Voice interaction method, device and vehicle

Abstract

This application provides a voice interaction method, device, and vehicle. The method includes: acquiring a first voice command; and, if the first voice command is not a command in a set of wake-up-free voice commands, updating the set of wake-up-free voice commands based on the usage frequency and a first recognition frequency of the first voice command, where the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state. This application can be applied to intelligent vehicles or electric vehicles, dynamically updating the voice command set according to user habits, reducing the possibility of the first voice command being misrecognized, thus improving the user's interaction experience with the vehicle and contributing to a safer driving environment.

Description

Voice interaction methods, devices and vehicles Technical Field

[0001] This application relates to the field of intelligent vehicles, and more specifically, to a voice interaction method, device, and vehicle. Background Technology

[0002] Wake-up-free operation, an important feature of smart cars, aims to allow users to operate the vehicle directly without using a wake-up word, improving interaction efficiency. However, wake-up-free voice commands are often relatively fixed, which can negatively impact the user experience.

[0003] How to provide more personalized wake-up-free functionality based on users' personal habits and needs is an urgent problem to be solved. Summary of the Invention

[0004] This application provides a voice interaction method, device, and vehicle. It facilitates the dynamic updating of the wake-up-free voice command set, enabling the set to be applicable to different users, thereby achieving efficient interaction between the user and the vehicle and improving the user experience.

[0005] In a first aspect, a voice interaction method is provided, the method comprising: acquiring a first voice command; and, if the first voice command is not a command in a set of wake-up-free voice commands, updating the set of wake-up-free voice commands based on the usage frequency of the first voice command and a first recognition frequency, wherein the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state.

[0006] In one possible implementation, the first voice command can be the user's voice input acquired while the vehicle is in a wake-up state. In this case, the user's voice input can be a command from the wake-up-free voice command set, or it can be a command not from the wake-up-free voice command set.

[0007] For example, the wake word is "Xiao A Xiao A". The user inputs the voice "Xiao A Xiao A", and then inputs the voice "Help me check the weather and road conditions from point A to point B". Then "Help me check the weather and road conditions from point A to point B" is the first voice command obtained.

[0008] The purpose of counting usage frequency is to assess whether the first voice command is a frequently used command by the user, and the purpose of counting the first recognition frequency is to assess whether the first voice command is a command that is easily misrecognized.

[0009] Based on the above technical solution, by simultaneously updating the wake-up-free voice command set according to the usage frequency and recognition frequency of the first voice command, the vehicle can respond quickly to frequently used user commands, and the misrecognition of the first voice command is reduced, thereby lowering the risk of the vehicle being accidentally woken up. This allows the wake-up-free voice command set to better meet the personalized needs of different users.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, the wake-up-free voice command set is updated based on the usage frequency and the first recognition frequency of the first voice command, including: when the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold, the first voice command is set as a command in the wake-up-free voice command set.

[0011] For example, a user used the "wipe the window" command a total of 20 times in the past month. Of these, 15 were recognized when the vehicle was awake, and 5 were recognized when the vehicle was not awake. Therefore, the usage frequency of this command is 20 times / month, and the first recognition frequency is 5 times / month. If the first preset threshold is 15 times / month and the second preset threshold is 8 times / month, then "wipe the window" can be added to the set of wake-free voice commands; that is, a new wake-free voice command, "wipe the window," is added.

[0012] Based on the above technical solution, frequently used commands can be added according to the usage frequency and recognition frequency of the first voice command. At the same time, the added commands can have higher recognition accuracy, thereby reducing the risk of false wake-up and improving the user experience.

[0013] In conjunction with the first aspect, in some implementations of the first aspect, updating the wake-up-free voice command set based on the usage frequency and the first recognition frequency of the first voice command includes: when the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold, the control prompting device prompts the user to add the first voice command to the wake-up-free voice command set; in response to obtaining input from the user confirming the addition of the first voice command to the wake-up-free voice command set, the first voice command is set as a command in the wake-up-free voice command set.

[0014] In one possible implementation, the vehicle issues a voice prompt to the user: "Do you want to set [Wipe the glass] as a wake-up-free voice command? Answer 'Confirm' to set it, answer 'Cancel' to stop the prompt." In response to the user's voice input "Confirm," "Wipe the glass" is set as a command in the wake-up-free voice command set.

[0015] In one possible implementation, the vehicle issues a voice command to the user: "Do you want to set [Navigate to the usual place] as a wake-up-free voice command? Answer 'Confirm' to set it, answer 'Cancel' to stop the prompt." In response to the user's voice input "Cancel," "Navigate to the usual place" is set as a command in the set of non-wake-up-free voice commands. This set of non-wake-up-free voice commands may include wake-up-free voice commands that the user does not wish to use. This set of non-wake-up-free voice commands can also be referred to as a third set of voice commands.

[0016] Based on the above technical solutions, in the process of adding new wake-up-free voice commands, the update of the wake-up-free voice command set can be made more targeted and flexible by taking into account the frequency of use and the first recognition frequency, while also taking into account the user's actual wishes. This will better match the user's usage preferences and further improve the user experience.

[0017] In conjunction with the first aspect, in some implementations of the first aspect, the wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. Setting the first voice command as an instruction in the wake-free voice command set includes: setting the first voice command as an instruction in the second voice command set.

[0018] In one possible implementation, the first set of voice commands may include pre-set wake-up-free voice commands in the vehicle, and the second set of voice commands may include wake-up-free voice commands determined by the user.

[0019] For example, the first set of voice commands can also be referred to as the "system preset list", and the second set of voice commands can also be referred to as the "whitelist".

[0020] In one possible implementation, the first voice command set and the second voice command set can be two mutually exclusive sets; that is, the first voice command set and the second voice command set include different wake-up-free voice commands. In this case, the commands in the second voice command set can be regarded as wake-up-free voice commands added by the user based on the vehicle's preset commands.

[0021] In one possible implementation, the first and second voice command sets may also include the same wake-up-free voice commands. In this case, "same command" indicates that the user has selected some voice commands from the vehicle's preset wake-up-free voice command set as the wake-up-free voice commands the user wishes to use.

[0022] For example, the wake-up-free voice command "heat seat" in the first voice command set may cause false wake-ups, but if the user chooses to add the command to the second voice command set, it means that even if the vehicle's preset command "heat seat" may cause false wake-ups, the user can accept this situation and choose to use the command.

