Voice interaction method and server

Abstract

The present application provides a voice interaction method and a server. The voice interaction method includes: after the network state of the vehicle switches from the offline state to the online state, receiving a current voice request forwarded by the vehicle; querying local execution results corresponding to historical voice requests executed by the vehicle in the offline state stored in the server; performing entity extraction on the local execution results to obtain entity information; performing information inheritance on the current voice request according to the entity information to obtain a context information completion result; and outputting an execution instruction according to the context information completion result and sending it to the vehicle. The solution provided by the present application can maintain the continuity of voice interaction, reduce the user operation burden, improve driving safety and improve the user experience.

Description

Technical Field

[0001] This application relates to the field of voice interaction technology, and in particular, to a voice interaction method and a server. Background Art

[0002] Voice interaction is a new generation of interaction mode based on voice input. With the continuous development of the automotive industry and human-computer interaction technology, intelligent vehicles also provide users with voice interaction functions.

[0003] In related technologies, when the vehicle enters special environments such as basements and tunnels, it will temporarily enter a network-free state. At this time, users can execute basic voice commands through the vehicle's in-vehicle local terminal. However, when the vehicle exits the current environment and resumes the network state, it cannot inherit the context semantics of the voice commands in the network-free state, resulting in users needing to issue secondary voice commands for clarification to complete voice control.

[0004] However, the need for users to issue secondary voice commands for clarification affects the continuity of voice interaction, increases the user operation burden, may distract the user's attention, affect driving safety, and also reduces the user experience. Summary of the Invention

[0005] To solve or partially solve the problems existing in related technologies, this application provides a voice interaction method and a server, which can maintain the continuity of voice interaction, reduce the user operation burden, improve driving safety, and improve the user experience.

[0006] The first aspect of this application provides a voice interaction method, including:

[0007] After the network state of the vehicle switches from a network-free state to a networked state, receiving a current voice request forwarded by the vehicle;

[0008] Querying the local execution result corresponding to the historical voice request executed by the vehicle in the network-free state stored in the server;

[0009] Performing entity extraction on the local execution result to obtain entity information;

[0010] Inheriting information from the current voice request according to the entity information to obtain a context information completion result;

[0011] Outputting an execution instruction according to the context information completion result and sending it to the vehicle.

[0012] The local execution result corresponding to the historical voice request executed by the vehicle in the network-free state is determined in the following manner:

[0013] After the vehicle receives a historical voice request in the network-free state, parsing the historical voice request to obtain entity information;

[0014] Output a historical execution instruction according to the historical voice request and the entity information to control the vehicle, and obtain a local execution result corresponding to the historical voice request.

[0015] The method further includes:

[0016] The vehicle dynamically stores the local execution result by setting a semantic inheritance interval identifier value and / or a time stamp, and sends it to the server after switching to the networked state.

[0017] Querying the local execution result corresponding to the historical voice request executed by the vehicle stored in the server in the offline state, including:

[0018] Compare the semantic inheritance interval identifier value of the local execution result corresponding to the historical voice request executed by the vehicle stored in the server in the offline state with an interval identifier value threshold;

[0019] According to the semantic inheritance interval identifier value being less than or equal to the interval identifier value threshold, use the queried local execution result as the local execution result to be subjected to entity extraction.

[0020] Querying the local execution result corresponding to the historical voice request executed by the vehicle stored in the server in the offline state, including:

[0021] Compare the time stamp of the local execution result corresponding to the historical voice request executed by the vehicle stored in the server in the offline state with a time stamp threshold;

[0022] According to the time stamp being less than or equal to the time stamp threshold, use the queried local execution result as the local execution result to be subjected to entity extraction.

[0023] The information inheritance of the current voice request according to the entity information to obtain a context information completion result includes:

[0024] Use the entity information as the entity information corresponding to the current voice request to obtain a context information completion result including entity information.

[0025] The entity extraction of the local execution result to obtain entity information includes:

[0026] Perform entity extraction on the local execution result to obtain action entity information and vehicle control entity information.

