Voice instruction execution method, device and system

Abstract

The invention discloses a voice instruction execution method, device and system, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of converting a received voice instruction of a user into a text instruction, determining corresponding context information, further determining a target instruction according to the context information, and executing the voice instruction by adopting an execution method corresponding to the target instruction. According to the implementation mode, the corresponding voice instruction can be executed by converting the voice instruction of the user into the text instruction and further determining the execution method of the target instruction corresponding to the text instruction, corresponding instruction operation is executed based on the natural language of the user, and a more intelligent man-machine interaction mode is provided.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device and system for executing a voice command. Background Art

[0002] With the development of Internet technology, voice recognition technology has gradually matured, and it is possible to control smart devices to perform corresponding functions through voice commands. For example, for augmented reality (AR) devices, there are a lot of voice commands that can realize various functions of AR devices. In AR devices, voice commands are hidden in pages at different levels. Functional commands are usually configured at multiple levels. Different functions have different voice commands, and users need to enter the voice commands set by the AR device according to the required functions.

[0003] However, the above method requires the user to remember the multifunctional instructions of the AR device itself when using the AR device in order to realize different functions when using it. Moreover, for different AR devices, the set voice instructions are also different. For users, on the basis of memorizing the voice instructions of different functions, they also need to remember the instructions of different AR devices, which is relatively difficult and the user experience is poor. For example, in industrial maintenance scenarios, maintenance personnel can better perform machine maintenance operations through AR glasses. Due to the diversity of fault problems and maintenance operations, a large number of voice commands need to be set to realize different functions. For maintenance personnel, it is difficult to remember all the instructions, and the role of AR glasses cannot be fully utilized. Summary of the invention

[0004] In view of this, the embodiments of the present invention provide a method, device and system for executing voice commands, which can realize the intelligent application of voice commands and achieve corresponding functions.

[0005] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for executing a voice command is provided, comprising:

[0006] Receive user's voice command;

[0007] Convert voice commands into text commands through voice recognition;

[0008] Determine context information according to the text instruction through the instruction recognition model, and determine the target instruction corresponding to the text instruction based on the context information;

[0009] An execution method corresponding to the target instruction is determined based on preset standard instruction data, and the voice instruction is executed based on the execution method.

[0010] Optionally, the instruction recognition model is a model trained based on a large language model structure and instruction training data of different terminals.

[0011] Optionally, determining a target instruction corresponding to the text instruction based on context information by using an instruction recognition model includes:

[0012] Generate a first target instruction corresponding to the text instruction based on the context information through the instruction recognition model, and

[0013] Based on the synonym database, it is determined that the target instruction corresponding to the first target instruction is the target instruction corresponding to the text instruction.

[0014] Optionally, the target instruction includes a target instruction name and a target instruction format;

[0015] Before determining the execution method corresponding to the target instruction based on the preset standard instruction data, the method further includes:

[0016] According to the correspondence between the preset instruction name and the instruction format, a consistency check is performed on the target instruction name and the target instruction format, and it is determined that the consistency check passes.

[0017] Optionally, the method further comprises:

[0018] In response to the consistency check failing, determining a first instruction list in the preset standard instruction data based on the target instruction, wherein the first instruction in the first instruction list satisfies a similarity condition with the target instruction;

[0019] In response to receiving information that a user selects any first instruction in the first instruction list, determining an execution method corresponding to any first instruction as an execution method corresponding to the target instruction.

[0020] Optionally, the method is performed by a target application deployed on an augmented reality device;

[0021] Before receiving the user's voice command, the method further includes:

[0022] After the target application is started, the preset standard instruction data is initialized in the target application, and the system instruction of the terminal is closed.

[0023] According to another aspect of an embodiment of the present invention, there is provided a device for executing a voice command, comprising:

[0024] A receiving module, used for receiving a user's voice command;

[0025] A conversion module, used for converting voice commands into text commands through voice recognition;

[0026] A determination module, used to determine context information according to the text instruction through an instruction recognition model, and determine a target instruction corresponding to the text instruction based on the context information;

[0027] The execution module is used to determine an execution method corresponding to the target instruction based on preset standard instruction data, and execute the voice instruction based on the execution method.