[0023] It should be understood that the reason why users can accept the wake-up-free voice commands in the second set of voice commands that may cause false wake-ups is because these commands can provide great convenience in the user's daily driving. Therefore, the occasional false wake-up is not enough to weaken the practicality of the wake-up-free command and will not cause serious problems or inconvenience.

[0024] Based on the above technical solution, even when the first voice command set does not support user modification, it can help meet the user's need to add new commands and flexibly realize the dynamic expansion of the wake-up-free voice command set.

[0025] In conjunction with the first aspect, in some implementations of the first aspect, before setting the first voice command as an instruction in the second set of voice commands, the method further includes: deleting the first voice command from the set of non-wake-up-free voice commands.

[0026] As mentioned earlier, the non-wake-free voice command set can also be referred to as the third voice command set. The third voice command set can include wake-free voice commands that the user does not wish to use. For example, the third voice command set can also be referred to as a "blacklist." For instance, the third voice command set includes the command "wipe the glass," which means that the command "wipe the glass" cannot be used for the vehicle's wake-free function.

[0027] Clearly, the third voice command set and the second voice command set are two mutually exclusive sets. Therefore, before adding a first voice command to the second voice command set, if the first voice command is already in the third voice command set, the conflict between the two should be resolved first. For example, if the command "Turn the air conditioner to the highest setting" is in the second voice command set, and it is necessary to set "Turn the air conditioner to the highest setting" as a command in the third voice command set, the command "Turn the air conditioner to the highest setting" should first be deleted from the second voice command set.

[0028] Based on the above technical solution, the conflict between the second and third voice command sets can be effectively resolved when new wake-up-free voice commands are added, thus improving the standardization of wake-up-free voice command updates.

[0029] In one possible implementation, before setting the first voice command as an instruction in the second set of voice commands, the method further includes: controlling the prompting device to prompt the user to delete the first voice command from the third set of voice commands; and in response to receiving input from the user that the first voice command has been deleted from the third set of voice commands, setting the first voice command as an instruction in the second set of voice commands.

[0030] For example, the vehicle issues a voice command to the user: "[Wipe the glass] is currently not a wake-up-free voice command supported by the vehicle. Please remove this command from the blacklist before you can add it to the whitelist. Confirm to remove [Wipe the glass] from the blacklist?" In response to the user's voice input "Confirm", "Wipe the glass" is set as a command in the second set of voice commands.

[0031] In this implementation, by prompting the user to resolve the conflict between the second and third voice command sets, the user can gain a clearer understanding of the need to update the wake-up-free voice command set, thus improving the user experience.

[0032] In conjunction with the first aspect, in some implementations of the first aspect, before updating the wake-free voice command set based on the usage frequency and the first recognition frequency of the first voice command, the method further includes: sending the first voice command and the text recognition result of the first voice command to the cloud; receiving a verification result sent by the cloud, the verification result being used to indicate whether the text recognition result is accurate and whether the first voice command is an instruction issued by the user to the vehicle; updating the wake-free voice command set based on the usage frequency and the first recognition frequency of the first voice command, including: obtaining a second recognition frequency according to the first recognition frequency and the verification result; and updating the wake-free voice command set based on the usage frequency and the second recognition frequency.

[0033] In one possible implementation, the verification result can be obtained based on a large cloud-based automatic speech recognition (ASR) model.

[0034] For example, the cloud-based ASR model can be used to determine the accuracy of text recognition results on the vehicle. For instance, within a month, with the vehicle in a non-wake-up state, the command "go pick up your child from school" was successfully decoded 18 times, meaning the first recognition frequency is 18 times / month. Of these, 10 times the cloud-based text recognition results differed from the vehicle-based text recognition results, indicating erroneous recognition in these 10 instances, meaning the second recognition frequency is 10 times / month.

[0035] For example, the cloud-based ASR model can also be used to determine whether the sound is from a human-to-human interaction, a human-to-vehicle interaction, or an electronic device playing a sound. For instance, within a week, when the vehicle was in a non-wake-up state, the command "Activate Joy Mode" was successfully decoded 7 times. Of these, 2 were from electronic devices, 2 were from conversations between passengers, and 3 were from users attempting to give instructions to the vehicle. Therefore, "Activate Joy Mode" was incorrectly identified 4 times, meaning the second recognition frequency was 4 times per week.

[0036] In one possible implementation, the ASR big model may include a first model and a second model. The first model is used to determine whether the text recognition result on the vehicle is accurate, and the second model is used to determine whether the sound scenario is a human-to-human interaction scenario, a human-to-vehicle interaction scenario, or a scenario where an electronic device is playing sound. For example, the first model may be a speech recognition big model, and the second model may be a sound big model.

[0037] Based on the above technical solution and combined with cloud verification results, the frequency of incorrect recognition of the first voice command can be accurately obtained.

[0038] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: when the first voice command is an instruction in the set of wake-up-free voice commands, updating the set of wake-up-free voice commands based on the cancellation frequency of the first voice command.

[0039] The purpose of calculating the cancellation frequency is to assess whether the commands in the wake-up-free voice command set are those that frequently cause the voice assistant to be mistakenly activated. The cancellation frequency can be defined as the frequency with which the user cancels the operation corresponding to a command by voice or gesture within a certain period of time after the vehicle begins to execute that operation.

[0040] For example, a user can cancel a wake-up-free voice command A by using voice command B. Voice command B can be "cancel command A" or a command that is the opposite of the action corresponding to command A.

[0041] For example, in the first instant, in response to the voice input "open the window," the vehicle begins to open the windows. Three seconds later, in the second instant, the user issues the command "close the window" or "cancel opening the sunroof," and the vehicle begins to close the front and rear windows. The command "close the window" or "cancel opening the sunroof" can be considered a cancellation of the command "open the window."

[0042] For example, a user can cancel the wake-up-free voice command A by taking action C, where action C can be the opposite of the action corresponding to command A.

[0043] For example, at the first moment, in response to the user's voice input "turn on the air conditioning," the vehicle's air conditioning is turned on. Five seconds later, at the second moment, the user manually turns off the air conditioning. The user's action of turning off the air conditioning can be considered a cancellation of the "turn on the air conditioning" command.

[0044] In one possible implementation, the total number of cancellations can be used as the cancellation frequency of the first voice command.

[0045] In one possible implementation, the total number of wake-ups and total number of cancellations of the first voice command can be counted, and the ratio of the total number of cancellations to the total number of wake-ups can be used as the cancellation frequency of the first voice command.

[0046] In one possible implementation, the revocation frequency can also be calculated using a time decay mechanism.