[0027] A second aspect of the present application provides a server, including:

[0028] A request receiving module, configured to receive a current voice request forwarded by a vehicle after the network state of the vehicle switches from an offline state to an online state;

[0029] A result query module, configured to query the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server;

[0030] An information extraction module, configured to perform entity extraction on the local execution result queried by the result query module to obtain entity information;

[0031] An information completion module, configured to perform information inheritance on the current voice request according to the entity information obtained by the information extraction module to obtain a context information completion result;

[0032] An instruction output module, configured to output an execution instruction to the vehicle according to the context information completion result obtained by the information completion module.

[0033] A third aspect of the present application provides a server, including:

[0034] A processor; and

[0035] A memory, storing executable code thereon, which when executed by the processor, causes the processor to execute the method as described above.

[0036] A fourth aspect of the present application provides a computer-readable storage medium, storing executable code thereon, which when executed by a processor of an electronic device, causes the processor to execute the method as described above.

[0037] The technical solution provided by the present application may include the following beneficial effects:

[0038] In the solution of the present application, after the network state of the vehicle switches from the offline state to the online state, if the current voice request forwarded by the vehicle is received, first query the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server, and then perform entity extraction on the local execution result to obtain entity information, then information inheritance can be performed on the current voice request according to the entity information to obtain a context information completion result. After obtaining the context information completion result, the semantics of the current voice request and the historical voice request have forward and backward relevance and continuity, realizing semantic inheritance. Even if the current voice request is incomplete or ambiguous, the semantic information can be complemented according to the local execution result corresponding to the historical voice request, so as to accurately analyze the meaning of the user's current voice request and realize precise control operations on the vehicle's controls. Therefore, the solution of the present application can maintain the continuity of voice interaction, reduce the user's operation burden, improve driving safety and improve the user experience.

[0039] It should be understood that the above general description and subsequent detailed description are only exemplary and explanatory, and cannot limit the present application. Brief Description of the Drawings

[0040] The above and other objects, features, and advantages of the present application will become more apparent by describing the exemplary embodiments of the present application in more detail in conjunction with the accompanying drawings, wherein, in the exemplary embodiments of the present application, the same reference numerals generally represent the same components.

[0041] Figure 1 is a schematic flowchart of the voice interaction method shown in the present application;

[0042] Figure 2 is a schematic flowchart of another voice interaction method shown in the present application;

[0043] Figure 3 is a schematic flowchart of another voice interaction method shown in the present application;

[0044] Figure 4 is a schematic diagram of the application framework of the voice interaction method shown in the present application;

[0045] Figure 5 is a schematic diagram of the structure of the server shown in the present application;

[0046] Figure 6 is a schematic diagram of the structure of another server shown in the present application;

[0047] Figure 7 is a schematic diagram of the structure of the voice interaction system shown in the present application;

[0048] Figure 8 is a schematic diagram of the structure of the vehicle in the voice interaction system of the present application;

[0049] Figure 9 is a schematic diagram of the structure of another server shown in the present application. Detailed Description of the Embodiments

[0050] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0051] The terms used in the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0052] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of this application, "a plurality" means two or more, unless otherwise specifically defined.

[0053] The technical solution of this application will be described in detail below with reference to the accompanying drawings.

[0054] Figure 1 It is a schematic flowchart of the voice interaction method shown in this application. This method can be applied to a server.

[0055] See Figure 1 , this method includes:

[0056] S101. After the network state of the vehicle switches from the offline state to the online state, receive the current voice request forwarded by the vehicle.

[0057] When the vehicle enters special environments such as an underground garage or a tunnel, it will temporarily enter the offline state, but when the vehicle exits the special environment, it will resume the online state. After switching to the online state, the server can receive the current voice request forwarded by the vehicle.

[0058] S102. Query the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server.

[0059] In S102, the semantic inheritance interval identification value of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server can be compared with the interval identification value threshold; according to the semantic inheritance interval identification value being less than or equal to the interval identification value threshold, the queried local execution result is used as the local execution result for entity extraction to be performed.

[0060] In S102, the timestamp of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server can be compared with the timestamp threshold; according to the timestamp being less than or equal to the timestamp threshold, the queried local execution result is used as the local execution result for entity extraction to be performed.

[0061] Among them, the local execution result corresponding to the historical voice request executed by the vehicle in the offline state can be determined in the following manner:

[0062] After the vehicle receives a historical voice request in the offline state, it parses the historical voice request to obtain entity information; it outputs a historical execution instruction according to the historical voice request and the entity information to control the vehicle, and obtains a local execution result corresponding to the historical voice request.