[0028] According to another aspect of an embodiment of the present invention, a method for executing a voice command is provided, comprising:

[0029] The augmented reality device receives the user's voice command, converts the voice command into a text command through voice recognition, and sends the text command to the command recognition model;

[0030] The command recognition model obtains context information based on the text command parsing, generates a target command based on the context information, and sends the target command to the augmented reality device;

[0031] The augmented reality device determines an execution method corresponding to the target instruction based on the target instruction and preset standard instruction data, and executes the voice instruction based on the execution method.

[0032] According to another aspect of an embodiment of the present invention, a system for executing a voice command is provided, including an augmented reality device and a command recognition model:

[0033] The augmented reality device is used to receive a user's voice command, convert the voice command into a text command through voice recognition, and send the text command to the command recognition model; receive a target command sent by the command recognition model, determine an execution method corresponding to the target command based on preset standard command data, and execute the voice command based on the execution method;

[0034] The instruction recognition model is used to determine context information according to the text instruction, and determine the target instruction corresponding to the text instruction based on the context information, and send the target instruction to the augmented reality device.

[0035] An embodiment of the above invention has the following advantages or beneficial effects: it can obtain text instructions through conversion according to the voice instructions input by the user, and finally determine the target instruction by identifying the context information of the text instruction, and execute the voice instruction using the method corresponding to the target instruction, thereby realizing intelligent recognition of the corresponding operation instructions based on the voice content input by the user and realizing the corresponding functions, without the need for the user to remember the device-specific instruction names, thereby improving the user experience.

[0036] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0038] Figure 1 is a schematic diagram of the main steps of a method for executing a voice command according to an embodiment of the present invention;

[0039] Figure 2 A flow chart of voice command recognition for a user terminal;

[0040] Figure 3 is a schematic diagram of main modules of a device for executing a voice command according to an embodiment of the present invention;

[0041] Figure 4 is a schematic diagram of main steps of another method for executing a voice command according to an embodiment of the present invention;

[0042] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0043] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention 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 invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0045] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution disclosed in the present invention are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0046] It should be noted that in the technical solution of the present invention, the collection, use, storage, sharing and transfer of user personal information involved are in compliance with the provisions of relevant laws and regulations, and it is necessary to inform the user and obtain the user's consent or authorization. When applicable, the user's personal information is de-identified and / or anonymized and / or encrypted.

[0047] Figure 1 is a schematic diagram of the main process of the method for executing a voice command according to an embodiment of the present invention, such as Figure 1 As shown, including:

[0048] Step S101, receiving a user's voice command.

[0049] After the user device terminal (the device, terminal, and device terminal mentioned in the following description all refer to the user device terminal) is started, the user's voice command is received through the user device. The user device can be AR glasses or other devices that can execute voice commands, without limitation.

[0050] Optionally, before receiving the user's voice command, it is necessary to first receive a wake-up command for waking up the command receiving function of the user device. The wake-up command can be set by the user / user device and there is no restriction on this.

[0051] Step S102: converting the voice command into a text command through voice recognition.

[0052] The received voice command is converted into text content through speech recognition technology to obtain text commands. Among them, commonly used speech recognition technologies include methods based on linguistics and acoustics, random model methods, neural network methods, and probabilistic grammar analysis methods.

[0053] Step S103: determining context information according to the text instruction through the instruction recognition model, and determining a target instruction corresponding to the text instruction based on the context information.

[0054] Based on the recognition of the content of the text instruction, the context information is determined, and combined with the word vector data, the target instruction in the text instruction is determined. For example, the text instruction obtained is "I need remote expert help to modify the file", and the context information obtained is "need remote expert help". Based on this and combined with the word vector, the target instruction is further determined to be the "remote collaboration" instruction.

[0055] Optionally, the target instruction is determined according to the text instruction by, for example, pre-training an instruction recognition model according to the large language model. Specifically, the instruction recognition model is further trained based on the general large language model in combination with the instruction data and the voice instruction data.

[0056] Step S104, determining an execution method corresponding to the target instruction based on preset standard instruction data, and executing the voice instruction based on the execution method.

