A voice processing method, device, equipment and storage medium

By determining the context information of the speech information and combining the semantic analysis results and selecting the target analysis results, the semantic miss call problem in speech interaction is solved, the precise positioning of user intentions is achieved, and the degree of intelligence and user satisfaction is improved.

CN113744728BActive Publication Date: 2024-11-19APOLLO INTELLIGENT CONNECTIVITY (BEIJING) TECH CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202111012283.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-08-31
Publication Date
2024-11-19
Estimated Expiration
2041-08-31

AI Technical Summary

Technical Problem

The prior art has semantic miscalling problems in voice interactions, making it difficult to accurately locate user intentions, especially when voice information has multiple meanings.

Method used

By determining the contextual information of the speech information and combining at least two semantic analysis results for statistical analysis or using a neural network model, the target analysis results are selected from them to accurately locate user intentions.

Benefits of technology

It effectively solves the problem of semantic miscalling in voice interaction, improves user satisfaction and the intelligence of voice processing, and ensures the smooth progress of human-computer interaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113744728B_ABST
    Figure CN113744728B_ABST
Patent Text Reader

Abstract

The present disclosure provides a speech processing method, device, equipment and storage medium, which relate to the field of artificial intelligence, and in particular to vehicle networking and smart cockpit technology. The specific implementation scheme is: determining the context information of the voice information input by the sender; obtaining at least two semantic analysis results of the voice information; and selecting a target analysis result from the at least two semantic analysis results according to the context information. According to the technology disclosed in the present disclosure, the user's intention can be accurately located.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, in particular to vehicle networking and smart cockpit technology, and specifically to a voice processing method, device, equipment and storage medium. Background Art

[0002] With the widespread popularity of artificial intelligence technology, human-computer interaction has been applied to various fields. At present, in the field of Internet of Vehicles, the main way for car owners to interact with the car computer is voice interaction. Therefore, accurate recognition of user intentions is crucial for voice interaction. Summary of the invention

[0003] The present disclosure provides a speech processing method, apparatus, device and storage medium.

[0004] According to one aspect of the present disclosure, there is provided a speech processing method, the method comprising:

[0005] Determining context information of the voice message input by the sender;

[0006] Obtaining at least two semantic analysis results of the speech information;

[0007] A target parsing result is selected from the at least two semantic parsing results according to the context information.

[0008] According to another aspect of the present disclosure, there is provided a speech processing device, the device comprising:

[0009] A context information determination module, used to determine the context information of the voice information input by the sender;

[0010] A parsing result acquisition module, used to acquire at least two semantic parsing results of the speech information;

[0011] A target result selection module is used to select a target parsing result from the at least two semantic parsing results according to the context information.

[0012] According to another aspect of the present disclosure, there is provided an electronic device, the electronic device comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the speech processing method described in any embodiment of the present disclosure.

[0016] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the speech processing method described in any embodiment of the present disclosure.

[0017] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the speech processing method described in any embodiment of the present disclosure is implemented.

[0018] According to the technology disclosed in the present invention, the user intention can be accurately located.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 is a flow chart of a speech processing method provided according to an embodiment of the present disclosure;

[0022] Figure 2 is a flow chart of another speech processing method provided according to an embodiment of the present disclosure;

[0023] Figure 3 is a flowchart of another speech processing method provided according to an embodiment of the present disclosure;

[0024] Figure 4 is a structural schematic diagram of a speech processing device provided according to an embodiment of the present disclosure;

[0025] Figure 5 is a block diagram of an electronic device for implementing the speech processing method of an embodiment of the present disclosure; DETAILED DESCRIPTION

[0026] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0027] Figure 1It is a flow chart of a voice processing method provided according to an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to the situation of how to process voice, and is particularly applicable to the situation of how to process user voice information in order to accurately locate user intentions when the user interacts with the vehicle terminal and the voice information input by the user to the vehicle terminal has multiple meanings (that is, the voice information does not clearly express the user's intention). This embodiment can be executed by a voice processing device configured in an electronic device, and the device can be implemented in software and / or hardware. Optionally, the electronic device can be a vehicle terminal, or a server terminal, etc. Figure 1 As shown, the speech processing method provided in this embodiment may include:

[0028] S101, determining context information of voice information input by a sender.