[0063] The vehicle can dynamically store the local execution result by setting a semantic inheritance interval identifier value and / or a timestamp, and send it to the server after switching to the online state.

[0064] S103. Perform entity extraction on the local execution result to obtain entity information.

[0065] Among them, after the corresponding local execution result is queried, entity extraction can be performed on the local execution result to obtain action entity information and vehicle control entity information. For example, "close" is the action entity information, and "window" is the vehicle control entity information.

[0066] S104. Inherit information from the current voice request according to the entity information to obtain a context information completion result.

[0067] Among them, the entity information can be used as the entity information corresponding to the current voice request to obtain a context information completion result containing the entity information. For example, the current voice request is "open", and the information is incomplete at this time. However, the entity information obtained by performing entity extraction on the local execution result is "window". Using "window" as the entity information corresponding to the current voice request, a context information completion result of "open the window" can be obtained.

[0068] S105. Output an execution instruction according to the context information completion result and send it to the vehicle.

[0069] For example, if the context information completion result is "open the window", then the execution instruction "open the window" is output and sent to the vehicle. The vehicle opens the window of the vehicle according to the received execution instruction.

[0070] In the solution of this application, after the network state of the vehicle switches from the offline state to the online state, if a current voice request forwarded by the vehicle is received, first query the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server, and then perform entity extraction on the local execution result to obtain entity information. Then, the current voice request can be information-inherited based on the entity information to obtain a context information completion result. After obtaining the context information completion result, the semantics of the current voice request and the historical voice request have forward and backward relevance and continuity, realizing semantic inheritance. Even if the current voice request is incomplete or ambiguous, the semantic information can be completed according to the local execution result corresponding to the historical voice request, so as to accurately analyze the meaning of the user's current voice request and realize precise control operations on the vehicle's controls. Therefore, the solution of this application can maintain the continuity of voice interaction, reduce the user operation burden, improve driving safety and improve the user experience.

[0071] Figure 2 It is a schematic diagram of the offline processing flow on the vehicle side in the voice interaction method shown in this application. This method can be applied to the vehicle side.

[0072] See Figure 2 , this method includes:

[0073] S201. After the vehicle receives a historical voice request in the offline state, parse the historical voice request to obtain entity information.

[0074] In the offline state, after the vehicle receives the user's historical voice request, it can call an internal module to parse the user's historical voice request to obtain the corresponding entity information, including action entity information, vehicle control entity information, etc.

[0075] S202. Output a historical execution instruction according to the historical voice request and the entity information to control the vehicle, and obtain the local execution result corresponding to the historical voice request.

[0076] Among them, according to the parsed entity information and the historical voice request, a historical execution instruction can be output, and the vehicle's controls can be controlled according to the historical execution instruction to obtain the local execution result corresponding to the historical voice request.

[0077] S203. The vehicle dynamically stores the local execution result by setting a semantic inheritance interval identifier value and / or a time stamp, and sends it to the server after switching to the online state.

[0078] The vehicle can dynamically store the local execution result by setting a semantic inheritance interval identifier value and / or a time stamp. When the network resumes from the offline state to the online state, the vehicle can send the local execution result to the server for storage.

[0079] This application outputs historical execution instructions based on historical voice requests and entity information to control the vehicle, obtains the local execution results corresponding to the historical voice requests and stores them, and then sends them to the server after switching to the networked state, which can provide a reference for the subsequent server to recognize the historical context semantics and implement semantic inheritance.

[0080] Figure 3 It is a schematic flowchart of a voice interaction method shown in another embodiment of this application. Figure 4 It is a schematic diagram of an application framework of the voice interaction method shown in this application. Figure 3 and Figure 4 In the process shown, the solution of this application is introduced through the interaction process between the server and the vehicle (including the local terminal and the in-vehicle device).