[0057] After obtaining the target instruction, the target instruction is queried according to the preset standard instruction data stored in the device, and after determining the target instruction, the corresponding execution method is found, and the voice instruction is executed in the device using the method. Among them, the preset standard instruction data stores the corresponding relationship between the operation instruction and the execution method, specifically the operation instruction name-instruction format-execution method. For example, as shown in Table 1, the operation instruction name is remote collaboration, the instruction is remote.click(), and the corresponding execution method is remote().

[0058]

[0059] According to the method for executing voice commands provided by the present invention, it is possible to obtain text commands based on the voice commands input by the user through conversion, and finally determine the target command by identifying the context information of the text command, and execute the voice command using the method corresponding to the target command, thereby realizing intelligent recognition of the corresponding operation command based on the voice content input by the user and realizing the corresponding function, without the need for the user to remember the device-specific command name, thereby improving the user experience.

[0060] Optionally, the instruction recognition model is a model trained based on a large language model structure and instruction training data of different terminals.

[0061] After determining the text instruction, the context information in the text instruction can be identified according to the instruction recognition model, wherein the instruction recognition model is a trained large model, through which the recognition is performed, and the corresponding output target instruction is determined based on the context information. Among them, the instruction training data of different terminals are, for example: the name of the operation instruction, the instruction format, the operation description or definition of each operation instruction, and the execution process description of each operation instruction obtained from the historical text data of different terminals. The instruction recognition model can output a model of the instruction in the specified format, for example, the output instruction is: remote collaboration remote.click().

[0062] Optionally, the target instruction may be determined by identifying a keyword in the context information, and the keyword determined by semantic recognition is the target instruction.

[0063] Optionally, determining a target instruction corresponding to a text instruction based on context information through an instruction recognition model includes: generating a first target instruction corresponding to the text instruction based on context information through an instruction recognition model, and determining, based on a synonym library, that the target instruction corresponding to the first target instruction is the target instruction corresponding to the text instruction.

[0064] The large model identifies the text instruction based on the context information to obtain the corresponding first target instruction, wherein the large model obtains keywords or synonyms through the reasoning ability of the large language model (LLM), and after obtaining the keywords or synonyms, determines the corresponding operation instruction name. Since the large model determines the instruction by identifying keywords and synonyms, the instruction is not necessarily the same as the instruction pre-stored in the device, and there may be similar situations, such as the first target instruction identified as remote assistance, while the stored instruction is remote collaboration. At this time, it is necessary to determine the target instruction corresponding to the first target instruction according to the stored synonym library, that is, the target instruction similar to the first target instruction. For example, I need help from a remote expert. Through identification and reasoning, it is concluded that he needs remote help. Remote help and remote collaboration have the same meaning, so the remote collaboration instruction is found.

[0065] Optionally, the target instruction includes a target instruction name and a target instruction format; before determining the execution method corresponding to the target instruction based on preset standard instruction data, the method also includes: performing a consistency check on the target instruction name and the target instruction format according to the correspondence between the preset instruction name and the instruction format, and determining that the consistency check passes.

[0066] Before executing the method corresponding to the determined target instruction, verification is required, including consistency verification of the target instruction name and the target instruction format. The verification can verify the target instruction according to the correspondence between the preset instruction name and the instruction format. When the two are consistent, the verification result is obtained. Among them, the verification is to prevent the model from generating wrong / mismatched operation instructions, causing the subsequent application to crash. Only when the verification passes will the corresponding execution method be determined according to the target instruction for execution; each preset instruction name has a unique corresponding instruction format. For example, the instruction name is search, and the corresponding instruction format is search(). Among them, regular verification can be used to perform the above verification process.

[0067] Optionally, the method also includes: in response to the consistency check failing, determining a first instruction list in the preset standard instruction data based on the target instruction, the first instruction in the first instruction list satisfies the similarity condition with the target instruction; in response to receiving information that the user selects any first instruction in the first instruction list, determining that the execution method corresponding to any first instruction is the execution method corresponding to the target instruction.

[0068] If the verification fails, it means that the instruction output by the model is illegal and needs to be re-determined. By selecting from the data of the preset standard instruction library, the similarity between the selected instruction and the instruction output by the model must meet the preset similarity condition. The similarity condition can specifically be that the similarity value of the two instructions is greater than a preset threshold, or the selected instruction is the instruction in the standard instruction library with the largest similarity value with the instruction output by the model. Determine that the selected instruction is the first instruction. The first instruction is output through a visual interface for the user to select. If information that the user selects the first instruction is received, the method corresponding to the first instruction is determined to be executed. If the user does not select the first instruction, the process ends.