[0029] In this embodiment, the so-called sender is any party that has the need to interact with the vehicle terminal, such as the driver, co-driver or other users in the vehicle. The voice information is a service request sent by the sender to the vehicle terminal through voice; further, in this embodiment, the voice information input by the sender to the vehicle terminal can express multiple meanings, such as "Red Sun", which has two meanings, namely "play the song "Red Sun"" and "introduce the Encyclopedia Red Sun".

[0030] The context information is used to characterize the situation in which the sender inputs the voice information to the vehicle terminal; optionally, the context information includes all factors that affect the semantic analysis results. Furthermore, the context information may include relevant information about the vehicle terminal status, such as the interface identifier currently displayed on the vehicle terminal, where the interface identifier is used to uniquely represent a specific interface, which may be an interface ID; the context information may also include the language and tone used by the sender to express the voice information; in addition, the context information may also include the relative distance between the sender and the vehicle terminal when the sender expresses the voice information.

[0031] Optionally, if the execution subject of this embodiment is the vehicle terminal, the vehicle terminal can collect the voice information input by the sender through a voice collection module (such as a microphone) and determine the context information of the voice information. For example, the vehicle terminal can determine the context information of the voice information according to the state of the vehicle terminal; or, when collecting the voice information of the sender, the vehicle terminal simultaneously collects the scene image including the sender (for example, when the sender inputs the voice information to the vehicle terminal, the sender is in the vehicle, then the vehicle terminal can synchronously collect the image of the entire vehicle interior), and then the collected scene image can be identified and analyzed to determine the context information of the voice information, etc.

[0032] Furthermore, if the executor of this embodiment is the server, the vehicle computer can collect the voice information input by the sender through a voice collection module (such as a microphone), and simultaneously collect the scene image including the sender; thereafter, the vehicle computer can send the collected voice information and scene image to the server, and the server determines the context information of the voice information input by the sender based on the scene image.

[0033] It should be noted that, considering the network quality, storage space, and computing power of the vehicle-side and the server-side, the execution subject of this embodiment is preferably the server-side. Furthermore, when the hardware facilities and network quality of the vehicle-side meet the requirements, the execution subject of this embodiment can also be the vehicle-side. Exemplarily, when the execution subject is the server-side, considering that the server-side directly determines the context information, the vehicle-side needs to transmit data (such as scene images, etc.), which will occupy the network resources of the vehicle-side, so the server-side can obtain the context information of the voice information transmitted by the vehicle-side.

[0034] S102, obtaining at least two semantic analysis results of the speech information.

[0035] In this embodiment, the so-called semantic analysis result is obtained by performing speech recognition on the speech information to obtain the text content corresponding to the speech information, and performing semantic analysis on the text content.

[0036] Optionally, if the executor of this embodiment is the vehicle terminal, the vehicle terminal can distribute the voice information input by the sender to at least two semantic analysis modules, and each semantic analysis module analyzes the voice information respectively, and then the vehicle terminal can obtain the semantic analysis results of each semantic analysis module on the voice information. Alternatively, the vehicle terminal can send a semantic request to the server, and the server distributes the voice information to different semantic analysis modules, and then the vehicle terminal can obtain at least two semantic analysis results of the voice information fed back by the server. Among them, the semantic analysis module can be configured in the vehicle terminal, or can be configured in other devices independently of the vehicle terminal.

[0037] Furthermore, if the execution subject of this embodiment is a server, the server can send a semantic request including voice information to at least two semantic analysis modules, and each semantic analysis module analyzes the voice information respectively; and then the server can obtain the semantic analysis results of the voice information by each semantic analysis module. Among them, the semantic analysis module can be configured in the server, or can be configured in other devices independently of the server.

[0038] S103: Select a target parsing result from at least two semantic parsing results according to the context information.

[0039] In this embodiment, the target parsing result is a semantic parsing result that can express the sender's true intention.

[0040] In one possible implementation, this embodiment can perform statistical analysis on at least two semantic parsing results based on context information to further determine a target parsing result.

[0041] In another possible implementation, the target parsing result may be determined by a neural network model. For example, the context information and at least two semantic parsing results are input into a pre-trained matching model, and the matching model outputs the target parsing result.