[0047] Based on the above technical solution, the cancellation frequency can be used to determine the wake-up-free voice commands that users do not want to use or that are easily misrecognized, thereby updating the wake-up-free voice command set according to the user's usage habits.

[0048] In conjunction with the first aspect, in some implementations of the first aspect, the wake-up-free voice command set is updated based on the cancellation frequency of the first voice command, including: deleting the first voice command from the wake-up-free voice command set when the cancellation frequency is greater than or equal to a third preset threshold.

[0049] If the cancellation frequency of the first voice command is greater than or equal to the third preset threshold, it means that the first voice command is frequently cancelled during use, indicating that the vehicle may frequently perform operations that the user does not need. Therefore, the cancellation frequency of the first voice command can reflect the probability of false wake-up caused by the first voice command to a certain extent. The higher the cancellation frequency, the higher the probability of false wake-up caused by the first voice command, which may seriously affect the user experience. If the first voice command continues to be used as a command in the set of wake-up-free voice commands, the user may have a poor experience due to frequent false wake-ups, and may even face more safety hazards during driving. Therefore, the first voice command can be removed from the set of wake-up-free voice commands.

[0050] Based on the above technical solution, commands that users do not want to use or commands that are easily misrecognized can be deleted according to the cancellation frequency, so that the set of wake-up-free voice commands can better adapt to the user's needs, and can also reduce the situation where the first voice command is misrecognized, thereby reducing the risk of the vehicle being accidentally woken up.

[0051] In conjunction with the first aspect, in some implementations of the first aspect, updating the wake-free voice command set based on the cancellation frequency of the first voice command includes: when the cancellation frequency is greater than or equal to a third preset threshold, the control prompting device prompts the user to delete the first voice command; in response to obtaining input from the user confirming the deletion of the first voice command, the first voice command is deleted from the wake-free voice command set.

[0052] In one possible implementation, the vehicle issues a voice prompt to the user: "Do you want to remove '[Open the fragrance]' from the set of wake-up-free voice commands? Answer 'confirm' to remove it, answer 'cancel' to stop the prompt." In response to the user's voice input "confirm", "Open the fragrance" is removed from the set of wake-up-free voice commands.

[0053] In one possible implementation, the vehicle issues a voice command to the user: "Do you want to remove '[Play music your parents like]' from the wake-up-free voice command set? Answer 'confirm' to remove it, answer 'cancel' to stop the prompt." In response to the user's voice input "cancel", "Navigate to the usual place" is set as a command in the second voice command set.

[0054] Based on the above technical solutions, in the process of deleting wake-up-free voice commands, the update of the wake-up-free voice command set can be made more accurate and flexible by taking into account the frequency of revocation and the user's actual wishes, thereby better adapting to the user's usage preferences and further improving the user experience.

[0055] In conjunction with the first aspect, in some implementations of the first aspect, deleting the first voice instruction from the set of wake-free voice instructions includes: setting the first voice instruction as an instruction in the set of non-wake-free voice instructions.

[0056] In one possible implementation, the first and third voice command sets can contain the same command. For example, the first voice command set might contain the wake-free voice command "Turn the air conditioning to the highest setting," and the third voice command set might also contain the same command "Turn the air conditioning to the highest setting." This means that the wake-free voice command "Turn the air conditioning to the highest setting" in the first voice command set is an invalid wake-free voice command. When the vehicle is in wake-free mode, if the user says "Turn the air conditioning to the highest setting," the vehicle will not perform the corresponding operation.

[0057] Based on the above technical solution, when the first voice command set does not support user modification, the effect of "indirectly deleting commands from the first voice command set" can be achieved by using a third voice command set. This satisfies the user's need to delete commands and flexibly enables the dynamic reduction of the wake-up-free voice command set.

[0058] In conjunction with the first aspect, in some implementations of the first aspect, the wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. Before setting the first voice command as an instruction in the non-wake-free voice command set, the method further includes: deleting the first voice command from the second voice command set.

[0059] As mentioned earlier, the non-wake-up-free voice command set and the second voice command set are two mutually exclusive sets. Therefore, when setting the first voice command as an instruction in the third voice command set, if the first voice command is in the second voice command set, the first voice command should be deleted from the second voice command set.

[0060] In one possible implementation, before setting the first voice command as an instruction in the third voice command set, the method further includes: controlling the prompting device to prompt the user to delete the first voice command from the second voice command set; and in response to receiving input from the user to delete the first voice command from the second voice command set, setting the first voice command as an instruction in the third voice command set.

[0061] For example, the vehicle issues a voice command to the user: "[Wipe the glass] is currently not a wake-up-free voice command supported by the vehicle. Please remove this command from the blacklist before you can add it to the whitelist. Confirm to remove [Wipe the glass] from the blacklist?" In response to the user's voice input "Confirm", "Wipe the glass" is set as a command in the second set of voice commands.

[0062] Based on the above technical solution, the conflict between the second and third voice command sets can be effectively resolved even when the wake-up-free voice commands are removed, thus improving the standardization of wake-up-free voice command updates.

[0063] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: standardizing the first voice instruction, wherein the standardization process includes at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

[0064] Based on the above technical solutions, voice commands can be made more concise and clear, and can be understood and executed more accurately.

[0065] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: displaying a management interface for a set of wake-free voice commands, the management interface displaying an updated set of wake-free voice commands.

[0066] Based on the above technical solution, by providing a management interface for the set of wake-free voice commands, users can more easily view and manage the set of wake-free voice commands, making the updates of the set of wake-free voice commands more convenient and transparent, thereby enhancing the user experience.

[0067] In a second aspect, a voice interaction device is provided, the device comprising: an acquisition unit, configured to: acquire a first voice command; and an update unit, configured to: update the wake-up-free voice command set based on the usage frequency of the first voice command and a first recognition frequency when the first voice command is not a command in the wake-up-free voice command set, wherein the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state.

[0068] In conjunction with the second aspect, in some implementations of the second aspect, the updating unit is further configured to: update the wake-up-free voice command set based on the cancellation frequency of the first voice command when the first voice command is a command in the wake-up-free voice command set.

[0069] In conjunction with the second aspect, in some implementations of the second aspect, the updating unit is further configured to: set the first voice command as a command in the wake-up-free voice command set when the usage frequency is greater than or equal to the first preset threshold and the first recognition frequency is less than or equal to the second preset threshold.