[0081] In the related art, when the vehicle enters special environments such as the basement or tunnel, it will temporarily enter the offline state. At this time, the user can execute basic voice instructions through the vehicle's in-vehicle local terminal. However, when the vehicle exits the special environment and resumes the networked state, the related technical solutions cannot inherit the context semantics of the user's voice instructions in the offline state. For example, when the user issues a voice request of "close the window" in the tunnel offline state and issues a voice request of "open" when exiting the tunnel, since the context semantics in the offline state are not inherited at this time, the specific semantics of this "open" voice request cannot be recognized, and the window cannot be opened for the user. The related art generally needs to ask the user what operation to perform on which control, and the user needs to issue a secondary voice request for clarification and confirmation. When the user issues a complete voice request of "open the window" for the second time, the true intention of the user can be determined, and then the vehicle control can be completed. However, the user expects that the operations on various controls during the process of switching from the offline state to the networked state are sustainable. The process of secondary clarification and the need for the user to reissue the request not only takes time but also reduces the overall user experience. In addition, this switching environment is generally at positions such as the basement exit or the tunnel entrance and exit. At this time, the driving environment of the vehicle is complex, and the need for the user to clarify the voice again may distract the user's attention and affect the user's driving safety.

[0082] The solution of this application provides a continuous voice interaction solution with semantic inheritance when the network state switches, meeting the user's need for continuous voice interaction with the in-vehicle device in the case of network state switching in special environments such as tunnels and basements. For example, when the vehicle is driving in the tunnel, the user can operate controls such as the window, headlights, and air conditioner through voice instructions. When the vehicle exits the tunnel, the user can perform two-round fuzzy operations, and can also realize the semantic continuous control of controls such as the headlights by the user. For example, when executing the voice request of "turn on the headlights" issued by the user in the tunnel and receiving the voice request of "close" when exiting the tunnel, at this time, according to semantic inheritance, it can be determined that the user's voice request is "turn off the headlights", so there is no need for the user to clarify the voice request again, and "turn off the headlights" is directly executed.

[0083] See Figure 3 and Figure 4 , the method includes:

[0084] S301. The local terminal of the vehicle receives a voice request in a network - free state, parses the voice request to obtain entity information, and outputs an execution instruction according to the voice request and the entity information, and sends the execution instruction to the in - vehicle unit of the vehicle.

[0085] It should be noted that, if considered from the stage where the voice request is received by the slave server in a network - connected state, the voice request received by the vehicle in a network - free state at this time can relatively be called a historical voice request, and the execution instruction can be a historical execution instruction.

[0086] The local terminal of the vehicle receives the user's voice request, and after being processed by the internal module of the local terminal, the execution instruction is sent to the in - vehicle unit.

[0087] In a network - free state, after the local terminal receives the user's voice request, it will call each sub - module in the internal module to parse the user's voice request. Usually, the sub - modules can include a local ASR (Automatic Speech Recognition) module, a local NLU (Natural Language Understanding) module, etc. This application can utilize intent recognition technology. Intent recognition means enabling the search engine to recognize the information most relevant to the query input by the user. The voice request input by the user can be used as a query. By using each sub - module, the domain and intent corresponding to the user's query can be processed and obtained. For the query input by the user, according to the statistical classification model, the probability of each intent can be determined, and finally the intent of the query is given. For example, when receiving the voice request "close the window" from the user in a tunnel, the local terminal will recognize that the domain of this voice request is "control" and the intent is "window_close", that is, at this time, the action entity information can be recognized as "close", and the vehicle control entity information is "window". Therefore, the local terminal sends an execution instruction with the domain of "control" and the intent of "window_close" to the in - vehicle unit.

[0088] S302. The in - vehicle unit of the vehicle receives the execution instruction, controls the vehicle's controls according to the execution instruction, and returns the local execution result corresponding to the voice request to the local terminal.

[0089] The in - vehicle unit receives the execution instruction, performs the corresponding action, controls the vehicle's controls, and transmits the final execution result, that is, the local execution result, back to the local terminal.

[0090] Since the execution instructions that the local terminal can execute are limited, the final local execution result of the in-vehicle device will be sent back to the local terminal. Among them, the local execution result usually can include information such as execution classification, final execution result, etc., which can provide a reference for subsequent instruction execution.

[0091] S303. The local terminal of the vehicle receives the local execution result returned by the in-vehicle device, dynamically stores the local execution result, and sends it to the server after switching to the networked state.