[0042] It should be pointed out that when the same voice information has multiple expressions, the existing technology directly determines the user intention based on the analysis result of the voice information by the semantic analysis module, which leads to the problem of semantic false call. However, this embodiment, by combining the context information of the voice information, can accurately locate the target analysis result, that is, the user's true intention, from different semantic analysis results, thereby effectively solving the problem of semantic false call and improving user satisfaction.

[0043] The technical solution provided by the disclosed embodiment can accurately locate the target analysis result, that is, the user's true intention, from different semantic analysis results by combining the context information of the voice information when the same voice information has multiple expression meanings, thereby effectively solving the semantic miscall problem existing in the existing voice interaction process and providing a guarantee for the smooth progress of human-computer interaction. In addition, the solution also improves the intelligence level of voice processing.

[0044] On the basis of the above-mentioned embodiment, as an optional method of the embodiment of the present application, after selecting the target parsing result from at least two semantic parsing results according to the context information, the target parsing result can also be executed and the execution result can be output to the sender.

[0045] Specifically, if the execution subject of this embodiment is the vehicle computer, the vehicle computer can directly execute the target analysis result after determining the target analysis result, and can output the execution result to the sender. The execution result can be output in the form of voice broadcast, interface display, or a combination of voice broadcast and interface display.

[0046] For example, the voice message is "XX supermarket", the semantic analysis result A is "Introducing Encyclopedia XX supermarket", and the semantic analysis result B is "Search for nearby XX supermarkets on the map"; if the target analysis result is the semantic analysis result B, the vehicle computer will execute the semantic analysis result B and display the search results of nearby XX supermarkets on the map interface of the vehicle computer for the sender to view.

[0047] For example, the voice message is "turn on the air conditioner". In the smart home environment of car-home interconnection, "turn on the air conditioner" can correspond to semantic analysis result C "turn on the air conditioner at home" and semantic analysis result D "turn on the air conditioner in the car"; if the target analysis result is semantic analysis result D, the car computer will execute the semantic analysis result D and can broadcast fixed words to the sender, such as "the air conditioner is turned on".

[0048] Furthermore, if the execution subject of this embodiment is the server, then after determining the target analysis result, the server can feed back the target analysis result to the vehicle computer, and the vehicle computer executes the target analysis result and outputs the execution result to the sender.

[0049] It should be noted that, after accurately locating the user's intention, this embodiment can output the result after executing the user's intention to the user, further improving the solution.

[0050] Figure 2 1 is a flow chart of another speech processing method provided according to an embodiment of the present disclosure. Based on the above embodiment, this embodiment further explains in detail how to "determine the context information of the sender's input speech information". Figure 2 As shown, the speech processing method provided in this embodiment may include:

[0051] S201, determining the context information of the voice information input by the sender according to the interface status of the vehicle terminal.

[0052] In this embodiment, the vehicle-side interface status may include the vehicle-side interface display and operation status when the sender inputs voice information to the vehicle-side.

[0053] In one possible implementation, when the sender inputs voice information to the vehicle terminal, the interface being displayed on the vehicle terminal display screen can be used as the target interface, and the target interface identifier can be used as the context information of the voice information.

[0054] Alternatively, when the sender inputs voice information to the vehicle terminal, the interface being displayed on the vehicle terminal display screen and the interface running in the background at this time can be used as the target interface, and the target interface identifier can be used as the context information of the voice information.

[0055] Another possible implementation method can determine the context information of the voice information input by the sender based on the interface state of the vehicle terminal, the location information of the sender and / or the historical interaction record. Among them, the historical interaction record may include the interaction record between the vehicle terminal and the sender over a period of time; considering factors such as memory resource usage, the historical interaction record in this embodiment preferably includes the interaction record between the vehicle terminal and the sender based on the previous voice information, which may specifically include the previous voice information input by the sender to the vehicle terminal, and the content output by the vehicle terminal to the sender based on the analysis result of the previous voice information. The so-called previous voice information is the voice information input by the sender to the vehicle terminal before the voice information in S201 is input to the vehicle terminal.