[0069] Optionally, there may be multiple first instructions, and the multiple first instructions constitute a corresponding first instruction list for the user to select. After the user selects any instruction in the list, the method corresponding to the instruction is executed. Among them, the multiple first instructions can be arranged in order from large to small in similarity, which is more convenient for the user to select and improves the accuracy of matching.

[0070] According to the method for executing voice commands provided by the present invention, the voice commands can be accurately recognized through the pre-trained model, thereby ensuring the accuracy of voice command execution. At the same time, by verifying the commands output by the model, the accuracy of voice command recognition can be further guaranteed, while taking into account the advantages of improving user experience.

[0071] Optionally, the method is executed by a target application deployed on an augmented reality device; before receiving the user's voice command, the method also includes: after the target application is started, initializing preset standard command data in the target application, and turning off the terminal's system commands. In order to ensure the application of the above-mentioned command execution method by user devices, the recognition of voice commands can be realized through a self-developed target application, which can ensure that unified standard commands can be output for different user devices and implement corresponding functions. Specifically, after the user device is started, the target application is opened, and the preset standard command data is initialized in the target application, and the system commands of the user device are turned off, that is, the user device's own voice command system is no longer used.

[0072] Optionally, the target application mainly includes two parts: a large model and an instruction operation parser. The large model is used to obtain the corresponding target instruction according to the text instruction recognition, and the instruction operation parser is used to determine the corresponding execution method according to the target instruction. The instruction operation parser is connected to the large model through an API. In order to ensure the normal application of the instruction operation parser, it is necessary to set the output format of the large model so that the instruction operation parser can recognize it.

[0073] Optionally, after shutting down the system of the user device, the large model is connected to the instruction operation parser, and speech recognition can be started using the target application.

[0074] According to the voice command execution method provided in the embodiment of the present invention, it is possible to perform the voice command recognition process of different user devices through a unified target application, and output the target command in the same standard format, further determine the execution method corresponding to the target command, and execute the corresponding voice command, thereby realizing the use of the same voice command to operate different user devices, without the need for users to memorize voice operation commands, thereby greatly improving the user experience.

[0075] like Figure 2 The figure shows a flow chart of voice command recognition of a user terminal. Taking AR glasses as an example, after the user starts the AR device and opens the target application, the operation instructions in the target application are initialized. The AR glasses system instructions are closed, and then the instruction operation parser is connected to start the voice recognition function. After receiving the user's input voice through the target application, the voice recognition operation is performed, the voice is converted into text, and the obtained text is passed to the large model. The large model understands the context of the text and queries the corresponding operation instructions based on the word vector, and inputs the operation instructions into the instruction operation parser to obtain the execution method, and finally completes the instruction execution.

[0076] Figure 3 is a schematic diagram of a voice command execution device 300 provided in an embodiment of the present invention, such as Figure 3 As shown, including:

[0077] Receiving module 301, used to receive a user's voice command;

[0078] A conversion module 302, used to convert the voice command into a text command through voice recognition;

[0079] A determination module 303 is used to determine context information according to the text instruction through an instruction recognition model, and determine a target instruction corresponding to the text instruction based on the context information;

[0080] The execution module 304 is used to determine an execution method corresponding to the target instruction based on preset standard instruction data, and execute the voice instruction based on the execution method.

[0081] The voice command execution device provided by the present invention can obtain text commands after conversion based on the voice commands input by the user, and finally determine the target command by identifying the context information of the text command, and execute the voice command using the method corresponding to the target command, thereby realizing intelligent recognition of the corresponding operation command based on the voice content input by the user and realizing the corresponding functions, without the need for the user to remember the device-specific command names, thereby improving the user experience.

[0082] Optionally, the instruction recognition model is a model trained based on a large language model structure and instruction training data of different terminals.

[0083] Optionally, the determination module 303 is further configured to:

[0084] Generate a first target instruction corresponding to the text instruction based on the context information through the instruction recognition model, and

[0085] Based on the synonym database, it is determined that the target instruction corresponding to the first target instruction is the target instruction corresponding to the text instruction.