[0092] The local terminal of the vehicle can dynamically store the local execution result by setting the semantic inheritance interval identifier value and / or timestamp. When the network resumes from the non-networked state to the networked state, the local terminal of the vehicle sends the local execution result to the database of the server, such as the data storage module, for storage.

[0093] In this application, the local terminal can dynamically store the local execution result by setting the semantic inheritance interval identifier value or setting the timestamp, or by setting both the semantic inheritance interval identifier value and the timestamp simultaneously.

[0094] First, the local terminal can dynamically set the historical query storage policy based on the statistical characteristics of the user's historical context semantic inheritance query, that is, obtain the dynamic storage policy including the dynamic query quantity and the maximum interval of dynamic semantic inheritance execution through the historical statistical characteristics of the semantic inheritance of the current user. Among them, this policy includes the user's dynamic query quantity and the maximum interval of the dynamic query. Then, according to the dynamic storage policy, determine the length of the historical execution queue of the local terminal, and in each interval polling, set according to the polling time, and dynamically store the interval identifier of each query in the queue in real time and use it as the basis for the context semantic input of the server. Among them, the local execution results of multiple rounds can be integrated and sent to the server. This policy needs to set the semantic inheritance interval identifier value or set the timestamp for the local execution result. This application can use the dynamic interval identifier and timestamp for dual judgment to meet the accurate inheritance needs of different users.

[0095] S304. After the network state of the vehicle switches from the non-networked state to the networked state, the server receives the current voice request forwarded by the vehicle.

[0096] It should be noted that if considering the stage when the vehicle receives the user's voice request in the networked state, the voice request forwarded by the vehicle received by the server at this time can be called the current voice request.

[0097] S305. Query the local execution result corresponding to the historical voice request executed by the vehicle in the non-networked state stored in the server.

[0098] After the server receives the user's current voice request in the networked state, it can first determine whether there is a local execution result corresponding to the historical voice request sent by the vehicle in the continuously monitored database, such as the data storage module. If it exists, the local execution result corresponding to the historical voice request is queried from the database.

[0099] Among them, the semantic inheritance interval identification value of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server can be compared with the interval identification value threshold; according to the semantic inheritance interval identification value being less than or equal to the interval identification value threshold, the queried local execution result is used as the local execution result to be subjected to entity extraction.

[0100] Among them, the timestamp of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server can be compared with the timestamp threshold; according to the timestamp being less than or equal to the timestamp threshold, the queried local execution result is used as the local execution result to be subjected to entity extraction.

[0101] For example, the server compares the interval identification value (semantic inheritance interval identification value) of each query in the result queue of the local execution result with the interval identification value threshold, and compares the timestamp of each query in the result queue of the local execution result with the timestamp threshold. The interval identification value threshold can be the maximum interval value of user semantic inheritance execution obtained from dynamic storage statistical features. If the conditions are met, that is, if the semantic inheritance interval identification value of the local execution result is less than or equal to the interval identification value threshold, and / or, the timestamp of the local execution result is less than or equal to the timestamp threshold, then the queried local execution result is used as the local execution result to be subjected to entity extraction, and subsequent entity extraction and query preprocessing can be performed.

[0102] For example, the maximum interval identification value of user semantic inheritance execution obtained by dynamic feature statistics is 5. There is a historical query information in the result queue of the local execution result, and the semantic inheritance interval identification value is 3, which is less than the maximum interval value of user semantic inheritance execution 5, and the timestamp is less than or equal to the timestamp threshold. Then the server can use the queried local execution result as the local execution result to be subjected to entity extraction.

[0103] S306. Perform entity extraction on the queried local execution result to obtain entity information.

[0104] In S306, entity extraction is performed on the locally executed results queried to obtain action entity information and vehicle control entity information. For example, the corresponding vehicle control entity information is identified, and at the same time, some meaningless words carried due to reasons such as user colloquialism are removed. For example, for "Help me open the window", the finally identified action entity information is "open", and the vehicle control entity information is "window".

[0105] S307. The server performs information inheritance on the current voice request according to the entity information to obtain the result of context information completion.

[0106] The server combines the entity information obtained in the previous step with the user's current voice request, that is, the current query, and calls the information inheritance module therein for fusion processing, that is, for missing information supplementation. For example, combination methods such as templates, knowledge bases, and neural network models can be used to supplement the missing information of the query to obtain the result of context information completion.

