play Bluetooth music', 'Pause music', 'Single loop'], semantic='{"singer":["Zhou XX"],"song":"Sunny Day"}'}。
[0080] 3. Automatically compare the semantic parsing results in the dataset
[0081] The comparison results are as follows:
[0082] (1) Open the recommended page of online music:
[0083] Offline parsing result: domain: nlu_reject intention: nlu_reject action: executetype: nlu_reject Semantic:"".
[0084] Online parsing result: domain: cmd intention: instructing action: execute type:open_net_music Semantic: {"label":"_KU_WO_RECOMMEND_"}。
[0085] It can be seen that the comparison results are judged inconsistently.
[0086] (2) I want to listen to "Sunny Day" by Zhou XX:
[0087] Online parsing result: domain: music intention: searching action: execute type:song Semantic: {"singer":"Zhou XX", "song":"Sunny Day"}。
[0088] Offline parsing result: domain: music intention: searching action: execute type:song Semantic: {"singer":"Zhou XX", "song":"Sunny Day"}。
[0089] It can be seen that the comparison results are judged consistently.
[0090] 4. Output all parsed semantics and comparison results to form a document for use in performing modular iterative maintenance of the vehicle, ensuring at least online and offline semantic consistency.
[0091] In view of this, the technical solution of this embodiment can quickly identify all existing semantics (those with discrepancies between offline and online versions). In subsequent versions, known discrepancies in semantics can be iteratively maintained to ensure consistency between online and offline semantics, providing users with the same experience (preventing inconsistencies in semantic execution logic due to network issues). When adding new semantics, the consistency between online and offline semantics can be checked in real time, improving the accuracy of instructions.
[0092] Figure 3 This is a schematic diagram of the structure of a semantic parsing device for in-vehicle voice according to an embodiment of the present invention. This embodiment is applicable to at least any scenario of iterative testing and maintenance of in-vehicle voice systems in any vehicle. The semantic parsing device for in-vehicle voice systems can be implemented in software and / or hardware. Figure 3 As shown, the in-vehicle voice semantic parsing device is used to execute the in-vehicle voice semantic parsing method described in any of the foregoing embodiments or implementations. The in-vehicle voice semantic parsing device includes at least:
[0093] The speech conversion module 110 is used to convert user-input speech into user-input text in response to user-input speech input;
[0094] The online parsing module 120 is used to perform online parsing operations on user input text to obtain online parsing semantics, and to perform offline parsing operations on user input text to obtain offline parsing semantics;
[0095] The first comparison module 130 is used to compare the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain the first comparison result;
[0096] The second comparison module 140 is used to compare the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result.
[0097] The iterative maintenance module 150 is used to perform sub-module iterative maintenance on the vehicle based on the first comparison result and the second comparison result, so as to at least ensure online and offline semantic consistency.
[0098] Optionally, it also includes:
[0099] The semantic execution module 160 is used to determine abnormal semantics based on the first comparison result and the second comparison result, and to determine the final semantics based on the abnormal semantics and preset business logic, so as to distribute the final semantics to at least one functional application and realize the user's expected operation.
[0100] Optionally, the multivariate parameter set may include at least domain parameters, behavior parameters, purpose parameters, and type parameters.
[0101] Optionally, the modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the ride comfort function module, and the infotainment function module.
[0102] The technical solution provided in this embodiment firstly, in response to the user's voice input, converts the user's input voice into user input text through a voice conversion module; further, performs online parsing on the user input text through an online / offline parsing module to obtain online parsing semantics, and performs offline parsing on the user input text to obtain offline parsing semantics; further, compares the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics through a first comparison module to obtain a first comparison result; further, compares the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics through a second comparison module to obtain a second comparison result; finally, performs modular iterative maintenance on the vehicle based on the first and second comparison results through an iterative maintenance module to at least ensure consistency between online and offline semantics.
[0103] Therefore, this embodiment can at least realize online and offline semantic parsing and difference comparison of in-vehicle voice, and then perform adaptive modular iterative maintenance on the vehicle based on the comparison results, ensuring that the user gets the same result regardless of whether online or offline semantic parsing is used, which helps to improve user satisfaction with voice interaction and reduce development costs.