[0086] Optionally, the target instruction includes a target instruction name and a target instruction format; the device further includes:

[0087] The verification module 305 is used to perform consistency verification on the target instruction name and the target instruction format according to the correspondence between the preset instruction name and the instruction format, and determine whether the consistency verification passes.

[0088] Optionally, the device further comprises:

[0089] The determination module 303 is also used to determine, in response to a consistency check failing, a first instruction list in the preset standard instruction data based on the target instruction, wherein the first instruction in the first instruction list satisfies a similarity condition with the target instruction; in response to receiving information that a user selects any first instruction in the first instruction list, determine that the execution method corresponding to any first instruction is the execution method corresponding to the target instruction.

[0090] Optionally, the apparatus is deployed in a target application on an augmented reality device, and the apparatus further includes: an initialization module 306, which is used to initialize preset standard instruction data in the target application after the target application is started, and to close the system instructions of the terminal.

[0091] The voice command execution device provided in the embodiment of the present invention can perform the voice command recognition process of different user devices through a unified target application, and output the target command in the same standard format, further determine the execution method corresponding to the target command, and execute the corresponding voice command, thereby realizing the use of the same voice command to operate different user devices, without the user having to memorize the voice operation command, thereby greatly improving the user experience.

[0092] Figure 4 is a schematic diagram of the main process of the method for executing a voice command according to an embodiment of the present invention, such as Figure 4 As shown, including:

[0093] Step S401: The augmented reality device receives the user's voice command, converts the voice command into a text command through voice recognition, and sends the text command to the command recognition model. The augmented reality device receives the voice command input by the user, and further converts the voice command into text content to obtain a text command. The obtained text command is sent to the command recognition model so that the command recognition model performs subsequent parsing operations.

[0094] Step S402, the instruction recognition model obtains context information based on the text instruction parsing, generates a target instruction based on the context information, and sends the target instruction to the augmented reality device. The trained instruction recognition model determines the context information based on the recognition of the content of the text instruction, and determines the target instruction in the text instruction in combination with the word vector data. For example, the text instruction obtained is "I need remote expert help to modify the file", and the context information obtained by recognition is "remote expert help is needed". Based on this and the word vector, it is further determined that the target instruction is a "remote collaboration" instruction.

[0095] Step S403, the augmented reality device determines the execution method corresponding to the target instruction based on the target instruction and the preset standard instruction data, and executes the voice instruction based on the execution method. After obtaining the target instruction, the target instruction is queried according to the preset standard instruction data stored in the device, and after determining the target instruction, the execution method corresponding to it is found, and the method is used in the augmented reality device to execute the voice instruction.

[0096] According to the method for executing voice commands provided in an embodiment of the present invention, it is possible to obtain text commands based on the voice commands input by the user through conversion, and finally determine the target command by identifying the context information of the text command, and execute the voice command using the method corresponding to the target command, thereby realizing intelligent recognition of corresponding operation commands based on the voice content input by the user and realizing corresponding functions, without the need for the user to remember device-specific command names, thereby improving the user experience.

[0097] Optionally, the target instruction includes multiple instructions, and before determining the execution method corresponding to the target instruction based on the target instruction and the preset standard instruction data, the method further includes: the augmented reality device receives the user's selection information based on the multiple instructions; the augmented reality device determines the execution method corresponding to the target instruction based on the target instruction and the preset standard instruction data, including: the augmented reality device determines a second instruction from the target instruction based on the selection information; the augmented reality device determines the execution method corresponding to the second instruction based on the second instruction and the preset standard instruction data. The instruction recognition model can obtain the most similar multiple instructions based on text content recognition, and feed them back to the user for the user to select, match the determined second instruction with the instructions in the preset standard instruction data, and determine that the execution method corresponding to the matched instruction is the execution method corresponding to the user input voice instruction.

[0098] Optionally, the instruction recognition model is a model trained based on a large language model structure and instruction training data of different terminals.

[0099] Optionally, the instruction recognition model determines the target instruction corresponding to the text instruction based on context information, including: the instruction recognition model generates a first target instruction corresponding to the text instruction based on the context information, and determines, based on a synonym library, that the target instruction corresponding to the first target instruction is the target instruction corresponding to the text instruction.