[0107] Taking the situation of a vehicle entering and exiting a tunnel as an example, when the "Increase temperature" voice request issued by the user is executed in a state where there is no network in the tunnel, and after exiting the tunnel, the "Decrease" voice request issued by the user is received in a networked state. When the vehicle exits the tunnel, the server queries the locally executed results corresponding to the historical voice request from the database, and it meets the maximum interval condition for user semantic inheritance. Therefore, entity extraction can be further performed to obtain the corresponding historical context entity information of "temperature" and "increase", that is, the historical action entity information is "increase", and the historical vehicle control entity information is "temperature".

[0108] Assume that the current voice request received in the networked state is "Decrease". Then, the current voice request "Decrease" and the obtained entity information, that is, the historical context entity information, can be passed into the information inheritance module. The information inheritance module can complete the context information supplementation to obtain the complete result of context information completion "Decrease temperature".

[0109] S308. Output an execution instruction according to the result of context information completion and send it to the vehicle.

[0110] Based on the complete result of context information completion obtained in the previous step being "Decrease temperature", the execution instruction "Decrease temperature" can be finally output.

[0111] The server outputs the execution instruction and sends it to the vehicle, and the in-vehicle computer of the vehicle operates the specific controls of the vehicle according to the execution instruction. For example, if the execution instruction is "Decrease temperature", the in-vehicle computer will decrease the temperature of the air-conditioning control of the vehicle.

[0112] In summary, the solution proposed in this application enables users to achieve continuous voice interaction with semantic inheritance when switching network environments, more accurately assisting users in precisely operating vehicle controls in complex environments. At the same time, this application also realizes the close integration of the vehicle's local terminal and the server. This application stores and updates the dynamic local execution results based on the user's historical features, obtaining a more user-semantic-habit-compliant end-cloud integration result. While meeting the maximum interval for semantic inheritance execution of users, this application expands the server context and further uses the information inheritance module to complete the user's semantics, not only fulfilling the requirement of continuous voice interaction with semantic inheritance when users switch network environments, enabling users to continuously and precisely operate vehicle controls in complex environments such as entering and exiting tunnels and basements, but also avoiding the problem of distracting users' attention caused by the need for users to clarify again in related technical solutions, improving the user experience and the driving safety of users.

[0113] Corresponding to the foregoing method for implementing application functions, this application also provides a server and a system.

[0114] Figure 5 It is a schematic structural diagram of the server shown in this application.

[0115] See Figure 5 , the server 50 provided in this application includes: a request receiving module 51, a result query module 52, an information extraction module 53, an information completion module 54, and an instruction output module 55.

[0116] The request receiving module 51 is used to receive the current voice request forwarded by the vehicle after the network status of the vehicle switches from the offline state to the online state. When the vehicle enters special environments such as basements and tunnels, it will temporarily enter the offline state, but when the vehicle exits the special environment, it will resume the online state. After switching to the online state, the request receiving module 51 can receive the current voice request forwarded by the vehicle.

[0117] The result query module 52 is used to query the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server 50. Among them, the local execution result corresponding to the historical voice request executed by the vehicle in the offline state can be determined in the following manner: after receiving the historical voice request in the offline state, the vehicle parses the historical voice request to obtain entity information; according to the historical voice request and the entity information, an historical execution instruction is output to control the vehicle, obtaining the local execution result corresponding to the historical voice request. The vehicle can dynamically store the local execution result by setting a semantic inheritance interval identifier value and / or a time stamp, and send it to the server 50 after switching to the online state.

[0118] An information extraction module 53, configured to perform entity extraction on the local execution results queried by the result query module 52 to obtain entity information.

[0119] An information completion module 54, configured to perform information inheritance on the current voice request according to the entity information obtained by the information extraction module 53 to obtain a context information completion result.

[0120] An instruction output module 55, configured to output an execution instruction to the vehicle according to the context information completion result obtained by the information completion module 54. For example, if the context information completion result obtained by the instruction output module 55 is "open the window", the execution instruction "open the window" is output to the vehicle, and the vehicle opens the window of the vehicle according to the received execution instruction.

[0121] Figure 6 It is a schematic structural diagram of another server shown in the present application.