[0070] In conjunction with the second aspect, in some implementations of the second aspect, the device further includes a control unit, which is configured to: when the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold, control the prompting device to prompt the user to add the first voice command to the wake-up-free voice command set; the updating unit is further configured to: in response to receiving input from the user confirming the addition of the first voice command to the wake-up-free voice command set, set the first voice command as a command in the wake-up-free voice command set.

[0071] In conjunction with the second aspect, in some implementations of the second aspect, the wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. The updating unit is also used to: set the first voice command as an instruction in the second voice command set.

[0072] In conjunction with the second aspect, in some implementations of the second aspect, the updating unit is also used to: delete the first voice command from the set of non-wake-up-free voice commands.

[0073] In conjunction with the second aspect, in some implementations of the second aspect, the device further includes a sending unit and a receiving unit. The sending unit is configured to: send a first voice command and a text recognition result of the first voice command to the cloud; the receiving unit is configured to: receive a verification result sent from the cloud, the verification result indicating whether the text recognition result is accurate and whether the first voice command is a command issued by the user to the vehicle; the acquiring unit is configured to: acquire a second recognition frequency based on the first recognition frequency and the verification result; and the updating unit is further configured to: update the wake-up-free voice command set based on the usage frequency and the second recognition frequency.

[0074] In conjunction with the second aspect, in some implementations of the second aspect, the updating unit is further configured to: delete the first voice command from the wake-up-free voice command set when the cancellation frequency is greater than or equal to a third preset threshold.

[0075] In conjunction with the second aspect, in some implementations of the second aspect, the device further includes a control unit, which is configured to: prompt the user to delete the first voice command when the cancellation frequency is greater than or equal to a third preset threshold; the update unit is further configured to: delete the first voice command from the wake-up-free voice command set in response to receiving input from the user confirming the deletion of the first voice command.

[0076] In conjunction with the second aspect, in some implementations of the second aspect, the updating unit is further configured to: set the first voice instruction as an instruction in the set of non-wake-up-free voice instructions.

[0077] In conjunction with the second aspect, in some implementations of the second aspect, the wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. The updating unit is also used to: delete the first voice command from the second voice command set.

[0078] In conjunction with the second aspect, in some implementations of the second aspect, the device further includes a processing unit for: standardizing the first voice instruction, the standardization process including at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

[0079] In conjunction with the second aspect, in some implementations of the second aspect, the device further includes a display unit for displaying a management interface for a set of wake-up-free voice commands, the management interface displaying an updated set of wake-up-free voice commands.

[0080] Thirdly, a voice interaction device is provided, comprising: a memory for storing a computer program; and a processor for executing the computer program stored in the memory, such that the device performs any of the possible methods in the first aspect.

[0081] Fourthly, a voice interaction system is provided, including a microphone and a computing platform, the computing platform including any of the possible devices of the second or third aspect.

[0082] Fifthly, a vehicle is provided, including any of the possible devices of the second or third aspect, or any of the possible systems of the fourth aspect.

[0083] In a sixth aspect, a computer-readable storage medium is provided having instructions stored thereon, which, when executed by a processor, cause the processor to implement any of the possible methods of the first aspect.

[0084] In a seventh aspect, a computer program product is provided, comprising computer program code that, when executed on a computer, causes the computer to implement any of the possible methods of the first aspect.

[0085] Eighthly, a chip is provided that includes circuitry for performing any of the possible methods in the first aspect. Attached Figure Description

[0086] Figure 1 is a functional block diagram of a vehicle 100 provided in an embodiment of this application.

[0087] Figure 2 is a schematic diagram of speech recognition.

[0088] Figure 3 is a schematic flowchart of the voice interaction method 300 provided in an embodiment of this application.

[0089] Figure 4 is a schematic flowchart of the voice interaction method 400 provided in an embodiment of this application.

[0090] Figure 5 is a schematic flowchart of a wake-up-free voice command update method 500 provided in an embodiment of this application.

[0091] Figures 6-8 are schematic diagrams illustrating a set of recommended wake-up-free functions provided in the embodiments of this application.

[0092] Figures 9-11 are schematic diagrams of a set of added wake-up-free voice commands provided in the embodiments of this application.

[0093] Figure 12 is a schematic flowchart of a wake-up-free voice command update method 1200 provided in an embodiment of this application.

[0094] Figure 13 is a graphical user interface (GUI) that provides prompts to the user according to an embodiment of this application.

[0095] Figures 14 and 15 show another GUI for prompting users, provided in an embodiment of this application.

[0096] Figure 16 illustrates a method 1600 using a verification model provided in an embodiment of this application.

[0097] Figure 17 is a schematic flowchart of a wake-up-free voice command update method 1700 provided in an embodiment of this application.

[0098] Figures 18-21 are schematic diagrams of a set of deletion wake-up-free voice commands provided in the embodiments of this application.

[0099] Figure 22 is a schematic flowchart of a wake-up-free voice command update method 2200 provided in an embodiment of this application.

[0100] Figure 23 is a GUI for prompting users provided in an embodiment of this application.

[0101] Figures 24-34 are a set of GUIs for managing wake-up-free voice commands provided in the embodiments of this application.

[0102] Figure 35 is a schematic block diagram of the voice interaction device 3500 provided in an embodiment of this application. Detailed Implementation

[0103] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0104] In the description of the embodiments of this application, unless otherwise stated, " / " means "or", for example, A / B can mean A or B; "and / or" in this document is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. "At least one" means one or more. For example, "at least one of A and B" is similar to "A and / or B", describing the association relationship between related objects, indicating that three relationships can exist. For example, at least one of A and B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0105] The prefixes such as "first" and "second" used in this application embodiment are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not constitute unnecessary restrictions due to the use of such prefixes. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0106] Figure 1 is a functional block diagram of a vehicle 100 provided in an embodiment of this application. As shown in Figure 1, the vehicle 100 may include a microphone 110, a computing platform 120, and a display device 130.

[0107] The in-cabin display devices 130 are mainly divided into two categories: the first is in-vehicle displays; the second is projection displays, such as head-up displays (HUDs). In-vehicle displays are physical displays and an important component of the in-vehicle infotainment system. Multiple displays can be installed in the cabin, such as digital instrument cluster displays, central control screens, displays in front of the front passenger, displays in front of the left and right rear passengers, and even the windows can be used as displays. Head-up displays, also known as head-up display systems, are mainly used to display driving information such as speed and navigation on a display device in front of the driver (e.g., the windshield). This reduces driver eye movement time, avoids pupil changes caused by eye movement, and improves driving safety and comfort.