[0104] This embodiment provides an electronic device. Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. See also: Figure 4The electronic device 1000 includes a processor 1001 and a memory 1002. The memory 1002 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 1001, the steps in any of the above-mentioned semantic parsing methods for in-vehicle voice are performed. Through the above technical solution, the processor 1001 and the memory 1002 are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanism (not shown). The memory 1002 stores a computer program executable by the processor. When the electronic device 1000 is running, the processor 1001 executes the computer program to execute the semantic parsing method for vehicle voice in any optional implementation of the above embodiments, so as to at least achieve the following functions: in response to the user's voice input, converting the user's input voice into user input text; performing online parsing operation on the user input text to obtain online parsing semantics, and performing offline parsing operation on the user input text to obtain offline parsing semantics; comparing the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result; comparing the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result; performing modular iterative maintenance on the vehicle according to the first comparison result and the second comparison result to at least ensure consistency between online and offline semantics.
[0105] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the semantic parsing method for in-vehicle voice as provided in all embodiments of this application: in response to a user's voice input, converting the user's input voice into user-input text; performing online parsing on the user-input text to obtain online parsing semantics, and performing offline parsing on the user-input text to obtain offline parsing semantics; comparing the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result; comparing the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result; and performing modular iterative maintenance on the vehicle based on the first and second comparison results to at least ensure consistency between online and offline semantics.
[0106] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.
[0107] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.
[0108] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.
[0109] Computer program code for performing the operations of this invention can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, and C++, as well as conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A semantic parsing method for in-vehicle voice, characterized in that, At least including: Responding to user voice input, converting user-inputted speech into user-inputted text; Perform online parsing on the user input text to obtain online parsing semantics, and perform offline parsing on the user input text to obtain offline parsing semantics; The tuple parameters of the online parsing semantics are compared with the tuple parameters of the offline parsing semantics to obtain a first comparison result; The specific semantic parameters of the online parsing semantics are compared with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result; Based on the first comparison result and the second comparison result, perform modular iterative maintenance on the vehicle to at least ensure online and offline semantic consistency.
2. The semantic parsing method for in-vehicle voice according to claim 1, characterized in that, After comparing the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result, the method further includes: Based on the first comparison result and the second comparison result, anomaly semantics are determined, and based on the anomaly semantics and preset business logic, final semantics are determined so as to distribute the final semantics to at least one functional application to realize the user's expected operation.
3. The semantic parsing method for in-vehicle voice according to claim 1, characterized in that, The multivariate parameter set includes at least domain parameters, behavior parameters, purpose parameters, and type parameters.
4. The semantic parsing method for in-vehicle voice according to claim 1, characterized in that, The modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the ride comfort function module, and the infotainment function module.
5. A semantic parsing device for in-vehicle voice, characterized in that, Used to perform the semantic parsing method for in-vehicle voice as described in any one of claims 1-4; The semantic parsing device for in-vehicle voice communication includes at least: The speech conversion module is used to respond to the user's voice input and convert the user's voice input into the user's text input. The online parsing module is used to perform online parsing operations on the user input text to obtain online parsing semantics, and to perform offline parsing operations on the user input text to obtain offline parsing semantics; The first comparison module is used to compare the tuple parameters of the online parsing semantics with the tuple parameters of the offline parsing semantics to obtain a first comparison result; The second comparison module is used to compare the specific semantic parameters of the online parsing semantics with the specific semantic parameters of the offline parsing semantics to obtain a second comparison result. The iterative maintenance module is used to perform sub-module iterative maintenance on the vehicle based on the first comparison result and the second comparison result, so as to at least ensure online and offline semantic consistency.
6. The semantic parsing device for in-vehicle voice according to claim 5, characterized in that, Also includes: The semantic execution module is used to determine abnormal semantics based on the first comparison result and the second comparison result, and to determine the final semantics based on the abnormal semantics and preset business logic, so as to distribute the final semantics to at least one functional application and realize the user's expected operation.
7. The semantic parsing device for in-vehicle voice according to claim 5, characterized in that, The multivariate parameter set includes at least domain parameters, behavior parameters, purpose parameters, and type parameters.
8. The semantic parsing device for in-vehicle voice according to claim 5, characterized in that, The modular iterative maintenance includes at least the iterative maintenance of the driving control function module, the ride comfort function module, and the infotainment function module.
9. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the semantic parsing method for in-vehicle voice according to any one of claims 1 to 4.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the semantic parsing method for in-vehicle voice according to any one of claims 1 to 4.