[0122] See Figure 6 , the server 50 provided by the present application includes: a request receiving module 51, a result query module 52, an information extraction module 53, an information completion module 54, and an instruction output module 55. The result query module 52 includes a first query sub-module 521 and a second query sub-module 522.

[0123] Among them, the functions of the request receiving module 51, the result query module 52, the information extraction module 53, the information completion module 54, and the instruction output module 55 can be seen in Figure 5 the description in.

[0124] The first query sub-module 521 is configured to compare the semantic inheritance interval identification value of the local execution result corresponding to the historical voice request executed by the vehicle stored in the server 50 with the interval identification value threshold; according to the semantic inheritance interval identification value being less than or equal to the interval identification value threshold, the queried local execution result is used as the local execution result to be subjected to entity extraction.

[0125] The second query sub-module 522 is configured to compare the time stamp of the local execution result corresponding to the historical voice request executed by the vehicle stored in the server 50 with the time stamp threshold; according to the time stamp being less than or equal to the time stamp threshold, the queried local execution result is used as the local execution result to be subjected to entity extraction.

[0126] The information completion module 54 can use the entity information as the entity information corresponding to the current voice request to obtain a context information completion result containing the entity information. For example, when the current voice request is "Open", the information is incomplete at this time. However, based on the entity information obtained by entity extraction from the local execution result, if the entity information is "window", using "window" as the entity information corresponding to the current voice request, the context information completion result "Open the window" can be obtained.

[0127] The information extraction module 53 can perform entity extraction on the local execution result to obtain action entity information and vehicle control entity information.

[0128] After the network status of the vehicle in the server of this application switches from the offline state to the online state, if it receives the current voice request forwarded by the vehicle, it first queries the local execution result corresponding to the historical voice request executed by the vehicle stored in the server, and then performs entity extraction on the local execution result to obtain entity information. Then, it can perform information inheritance on the current voice request according to the entity information to obtain a context information completion result. After obtaining the context information completion result, it makes the semantics of the current voice request have forward and backward relevance and continuity with the historical voice request, realizing semantic inheritance. Even if the current voice request is incomplete or ambiguous, the semantic information can be completed according to the local execution result corresponding to the historical voice request, so as to accurately analyze the meaning of the user's current voice request and realize precise control operations on the vehicle's controls. Therefore, the solution of this application can maintain the continuity of voice interaction, reduce the user's operation burden, improve driving safety and improve the user experience.

[0129] Figure 7 It is a schematic structural diagram of the voice interaction system shown in this application.

[0130] See Figure 7 , the voice interaction system 70 provided by this application includes: a server 71 and a vehicle 72.

[0131] The server 71 is used to, after the network status of the vehicle switches from the offline state to the online state, receive the current voice request forwarded by the vehicle; query the local execution result corresponding to the historical voice request executed by the vehicle stored in the server 71; perform entity extraction on the local execution result to obtain entity information; perform information inheritance on the current voice request according to the entity information to obtain a context information completion result; and output an execution instruction to the vehicle according to the context information completion result.

[0132] The vehicle 72 is used to, after the network status of the vehicle switches from the offline state to the online state, forward the current voice request to the server 71.

[0133] Among them, the structure and function of the server 71 can be referred to Figure 5 or Figure 6Description of server 50

[0134] Figure 8 It is a schematic structural diagram of a vehicle in the voice interaction system of the present application

[0135] See Figure 8 , the vehicle 72 provided by the present application includes: an information parsing module 721, an execution result module 722, and a storage setting module 723

[0136] The information parsing module 721 is configured to parse the historical voice request to obtain entity information after the vehicle 72 receives the historical voice request in the offline state

[0137] The execution result module 722 is configured to output a historical execution instruction according to the historical voice request and the entity information to control the vehicle 72, and obtain a local execution result corresponding to the historical voice request

[0138] The storage setting module 723 is configured to dynamically store the local execution result by setting a semantic inheritance interval identifier value and / or a time stamp, and send it to the server after switching to the online state

[0139] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here

[0140] Figure 9 It is a schematic structural diagram of another server shown in the present application

[0141] See Figure 9 , the server 1000 includes a memory 1010 and a processor 1020

[0142] The processor 1020 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc

[0143] The memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, the ROM may store static data or instructions required by the processor 1020 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even when the computer is powered off. In some embodiments, the permanent storage device employs a mass storage device (such as a magnetic or optical disk, flash memory) as the permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as dynamic random access memory. The system memory may store some or all of the instructions and data required by the processor during operation. In addition, the memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and magnetic disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a removable storage device that is readable and / or writable, such as a compact disc (CD), read-only digital versatile disc (such as DVD-ROM, dual-layer DVD-ROM), read-only Blu-ray disc, super density disc, flash memory card (such as SD card, min SD card, Micro-SD card, etc.), magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and instantaneous electronic signals transmitted wirelessly or wired.