[0108] The above description of the display device 130 uses an in-vehicle display screen and a projection display screen as examples, but the embodiments of this application are not limited thereto. For example, the display device 130 can also be a light display screen or a projection screen.

[0109] Some or all of the functions of vehicle 100 can be controlled by computing platform 120. Computing platform 120 may include one or more processors, such as processors 121 to 12n (n being a positive integer). A processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a central processing unit (CPU), microprocessor, graphics processing unit (GPU) (which can be understood as a type of microprocessor), or digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field-programmable gate array (FPGA). In reconfigurable hardware circuits, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement some or all of the functions of the aforementioned units. Furthermore, the processor can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), tensor processing unit (TPU), deep learning processing unit (DPU), etc. In addition, the computing platform 120 may also include a memory for storing instructions. Some or all of the processors 121 to 12n can call the instructions in the memory to implement the corresponding functions.

[0110] Optionally, the structure of the vehicle 100 described above is merely illustrative. In actual applications, various components of the vehicle 100 may be added or removed as needed.

[0111] Vehicle 100 may include an intelligent driving system, which may include an advanced driving assistant system (ADAS) and an autonomous driving system (ADS). The intelligent driving system can use various sensors on the vehicle (including but not limited to: lidar, millimeter-wave radar, camera devices, ultrasonic sensors, global positioning system, inertial measurement unit) to acquire information from the vehicle's surroundings, and analyze and process the acquired information to achieve functions such as obstacle perception, target recognition, vehicle positioning, path planning, and driver monitoring / alert, thereby improving the safety, automation, and comfort of driving the vehicle.

[0112] The vehicle 100 may also include a voice operating system, which can use a microphone 110 on the vehicle to acquire the user's voice, and analyze and process the acquired voice to achieve the purpose of controlling the vehicle (including but not limited to vehicle settings, map navigation, music and multimedia applications, air conditioning control, and weather inquiries) through technologies such as voice recognition and semantic understanding.

[0113] Voice interaction in intelligent driving typically refers to a user speaking a command, which the vehicle then responds to and executes accordingly. Users can control the vehicle without touching it or shifting their gaze, thus achieving a convenient and safe driving experience.

[0114] To facilitate a clear description of the technical solutions in the embodiments of this application, the terminology involved in the embodiments of this application will be explained as follows:

[0115] ASR: ASR is a technology that can convert human speech into text information and is an important component of voice interaction. It collects and analyzes speech signals, extracts the feature information within them, and then converts them into corresponding text.

[0116] Acoustic Model: In ASR, the acoustic model is responsible for extracting feature information from the raw speech signal and estimating its probability distribution, providing input to the decoder.

[0117] Decoder: The decoder is used to parse the probability matrix output by the acoustic model and convert it into understandable text.

[0118] Decoding path: The decoding path can represent the mapping of speech signals to different "specific texts". Different decoding paths can represent different specific statements.

[0119] Figure 2 illustrates a speech recognition diagram. As shown in Figure 2, the acoustic model can output a probability matrix of the user's spoken words, and the decoder can determine whether the user has spoken "specific text". Each "specific text" can be added to a different decoding path by the voice operating system. The decoder can calculate a score for each decoding path (i.e., the score for each "specific text") based on the probability matrix. When the score of a decoding path exceeds a preset wake-up threshold, the voice operating system can consider that the user has spoken the "specific text" corresponding to that decoding path, that is, the "specific text" has been successfully decoded.

[0120] In one voice interaction implementation, the user's spoken voice can be divided into two parts: a wake word and a command sentence.

[0121] Wake-up word: The function of a wake-up word is to activate the vehicle's voice operating system. Wake-up words can be preset by the car manufacturer, such as "Xiao A Xiao A", "Xiao L Student", "Hello, Xiao Y", etc., or they can be customized by the user.

[0122] Command phrases: Command phrases can be actions that users want the vehicle to perform, such as "turn on the air conditioning", "navigate to the company", "play a song by a certain singer", etc.

[0123] Wake-up State: In this embodiment, when the user speaks a specific wake-up word, the voice operating system is activated. At this time, the vehicle enters a state ready to acquire and process the user's voice; this state is the wake-up state. When the vehicle is in the wake-up state, after the user speaks a command, the vehicle can analyze the user's intent based on the command and ultimately execute the operation corresponding to the user's intent.

[0124] For example, if the wake word is "Xiao A Xiao A", in response to the user's voice input "Xiao A Xiao A", the voice assistant can reply "I'm here", or display animations or text on the screen to indicate that the vehicle is waiting to receive user commands. Next, in response to the user's voice input "I want coffee", the voice assistant can reply "Okay, start navigating to the nearest coffee shop for you".

[0125] It should be noted that the wake-up state usually has a time limit. If the user does not issue a command within a certain period of time, the vehicle will switch from the wake-up state to the non-wake-up state.

[0126] Non-wake-up state: In this embodiment of the application, the vehicle is in a non-wake-up state when the user does not utter a specific wake-up word.

[0127] For example, in a scenario where a user is conversing with a rear passenger, even if the user says "I want to drink coffee," the vehicle will not respond because the user did not utter the wake word "Xiao A Xiao A." Therefore, when the vehicle is in a non-wake-up state, it will not actively analyze the user's intention in making a voice statement.

[0128] It should be noted that the non-wake-up state includes a special case, namely the wake-free state. Before introducing the wake-free state, we will first introduce the wake-free state.

[0129] Wake-up-free functionality: The process of saying a wake-up word followed by a command is cumbersome and negatively impacts the user experience. Some solutions allow users to control the vehicle directly by speaking commands without saying a wake-up word; this is called wake-up-free functionality, and the command phrases that enable it are called wake-up-free voice commands.

[0130] Wake-up-free operation can provide users with a more convenient and natural interactive experience.

[0131] In some implementations, users can switch between wake-up-free states by turning the wake-up-free function on or off.

[0132] Wake-up-free state: In this embodiment of the application, when the user does not output a specific wake-up word, the voice operating system remains active, and the vehicle can recognize and respond to preset wake-up-free voice commands. At this time, the vehicle is in a wake-up-free state.

[0133] For example, when the vehicle is in wake-free mode, the preset wake-free voice command is "Navigate to my usual coffee shop". In response to the user's voice input "Navigate to my usual coffee shop", the voice assistant can reply "Okay, start navigating for you". The location of the coffee shop is pre-stored in the vehicle.