[0056] For example, it can be determined whether the interface currently displayed on the vehicle-side display screen when the sender inputs voice information to the vehicle-side is the same as the interface displayed by the vehicle-side after the vehicle-side responds to the sender's previous voice information; if they are the same, the interface currently displayed on the vehicle-side display screen is used as the target interface; otherwise, the interface currently displayed on the vehicle-side display screen and the interface displayed by the vehicle-side after the vehicle-side responds to the sender's previous voice information are used as the target interfaces together; the target interface identifier is used as the context information of the voice information.

[0057] For example, when the sender inputs voice information to the vehicle terminal, the interface displayed on the vehicle terminal display screen can be used as the target interface; at the same time, the relative distance between the sender and the vehicle terminal is determined based on the sender's location information; thus, the target interface identifier and the relative distance can be used as context information.

[0058] Alternatively, it can be determined whether the interface currently displayed on the vehicle-side display screen when the sender inputs voice information to the vehicle-side terminal is the same as the interface displayed by the vehicle-side terminal after the vehicle-side terminal responds to the sender's previous voice information; if they are the same, the interface currently displayed on the vehicle-side display screen is used as the target interface; otherwise, the interface currently displayed on the vehicle-side display screen and the interface displayed by the vehicle-side terminal after the vehicle-side terminal responds to the sender's previous voice information are used as the target interfaces; at the same time, the relative distance between the sender and the vehicle-side terminal is determined based on the sender's location information; thereby, the target interface identifier and the relative distance can be used as context information together.

[0059] S202, obtaining at least two semantic analysis results of the speech information.

[0060] S203: Select a target parsing result from at least two semantic parsing results according to the context information.

[0061] The technical solution provided by the disclosed embodiment can determine the context information of the voice information input by the sender by combining the interface status of the vehicle terminal, providing an optional way to determine the context information, and at the same time providing data support for the subsequent combination of the context information of the voice information to accurately locate the target parsing result from different semantic parsing results. In addition, when determining the context information, this embodiment also combines other factors such as the sender's location information and / or historical interaction records, so that the context information is more accurate and comprehensive, thereby further improving the accuracy of the target parsing result finally determined.

[0062] Figure 3 is a flowchart of another speech processing method provided according to an embodiment of the present disclosure. Based on the above embodiment, this embodiment explains in detail how to "select a target parsing result from at least two semantic parsing results according to the context information" when the context information includes a target interface identifier. Figure 3 As shown, the speech processing method provided in this embodiment may include:

[0063] S301, determining context information of voice information input by a sender, wherein the context information includes a target interface identifier.

[0064] S302, obtaining at least two semantic analysis results of the speech information.

[0065] S303: Determine a target scene associated with the target interface according to the target interface identifier.

[0066] Optionally, this embodiment may search from a pre-established context mapping relationship table according to the target interface identifier to determine a target scene associated with the target interface.

[0067] In one possible implementation, this embodiment can construct a context mapping relationship table based on the services (or vertical categories) supported by the vehicle terminal, the functions (or intents) contained in each service, and the interface that the vehicle terminal can display to the user. Optionally, a scene in the context mapping relationship table can consist of a service and a function under the service. The services include but are not limited to maps, weather, music, phone calls, and games; for example, for maps, the map service can have multiple functions such as navigation, surrounding search, and along-the-way search.

[0068] Exemplarily, an interface in the context mapping table can be associated with one or more scenes. For example, as shown in Table 1, card_type represents service, and intent represents function. For the car map interface, you can associate the scene composed of map (map) and surrounding search (nearby), and you can also associate the scene composed of music (music) and play (play), etc. Furthermore, a scene can also be associated with one or more interfaces. For example, for the map search along the way scene, you can associate the car map interface, and you can also associate the car map navigation interface, etc.

[0069] Table 1 Context mapping relationship table

[0070] card_type intent illustrate interface map nearby Map nearby search Car map interface music Play Music playback Car map interface passing_point search Map search along the way Car map interface passing_point search Map search along the way Car map navigation interface map route navigation Car map navigation interface music play Music playback Music playback interface music collect Music albums Music playback interface

[0071] It should be noted that Table 1 simply lists some correspondences between scenes and interfaces for illustration, and the context mapping relationship table may also include correspondences between other interfaces and scenes, which is not limited in this embodiment.

[0072] In order to more comprehensively cover a variety of scenarios and more accurately locate any intention of the user, further, when constructing the context mapping relationship table, the historical interaction records between the user and the vehicle terminal can also be combined.