[0100] The large model identifies the text instruction based on the context information to obtain the corresponding first target instruction, wherein the large model obtains keywords or synonyms through the reasoning ability of the large language model (LLM), and after obtaining the keywords or synonyms, determines the corresponding operation instruction name. Since the large model determines the instruction by identifying keywords and synonyms, the instruction is not necessarily the same as the instruction pre-stored in the device, and there may be similar situations, such as the first target instruction identified is remote assistance, while the stored instruction is remote collaboration. At this time, it is necessary to determine the target instruction corresponding to the first target instruction based on the stored synonym library.

[0101] Optionally, the target instruction includes a target instruction name and a target instruction format; before the augmented reality device determines the execution method corresponding to the target instruction based on the preset standard instruction data, the method also includes: the augmented reality device performs a consistency check on the target instruction name and the target instruction format according to the correspondence between the preset instruction name and the instruction format, and determines that the consistency check passes. Before executing the method corresponding to the determined target instruction, a check is required, including a consistency check on the target instruction name and the target instruction format. The check can check the target instruction according to the correspondence between the preset instruction name and the instruction format, and when the two are consistent, a check result of passing is obtained.

[0102] Optionally, the method also includes: in response to the consistency check failing, the augmented reality device determines a first instruction list in the preset standard instruction data based on the target instruction, and the first instruction in the first instruction list satisfies the similarity condition with the target instruction; in response to receiving information that the user selects any first instruction in the first instruction list, the augmented reality device determines that the execution method corresponding to any first instruction is the execution method corresponding to the target instruction. If the check fails, it means that the instruction output by the model is illegal and needs to be re-determined. By selecting from the data of the preset standard instruction library, it is necessary to meet the similarity between the selected instruction and the instruction output by the model to meet the preset similarity condition. When the information that the user selects the first instruction is received, the method corresponding to the first instruction is determined to be executed.

[0103] Optionally, before the augmented reality device receives the user's voice command, the method further includes: after the target application is started, the augmented reality device initializes preset standard command data in the target application and closes the terminal's system command. After the user device is started, the target application is opened, the preset standard command data is initialized in the target application, and the system command of the user device is closed, that is, the user device's own voice command system is no longer used.

[0104] According to the voice command execution method provided in the embodiment of the present invention, it is possible to perform the voice command recognition process of different user devices through a unified target application, and output the target command in the same standard format, further determine the execution method corresponding to the target command, and execute the corresponding voice command, thereby realizing the use of the same voice command to operate different user devices, without the need for users to memorize voice operation commands, thereby greatly improving the user experience.

[0105] The present invention provides a voice command execution system, including an augmented reality device and a command recognition model:

[0106] The augmented reality device is used to receive a user's voice command, convert the voice command into a text command through voice recognition, and send the text command to the command recognition model; receive a target command sent by the command recognition model, determine an execution method corresponding to the target command based on preset standard command data, and execute the voice command based on the execution method;

[0107] The instruction recognition model is used to determine context information according to the text instruction, and determine the target instruction corresponding to the text instruction based on the context information, and send the target instruction to the augmented reality device.

[0108] Figure 5 An exemplary system architecture 500 is shown to which the method for executing a voice command or the apparatus for executing a voice command according to the embodiment of the present invention can be applied.

[0109] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0110] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0111] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0112] The server 505 may be a server that provides various services, such as a background management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 501, 502, and 503. The background management server may analyze and process the received data such as the identification request, and feed back the processing result (such as the execution method - only an example) to the terminal device.

[0113] It should be noted that the method for executing the voice command provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the execution device for the voice command is generally disposed in the server 505 .

[0114] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0115] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device or a server suitable for implementing an embodiment of the present invention. Figure 6 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0116] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0117] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0118] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0119] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, 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 above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0120] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0121] The units or modules involved in the embodiments of the present invention may be implemented in software or hardware. The units or modules described may also be provided in a processor. For example, they may be described as follows: a processor includes a receiving module, a conversion module, a determination module, and an execution module. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves. For example, the receiving module may also be described as a "module for receiving a user's voice command."

[0122] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0123] Receive user's voice command;

[0124] Convert voice commands into text commands through voice recognition;

[0125] Determine context information according to the text instruction through the instruction recognition model, and determine the target instruction corresponding to the text instruction based on the context information;

[0126] An execution method corresponding to the target instruction is determined based on preset standard instruction data, and the voice instruction is executed based on the execution method.