[0134] Wake-free voice command set: One or more preset wake-free voice commands can form a wake-free voice command set. If the user's voice input matches a command in the wake-free voice command set, the vehicle can perform the corresponding operation. That is, the user does not need to pre-activate the vehicle's voice operating system using a wake-up word; they can control the vehicle simply by speaking a wake-free voice command from the wake-free voice command set. For example, when the vehicle is in wake-free mode, in response to the user's voice input "turn on the air conditioning," since "turn on the air conditioning" is a command in the wake-free voice command set, the vehicle can perform the operation of turning on the air conditioning.

[0135] In some implementations, the set of wake-up-free voice commands can be pre-installed in the vehicle by the manufacturer. However, this set of commands has a limited number of instructions, fixed sentence structures, and cannot be modified by the user. This forces users to strictly adhere to fixed expressions to achieve wake-up-free operation, rather than adjusting the commands according to different users' expression habits, resulting in a poor user experience.

[0136] In addition, the aforementioned set of wake-up-free voice commands may also include commands that users use less frequently or not at all, which may prevent the vehicle from acquiring enough voice data to optimize the model used to recognize the voice commands, resulting in low recognition accuracy of these commands and false wake-ups.

[0137] Accidental wake-up: Accidental wake-up can occur when the vehicle or voice assistant is woken up incorrectly. Specifically, the vehicle responds incorrectly to voice commands that are not expected by the user, or the vehicle incorrectly executes an operation that the user did not give a command to.

[0138] For example, even if the user does not say "open the window", the vehicle recognizes other sounds in the environment as the voice command "open the window" and executes the "open the window" operation.

[0139] For example, when a user says "play a song by artist J", the vehicle might execute the command "play a song by artist H".

[0140] In order to develop personalized wake-free voice command sets for different users, the commands in the above wake-free voice command set can be updated.

[0141] One solution involves using frequently used voice commands as wake-up-free voice commands, while less frequently used commands can be removed, resulting in a set of wake-up-free voice commands that better aligns with user expression habits. However, this solution might include voice commands that are prone to false wake-ups, thus impacting the user experience.

[0142] Therefore, how to reduce the risk of accidental wake-up after updating the wake-up-free voice command is an urgent problem to be solved.

[0143] The present application provides a voice interaction method, device, and vehicle. The voice interaction method provided by the present application helps to reduce the risk of false wake-up while dynamically adjusting the set of wake-up-free voice commands, and also helps to provide users with a personalized voice interaction experience.

[0144] Figure 3 shows a schematic flowchart of a voice interaction method 300 provided in an embodiment of this application. The method 300 includes:

[0145] S301: Obtain the first voice command.

[0146] This application does not limit the specific method of obtaining the first voice command in any way.

[0147] For example, the first voice command could be the user's voice input obtained when the vehicle is in a wake-free state. In this case, the user's voice input is a command from the set of wake-free voice commands.

[0148] For example, the first voice command can also be the user's voice input obtained when the vehicle is in a wake-up state. In this case, the user's voice input can be a command from the wake-up-free voice command set, or it can be a command not from the wake-up-free voice command set.

[0149] S302: If the first voice command is not a command in the wake-free voice command set, update the wake-free voice command set based on the usage frequency and the first recognition frequency of the first voice command. The first recognition frequency is the frequency at which the vehicle recognizes the first voice command in the non-wake-up state.

[0150] In this application embodiment, the purpose of statistically analyzing the usage frequency is to assess whether the first voice command is a frequently used command by the user, and the purpose of statistically analyzing the first recognition frequency is to assess whether the first voice command is a command that is easily misrecognized.

[0151] Optionally, the wake-free voice command set is updated based on the usage frequency and the recognition frequency of the first voice command, including: when the usage frequency is greater than or equal to a first preset threshold and the recognition frequency is less than or equal to a second preset threshold, the first voice command is set as a command in the wake-free voice command set.

[0152] The first preset threshold can be a pre-set usage frequency threshold, and the second preset threshold can be a pre-set misidentification frequency threshold.

[0153] If the frequency of use of the first voice command is greater than or equal to the first preset threshold, it means that the user frequently uses the first voice command; if the first recognition frequency of the first voice command is less than or equal to the second preset threshold, it means that the probability of the first voice command being misrecognized is low. Therefore, when the frequency of use is greater than or equal to the first preset threshold and the first recognition frequency is less than or equal to the second preset threshold, setting the first voice command as a command in the wake-up-free voice command set can improve the efficiency of user interaction with the vehicle without bringing an excessively high risk of misrecognition.

[0154] It should be noted that if the first recognition frequency of the first voice command is greater than the second preset threshold, it means that the first voice command is easily misrecognized.

[0155] For example, command B is not a wake-up-free voice command; command B should be recognized when the vehicle is in a wake-up state. If command B is frequently recognized when the vehicle is not in a wake-up state, it may be because some ambient noise is similar to certain words or syllables in command B, causing the vehicle to mistakenly identify the ambient noise as command B. This means that command B has a high false recognition rate. Therefore, if command B is included in the wake-up-free voice command set, there is a possibility that the vehicle or voice assistant may be incorrectly woken up.

[0156] Therefore, the embodiments of this application simultaneously consider the usage frequency and the first recognition frequency to update the wake-up-free voice command set, which not only helps the vehicle to respond quickly to the user's frequently used commands, but also helps to reduce the risk of false wake-up.

[0157] Furthermore, the embodiments of this application do not limit the definition and statistical methods of usage frequency and first identification frequency.

[0158] For example, the usage frequency can be the total number of times a user uses the first voice command within a certain time range. For instance, a user used the command "wipe the window" a total of 20 times in the past month. Of these, 15 times were recognized when the vehicle was in a wake-up state, and 5 times were recognized when the vehicle was not in a wake-up state. Therefore, the usage frequency of this command is 20 times / month, and the first recognition frequency of this command is 5 times / month. If the first preset threshold is 15 times / month and the second preset threshold is 8 times / month, then "wipe the window" can be set as a command in the wake-up-free voice command set; that is, a new wake-up-free voice command, "wipe the window," is added.

[0159] Optionally, the wake-free voice command set is updated based on the usage frequency and the recognition frequency of the first voice command, including: when the usage frequency is greater than or equal to a first preset threshold and the recognition frequency is less than or equal to a second preset threshold, the control prompting device prompts the user to add the first voice command to the wake-free voice command set; in response to obtaining the user's input confirming the addition of the first voice command to the wake-free voice command set, the first voice command is set as a command in the wake-free voice command set.

[0160] In some embodiments, user input is required before adding new wake-up-free voice commands.