[0144] Executable code is stored on the memory 1010, and when the executable code is processed by the processor 1020, it may cause the processor 1020 to execute some or all of the methods described above.

[0145] In addition, the method according to the present application may also be implemented as a computer program or computer program product, which includes computer program code instructions for executing some or all of the above steps of the method according to the present application.

[0146] Alternatively, the present application may also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium), on which executable code (or computer program or computer instruction code) is stored. When the executable code (or computer program or computer instruction code) is executed by a processor of an electronic device (or a server, etc.), it causes the processor to execute some or all of the steps of the above method according to the present application.

[0147] The embodiments of the present application have been described above. The above description is exemplary and not exhaustive, and is also not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to technologies in the market, or to enable other ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A voice interaction method, characterized in that, it includes: After the network state of the vehicle switches from the offline state to the online state, receiving the current voice request forwarded by the vehicle; Querying the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server; Performing entity extraction on the local execution result to obtain entity information; Inheriting information from the current voice request according to the entity information to obtain a context information completion result; Outputting an execution instruction according to the context information completion result and sending it to the vehicle.

2. The method according to claim 1, characterized in that, The local execution result corresponding to the historical voice request executed by the vehicle in the offline state is determined in the following manner: After the vehicle receives a historical voice request in the offline state, parsing the historical voice request to obtain entity information; Outputting a historical execution instruction according to the historical voice request and the entity information to control the vehicle, and obtaining the local execution result corresponding to the historical voice request.

3. The method according to claim 2, characterized in that, The method further includes: The vehicle dynamically stores the local execution result by setting a semantic inheritance interval identifier value and / or a timestamp, and sends it to the server after switching to the online state.

4. The method according to claim 1, characterized in that, Querying the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server includes: Comparing the semantic inheritance interval identifier value of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server with an interval identifier value threshold; According to the semantic inheritance interval identifier value being less than or equal to the interval identifier value threshold, using the queried local execution result as the local execution result to be subjected to entity extraction.

5. The method according to claim 1, characterized in that, Querying the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server includes: Comparing the timestamp of the local execution result corresponding to the historical voice request executed by the vehicle in the offline state stored in the server with a timestamp threshold; According to the timestamp being less than or equal to the timestamp threshold, using the queried local execution result as the local execution result to be subjected to entity extraction.

6. The method according to any one of claims 1 to 5, characterized in that, Inheriting information from the current voice request according to the entity information to obtain a context information completion result includes: Using the entity information as the entity information corresponding to the current voice request to obtain a context information completion result containing entity information.

7. The method according to any one of claims 1 to 5, characterized in that, Performing entity extraction on the local execution result to obtain entity information includes: Performing entity extraction on the local execution result to obtain action entity information and vehicle control entity information.

8. A server, characterized in that, it includes: A request receiving module, configured to receive the current voice request forwarded by the vehicle after the network state of the vehicle switches from the offline state to the online state; A result query module, configured to query the local execution results corresponding to the historical voice requests executed by the vehicle stored in the server in a network - free state; An information extraction module, configured to perform entity extraction on the local execution results queried by the result query module to obtain entity information; An information completion module, configured to perform information inheritance on the current voice request according to the entity information obtained by the information extraction module to obtain a context information completion result; An instruction output module, configured to output an execution instruction to the vehicle according to the context information completion result obtained by the information completion module.

9. A server, characterized in that, it includes: a processor; and a memory, on which executable code is stored, and when the executable code is executed by the processor, the processor is caused to execute the method according to any one of claims 1 - 7.

10. A computer - readable storage medium, on which executable code is stored, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method according to any one of claims 1 - 7.

Images