[0073] For example, the correspondence between scenes and interfaces in the context mapping table can also be dynamically adjusted according to the addition, deletion, and upgrade of vehicle-side service functions, new interaction records between users and the vehicle-side, etc. For example, a new correspondence between scenes and interfaces is added.

[0074] The execution subject of this embodiment is the server as an example. The server can use the target interface representation as an index and search from the pre-established context mapping relationship table to obtain the target scene associated with the target interface. The number of target scenes can be one or more, and preferably multiple in this embodiment.

[0075] S304: Select a target parsing result from the at least two semantic parsing results according to the matching result between the at least two semantic parsing results and the target scene.

[0076] Optionally, if the semantic parsing result obtained includes relevant fields such as card_type, intent, and data, the semantic parsing result can be directly matched with the target scene; if the semantic parsing result obtained is not composed of fields, the semantic parsing result can be parsed into a form including relevant fields such as card_type, intent, and data. Among them, data is used to represent the specific data obtained by the voice request. For example, if the semantic parsing result is "Search for nearby XX supermarkets on the map", it can be parsed into a field form including card_type = map, intent = surrounding search, data = XX supermarket.

[0077] Optionally, after determining the target scene associated with the target interface, each semantic parsing result may be matched with the target scene; if there is only one semantic parsing result that matches the target scene, the matching semantic parsing result is used as the target parsing result.

[0078] Exemplarily, if there are at least two semantic parsing results that match the target scene, the priority of the matching semantic parsing results is determined according to the priority of the target scene; and the target parsing result is determined from the matching semantic parsing results according to the priority of the matching semantic parsing results.

[0079] In this embodiment, one interface can be associated with multiple scenes, and each scene has a certain priority, which represents the order in which each scene is recalled. Optionally, the priority between scenes can be determined based on the order of various functions supported by the service, and the historical interaction records between the user and the vehicle terminal.

[0080] If there is only one target interface identifier, the target scenes obtained at this time are multiple scenes under one target interface. Optionally, one semantic parsing result matches at most one target scene. Furthermore, if there are at least two semantic parsing results that match the target scene, such as semantic parsing result A matches target scene a, and semantic parsing result B matches target scene b, then the priority of the target scene under the target interface can be used as the priority of the matching semantic parsing result, such as setting the priority "first" of target scene a as the priority of semantic parsing result A, and setting the priority "second" of target scene b as the priority of semantic parsing result B; and then according to the priority of the matching semantic parsing results, one is selected from the matching semantic parsing results as the target parsing result. For example, the semantic parsing result with the highest priority among the matching semantic parsing results is used as the target parsing result. Furthermore, if the priorities of the matching semantic parsing results are the same, each matching semantic parsing result can be scored based on the set scoring rules, and the semantic parsing result with the highest score can be used as the target parsing result.

[0081] For another example, if the target interface identifier includes two or more, the target scene obtained at this time is the scene under two or two target interfaces. Furthermore, if there are at least two semantic parsing results that match the target scene, and the target scenes that match the semantic parsing results correspond to different target interfaces, such as semantic parsing result C matches target scene c, and semantic parsing result D matches target scene d, where target scene c corresponds to target interface 1, and target scene d corresponds to target interface 2, then the priority of target scene c and target scene d can be determined first, and then the priority of target scene c can be used as the priority of semantic parsing result C, and the priority of target scene d can be used as the priority of semantic parsing result D; and then according to the priority of the matching semantic parsing results, one is selected from the matching semantic parsing results as the target parsing result. Among them, if the priority of target scene c under target interface 1 is determined based on the context mapping relationship table, and is the same as the priority of target scene d under target interface 2, then the priority of target scene c and target scene d can be determined according to the priority between target interface 1 and target interface 2. Optionally, in this embodiment, the priority between interfaces can be determined based on the priority between the service providing the interface (for example, the service providing the vehicle map interface is a map) and the service.

[0082] Furthermore, if the acquired target scene is a scene under two or two target interfaces, and the same scene exists under different target interfaces, at this time if a target scene matched by a semantic parsing result corresponds to multiple target interfaces, then based on the priority between interfaces, one target interface can be selected from the multiple target interfaces as the target interface corresponding to the target scene matched by the semantic parsing result.