[0127] According to the technical solution of the embodiment of the present invention, it is possible to obtain text instructions through conversion based on the voice instructions input by the user, and finally determine the target instruction by identifying the context information of the text instruction, and execute the voice instruction using the method corresponding to the target instruction, thereby realizing intelligent recognition of the corresponding operation instructions based on the voice content input by the user and realizing the corresponding functions, without the need for the user to remember the device-specific instruction names, thereby improving the user experience.

[0128] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for executing a voice command, characterized in that: include: Receive user's voice command; Converting the voice command into text command through voice recognition; Determining context information according to the text instruction through an instruction recognition model, and determining a target instruction corresponding to the text instruction based on the context information; An execution method corresponding to the target instruction is determined based on preset standard instruction data, and the voice instruction is executed based on the execution method.

2. The method according to claim 1, characterized in that The instruction recognition model is a model obtained by training based on a large language model structure and instruction training data of different terminals.

3. The method according to claim 2, characterized in that The determining, by means of an instruction recognition model, a target instruction corresponding to the text instruction based on the context information includes: Generate a first target instruction corresponding to the text instruction based on the context information through the instruction recognition model, and Based on the synonym library, it is determined that the target instruction corresponding to the first target instruction is the target instruction corresponding to the text instruction.

4. The method according to any one of claims 1 to 3, characterized in that: The target instruction includes a target instruction name and a target instruction format; Before determining the execution method corresponding to the target instruction based on the preset standard instruction data, the method further includes: According to the correspondence between the preset instruction name and the instruction format, a consistency check is performed on the target instruction name and the target instruction format, and it is determined that the consistency check passes.

5. The method according to claim 4, characterized in that The method further comprises: In response to the consistency check failing, determining a first instruction list in the preset standard instruction data based on the target instruction, wherein a first instruction in the first instruction list satisfies a similarity condition with the target instruction; In response to receiving information that a user selects any first instruction in the first instruction list, determining that an execution method corresponding to any first instruction is an execution method corresponding to the target instruction.

6. The method according to claim 1, characterized in that The method is executed by a target application deployed on an augmented reality device; Before receiving the user's voice command, the method further includes: After the target application is started, preset standard instruction data is initialized in the target application, and the system instruction of the terminal is closed.

7. A device for executing a voice command, characterized in that: include: A receiving module, used for receiving a user's voice command; A conversion module, used for converting the voice command into a text command through voice recognition; A determination module, configured to determine context information according to the text instruction through an instruction recognition model, and determine a target instruction corresponding to the text instruction based on the context information; An execution module is used to determine an execution method corresponding to the target instruction based on preset standard instruction data, and execute the voice instruction based on the execution method.

8. A method for executing a voice command, characterized in that: include: The augmented reality device receives the user's voice command, converts the voice command into a text command through voice recognition, and sends the text command to the command recognition model; The instruction recognition model obtains context information based on the text instruction parsing, generates a target instruction based on the context information, and sends the target instruction to the augmented reality device; The augmented reality device determines an execution method corresponding to the target instruction based on the target instruction and preset standard instruction data, and executes the voice instruction based on the execution method.

9. The method according to claim 8, characterized in that The target instruction includes a plurality of instructions. Before determining the execution method corresponding to the target instruction based on the target instruction and preset standard instruction data, the method further includes: The augmented reality device receives selection information of the user based on the multiple instructions; The augmented reality device determines an execution method corresponding to the target instruction based on the target instruction and preset standard instruction data, including: The augmented reality device determines a second instruction from the target instruction based on the selection information; The augmented reality device determines an execution method corresponding to the second instruction based on the second instruction and the preset standard instruction data.

10. A voice command execution system, comprising an augmented reality device and a command recognition model: The augmented reality device is used to receive a user's voice command, convert the voice command into a text command through voice recognition, and send the text command to a command recognition model; receive a target command sent by the command recognition model, determine an execution method corresponding to the target command based on preset standard command data, and execute the voice command based on the execution method; The instruction recognition model is used to determine context information according to the text instruction, and determine the target instruction corresponding to the text instruction based on the context information, and send the target instruction to the augmented reality device.

Images