[0161] For example, the vehicle might issue a voice prompt to the user: "Do you want to set [Wipe the glass] as a wake-up-free voice command? Answer 'Confirm' to set it, or answer 'Cancel' to stop the prompt." In response to the user's voice input "Confirm," "Wipe the glass" is set as a command in the wake-up-free voice command set.

[0162] In this way, the set of wake-up-free voice commands can be made more in line with user preferences, taking into account usage frequency and primary recognition frequency, as well as the user's actual wishes.

[0163] Optionally, the wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. Setting the first voice command as a command in the wake-free voice command set includes setting the first voice command as a command in the second voice command set.

[0164] For example, the first set of voice commands may include pre-set wake-up-free voice commands in the vehicle, and the second set of voice commands may include wake-up-free voice commands determined by the user.

[0165] In some implementations, the first set of voice commands may also be referred to as the "system preset list," and the second set of voice commands may also be referred to as the "whitelist." Since the first set of voice commands cannot be modified by the user, the user can add any new commands to the second set of voice commands, thus achieving the effect of "adding a new command."

[0166] For example, the first voice command set and the second voice command set can be two mutually exclusive sets, that is, the first voice command set and the second voice command set include different wake-up-free voice commands. In this case, the commands in the second voice command set can be regarded as wake-up-free voice commands added by the user based on the vehicle's preset commands.

[0167] For example, the first and second voice command sets may also include the same wake-up-free voice commands. In this case, the same command indicates that the user has selected some voice commands from the vehicle's preset wake-up-free voice command set as the wake-up-free voice commands that the user wishes to use.

[0168] In some embodiments, the wake-up-free voice command "heat seat" in the first voice command set may cause a false wake-up, but the user chooses to add the command to the second voice command set. This means that even if the vehicle's preset command "heat seat" may cause a false wake-up, the user can accept this situation and choose to use the command.

[0169] It should be understood that the reason why users can accept the wake-up-free voice commands in the second set of voice commands that may cause false wake-ups is because these commands can provide great convenience in the user's daily driving. Therefore, the occasional false wake-up is not enough to weaken the practicality of the wake-up-free command and will not cause serious problems or inconvenience.

[0170] Optionally, before setting the first voice command as an instruction in the second set of voice commands, method 300 further includes: deleting the first voice command from the set of non-wake-free voice commands.

[0171] For example, the set of non-wake-up-free voice commands may include wake-up-free voice commands that the user does not wish to use, and the set of non-wake-up-free voice commands may also be referred to as the third voice command set. For example, the third voice command set may include the command "wipe the glass," which means that the command "wipe the glass" cannot be used for the vehicle's wake-up-free function.

[0172] In some implementations, the third voice command can also be referred to as a "blacklist". Since the first voice command set does not support user modification, users can add commands they do not want to use to the third voice command set. If the command is also a command in the first voice command set, it can achieve the effect of "indirectly deleting commands in the first voice command set".

[0173] For example, the first voice command set includes the wake-free voice command "Turn the air conditioning to the highest setting," and the third voice command set also includes the same command "Turn the air conditioning to the highest setting." This means that the wake-free voice command "Turn the air conditioning to the highest setting" in the first voice command set is an invalid wake-free voice command. When the vehicle is in wake-free mode, if the user says "Turn the air conditioning to the highest setting," the vehicle will not perform the corresponding operation.

[0174] Clearly, the third set of voice commands and the second set of voice commands are two mutually exclusive sets. Therefore, before adding a first voice command to the second set of voice commands, if the first voice command is already in the third set of voice commands, the conflict between the two should be resolved first. That is, the first voice command should be deleted from the third set of voice commands.

[0175] Optionally, before setting the first voice command as an instruction in the second voice command set, method 300 further includes: controlling the prompting device to prompt the user to delete the first voice command from the third voice command set; and in response to receiving input from the user to delete the first voice command from the third voice command set, setting the first voice command as an instruction in the second voice command set.

[0176] For example, before adding a first voice command to the second voice command set, the user needs to be informed of the conflict and prompted to manually delete the first voice command from the third voice command set. Alternatively, the vehicle can automatically delete the first voice command from the third voice command set.

[0177] For example, the vehicle might issue a voice command to the user saying, "[Wipe the glass] is currently not a wake-up-free voice command supported by the vehicle. Please remove this command from the blacklist before adding it to the whitelist. Confirm to remove [Wipe the glass] from the blacklist?" In response to the user's voice input "Confirm," "Wipe the glass" is set as a command in the second set of voice commands.

[0178] As can be seen from the above, some embodiments of this application can update the vehicle's preset, unmodifiable first voice command set based on the second and third voice command sets, thereby optimizing the user's voice interaction experience.

[0179] Figure 4 shows a schematic flowchart of the voice interaction method 400 provided in an embodiment of this application. The flowchart illustrates how the first voice command set, the second voice command set, and the third voice command set are used in conjunction in a wake-free state. As shown in Figure 4, the method 400 includes:

[0180] S401: Obtain the first voice command.

[0181] S402: The first voice command successfully matched the command in the wake-up-free voice command set.

[0182] The wake-up-free voice command set may include a first voice command set and a second voice command set. The definitions of the first and second voice command sets are as described above and will not be repeated here.

[0183] A successful match can mean that the user's voice is successfully recognized and matches a command in the set of wake-up-free voice commands. For example, a first voice command can be obtained based on data collected by the microphone 110 in the vehicle's cabin. Using voice recognition technology, the first voice command can be converted into text or other command formats. If the converted text or command matches the wake-up-free voice commands supported by the vehicle, then the first voice command is considered to have successfully matched the wake-up-free voice commands supported by the vehicle.

[0184] S403: Determine whether the first voice command is a command in the second voice command set.

[0185] If the first voice command is a command from the second set of voice commands, execute S405. Otherwise, execute S404.

[0186] S404: Determine whether the first voice command is a command in the third voice command set.

[0187] If the first voice command is a command from the third set of voice commands, execute S406. Otherwise, execute S405.

[0188] S405: Execute the operation corresponding to the first voice command.

[0189] S406: Do not execute the operation corresponding to the first voice command.

[0190] Optionally, method 300 further includes: standardizing the first speech instruction, wherein the standardization process includes at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

[0191] This application does not limit the standardization processing method. Exemplary methods for standardization processing may include the following:

[0192] 1. Remove interjections: For example, remove the interjections "too cold", "please", and "ba" from the instruction "It's too cold, please adjust the air conditioner temperature to 22 degrees". The processed instruction is "Adjust the air conditioner temperature to 22 degrees".