[0083] Exemplarily, if there is no semantic parsing result matching the target scene, at least two semantic parsing results are scored; and according to the scoring result, a target parsing result is selected from the at least two semantic parsing results.

[0084] Specifically, if any semantic analysis result does not match the target scene, each semantic analysis result can be scored based on a pre-set scoring rule, and the semantic analysis result with the highest score can be used as the target analysis result. The scoring rule can be based on the source credibility of the semantic analysis result, or based on the semantic recall of the voice information in the historical interaction record, etc.

[0085] It should be noted that the present embodiment introduces the association relationship between the interface and the scene, and selects the semantic parsing results that match the scene from multiple semantic parsing results for recall, that is, the semantic parsing results that match the contextual information are recalled first, which can make the recalled semantic parsing results more in line with the user's intentions; at the same time, in the absence of semantic parsing results that match the contextual information, different semantic parsing results are scored and recalled based on the bottom-up strategy, which increases the flexibility and completeness of the solution.

[0086] The technical solution provided by the embodiments of the present disclosure uses interface identifiers to represent context information. Based on the interface identifiers, the scene can be quickly located, and based on the located scene, the target parsing result, that is, the user's true intention, can be accurately located from multiple semantic parsing results, thereby effectively solving the semantic miscall problem existing in the existing voice interaction process and improving the response efficiency.

[0087] Figure 4 : is a structural diagram of a speech processing device provided according to an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to the situation of how to process speech, and is particularly applicable to the situation of how to process user speech information in the scenario of user interaction with the vehicle terminal to accurately locate the user's intention. The device can be implemented by software and / or hardware, and the device can implement the speech processing method described in any embodiment of the present disclosure. Figure 4 As shown, the speech processing device comprises:

[0088] A context information determination module 401, used to determine the context information of the voice information input by the sender;

[0089] The analysis result acquisition module 402 is used to acquire at least two semantic analysis results of the speech information;

[0090] The target result selection module 403 is used to select a target parsing result from at least two semantic parsing results according to context information.

[0091] The technical solution provided by the disclosed embodiment can accurately locate the target analysis result, that is, the user's true intention, from different semantic analysis results by combining the context information of the voice information when the same voice information has multiple expression meanings, thereby effectively solving the semantic miscall problem existing in the existing voice interaction process and providing a guarantee for the smooth progress of human-computer interaction. In addition, the solution also improves the intelligence level of voice processing.

[0092] Exemplarily, the context information determination module 401 includes:

[0093] The context information determination unit is used to determine the context information of the voice information input by the sender according to the interface state of the vehicle terminal.

[0094] Exemplarily, the context information determination unit is specifically used for:

[0095] The context information of the voice information input by the sender is determined based on the vehicle terminal interface status, the sender's location information and / or historical interaction records.

[0096] Exemplarily, the context information includes a target interface identifier; the target result selection module 403 includes:

[0097] A target scene determination unit, used to determine a target scene associated with a target interface according to a target interface identifier;

[0098] The target result selection unit is used to select a target parsing result from at least two semantic parsing results according to a matching result between the at least two semantic parsing results and a target scene.

[0099] Exemplarily, the target result selection unit is specifically used for:

[0100] If there is a semantic parsing result that matches the target scene, the priority of the matching semantic parsing result is determined according to the priority of the target scene;

[0101] According to the priorities of the matched semantic parsing results, a target parsing result is determined from the matched semantic parsing results.

[0102] Exemplarily, the target result selection unit is specifically used for:

[0103] If there is no semantic parsing result matching the target scene, score at least two semantic parsing results;

[0104] According to the scoring results, a target parsing result is selected from at least two semantic parsing results.

[0105] Exemplarily, the above device further includes:

[0106] The execution module is used to execute the target parsing result and output the execution result to the sender.

[0107] In the technical solution disclosed in the present invention, the acquisition, storage and application of the sender's voice information and semantic analysis results, etc., all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0108] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0109] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0110] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0111] Multiple components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as a speech processing method. For example, in some embodiments, the speech processing method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the speech processing method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the speech processing method in any other appropriate manner (e.g., by means of firmware).

[0113] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0114] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0115] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0116] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0117] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0118] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0119] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0120] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A speech processing method, comprising: Determining context information of the voice message input by the sender; Obtain at least two semantic analysis results of the speech information; Selecting a target parsing result from the at least two semantic parsing results according to the context information; The context information includes a target interface identifier; and selecting a target parsing result from the at least two semantic parsing results according to the context information includes: Determining a target scene associated with the target interface from a pre-established context mapping relationship table according to the target interface identifier, wherein the context mapping relationship table is constructed by services supported by the vehicle terminal, functions included in each service, and interfaces that the vehicle terminal can display to the user; A target parsing result is selected from the at least two semantic parsing results according to a matching result between the at least two semantic parsing results and the target scene.

2. The method according to claim 1, wherein: The determining of the context information of the voice information input by the sender includes: According to the interface status of the vehicle terminal, the context information of the voice information input by the sender is determined.

3. The method according to claim 2, wherein: The determining, according to the vehicle terminal interface state, the context information of the voice information input by the sender includes: The context information of the voice information input by the sender is determined based on the vehicle terminal interface status, the sender's location information and / or historical interaction records.

4. The method according to claim 3, wherein: The selecting a target parsing result from the at least two semantic parsing results according to the matching result between the at least two semantic parsing results and the target scene comprises: If there is a semantic parsing result matching the target scene, determining the priority of the matching semantic parsing result according to the priority of the target scene; According to the priorities of the matched semantic parsing results, a target parsing result is determined from the matched semantic parsing results.

5. The method according to claim 3, wherein: The selecting a target parsing result from the at least two semantic parsing results according to the matching result between the at least two semantic parsing results and the target scene comprises: If there is no semantic parsing result matching the target scene, scoring the at least two semantic parsing results; According to the scoring result, a target parsing result is selected from the at least two semantic parsing results.

6. The method according to claim 1, after selecting a target parsing result from the at least two semantic parsing results according to the context information, further comprising: Execute the target parsing result and output the execution result to the sender.

7. A speech processing device, comprising: A context information determination module, used to determine the context information of the voice information input by the sender; A parsing result acquisition module, used to acquire at least two semantic parsing results of the speech information; A target result selection module, configured to select a target parsing result from the at least two semantic parsing results according to the context information; The context information includes a target interface identifier; the target result selection module includes: A target scene determination unit, configured to determine a target scene associated with a target interface from a pre-established context mapping relationship table according to the target interface identifier, wherein the context mapping relationship table is constructed by services supported by the vehicle terminal, functions included in each service, and interfaces that the vehicle terminal can display to the user; A target result selection unit is used to select a target parsing result from the at least two semantic parsing results according to a matching result between the at least two semantic parsing results and the target scene.

8. The device according to claim 7, wherein: The context information determination module includes: The context information determination unit is used to determine the context information of the voice information input by the sender according to the interface state of the vehicle terminal.

9. The device according to claim 8, wherein: The context information determination unit is specifically used for: The context information of the voice information input by the sender is determined based on the vehicle terminal interface status, the sender's location information and / or historical interaction records.

10. The device according to claim 7, wherein: The target result selection unit is specifically used for: If there is a semantic parsing result matching the target scene, determining the priority of the matching semantic parsing result according to the priority of the target scene; According to the priorities of the matched semantic parsing results, a target parsing result is determined from the matched semantic parsing results.

11. The device according to claim 7, wherein: The target result selection unit is specifically used for: If there is no semantic parsing result matching the target scene, scoring the at least two semantic parsing results; According to the scoring result, a target parsing result is selected from the at least two semantic parsing results.

12. The apparatus according to claim 7, further comprising: The execution module is used to execute the target parsing result and output the execution result to the sender.

13. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the speech processing method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable a computer to execute the speech processing method according to any one of claims 1 to 6.

15. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the speech processing method according to any one of claims 1 to 6.

Citation Information

Patent Citations

  • Vertical search engine system and method using semantic analysis

    CN102200975A

  • Intelligent service method and system based on natural language interaction

    CN104199810A

  • Speech recognition method and vehicle-mounted device

    CN104966514A

  • Dialogue system, vehicle having the same and dialogue processing method

    CN108346430A

  • Intention acquiring method and system combining with context

    CN109979462A