[0193] 2. Remove redundant information: For example, the instruction "Raise the air conditioner temperature to 22 degrees" contains two instances of "raise", which can be simplified to "Raise the air conditioner temperature to 22 degrees".

[0194] 3. Correct typos: For example, correct the instruction "adjust the air temperature" to "adjust the air conditioner temperature".

[0195] 4. Adjust word order: For example, change the instruction "Go to the company navigation" to "Navigate to the company" to make the sentence more in line with standard expression.

[0196] The standardized processing method provided in the embodiments of this application can make voice commands more concise, clear, and more accurately understood and executed.

[0197] Figure 5 shows a schematic flowchart of a wake-up-free voice command update method 500 provided in an embodiment of this application. As shown in Figure 5, the method 500 includes:

[0198] S501: Obtain the first voice command when the vehicle is in a wake-up state.

[0199] S502: Standardize the first voice command to obtain the second voice command.

[0200] For example, the wake-up word is "Little L". In response to the user's voice input "Little L", the vehicle enters the wake-up state. Then, in response to the user's first voice command "I'm so hungry, I want to go to a restaurant, I want to eat that stir-fry", the vehicle can process the voice and get the standardized second voice command "I want to go to a restaurant to eat stir-fry".

[0201] S503: Determine whether the second voice command is an instruction in the third set of voice commands.

[0202] If the second voice command is an instruction in the third set of voice commands, execute S504. Otherwise, execute S505.

[0203] S504: End the process.

[0204] As described above, the third set of voice commands includes voice commands without wake-up that the user does not want to use. If the second voice command is an instruction in the third set of voice commands, it indicates that the user has previously expressed that they do not want to use the second voice command as a voice command without wake-up. Therefore, there is no need to enter the subsequent process of updating the voice commands without wake-up.

[0205] S505: Determine whether the length of the second voice command is greater than a preset length threshold.

[0206] If it is greater than the preset length threshold, execute S506. Otherwise, execute S504.

[0207] Generally, voice commands without wake-up with shorter lengths may be similar in pronunciation to other phrases, which causes the vehicle to be easily woken up by mistake. Therefore, in the embodiments of the present application, if the length of the second voice command is too short, it will not be considered as a voice command without wake-up.

[0208] [[ID=ZI]]The embodiments of the present application do not limit the preset length threshold, and this threshold may vary according to different application scenarios, language characteristics, etc. Exemplarily, in the Chinese expression scenario, one Chinese character is equal to 2 characters, and the preset length threshold can be 6 characters. For example, when the second voice command is "go home", the length of the second voice command is 4 characters, so the process of method 500 ends.

[0209] S506: Determine whether the second voice command is an instruction in the set of voice commands without wake-up.

[0210] If the second voice command is an instruction in the set of voice commands without wake-up, execute S507. Otherwise, execute S508.

[0211] S507: Recommend to the user to use the instructions in the set of voice commands without wake-up.

[0212] In some intelligent driving scenarios, the user may not know which instructions belong to the voice commands without wake-up. Therefore, when the user operates the vehicle by using the wake-up word combined with a command sentence, the vehicle can give a timely prompt to inform the user that they can use the corresponding voice commands without wake-up in the future to simplify the operation process.

[0213] Figures 6 - 8 show a set of schematic diagrams for recommending the use of the wake-up-free function provided by the embodiments of the present application.

[0214] As shown in Figure 6, in response to the user's voice input "Hey A, Hey A, turn on the air conditioning," the vehicle can perform the operation of turning on the air conditioning. As shown in Figure 7, after the vehicle performs the corresponding operation, the vehicle can issue a voice message to the user: "Hey A kindly reminds you, [turn on the air conditioning] is a wake-up-free voice command. Next time, you can try saying '[turn on the air conditioning]' directly to me." For example, screen 701 can display the GUI shown in Figure 8.

[0215] As shown in Figure 8, screen 701 can display a "Recommended Tip for Wake-up-Free Function" and the specific prompt content: "It was detected that you performed the operation '[Turn on the air conditioner]' through the voice assistant. You can directly say the command '[Turn on the air conditioner]' next time to help you get a more convenient experience." and a confirmation control. If the user clicks the confirmation control, the prompt box can be closed, or it can be automatically closed after the prompt box appears for a period of time.

[0216] S508: Add the second voice command to the decoding path.

[0217] As mentioned earlier, adding the second voice command to the decoding path allows for the calculation of its score in subsequent processes. When the score of the second voice command exceeds a preset wake-up threshold, it can be determined that the second voice command was successfully decoded, that is, that the user has spoken the second voice command.

[0218] By counting the number of times the second voice command was successfully decoded, we can obtain the number of times the user spoke the second voice command.

[0219] S509: Statistically measure the frequency of use of the second voice command and the frequency of recognition of the first voice command.

[0220] In some implementations, the frequency of use of the second voice command can be divided into two cases for statistical analysis: the first case is the frequency of use of the second voice command when the vehicle is in a non-wake-up state, which can also be called the first recognition frequency; the second case is the frequency of use of the second voice command when the vehicle is in a wake-up state, which can also be called the wake-up frequency; the sum of the first recognition frequency and the wake-up frequency can represent the frequency of use of the second voice command.

[0221] For example, statistics show that the "View Stocks" function was successfully decoded 30 times in one month. Of these, 25 were successfully decoded when the vehicle was in a wake-up state, and the other 5 were successfully decoded when the vehicle was not in a wake-up state. Therefore, the usage frequency of "View Stocks" is 30 times / month, the primary recognition frequency is 5 times / month, and the wake-up frequency is 25 times / month.

[0222] It should be noted that since the second voice command is not part of the wake-up-free voice command set, when a user wants the vehicle to perform the corresponding operation, they should first use a wake-up word to wake the vehicle before speaking the second voice command. Theoretically, the second voice command should only be successfully decoded when the vehicle is awake. Therefore, if the second voice command is successfully decoded when the vehicle is not awake, it indicates that the second voice command was incorrectly recognized.

[0223] S510: Determine whether the usage frequency of the second voice command is greater than or equal to the first preset threshold, and whether the first recognition frequency of the second voice command is less than or equal to the second preset threshold.

[0224] If the first recognition frequency of the second voice command is greater than the second preset threshold, execute S515.

[0225] For example, statistics show that the first recognition frequency of "open it" is 20 times per month. If the second preset threshold is 15 times per month, this means that "open it" is often incorrectly recognized when the vehicle is in a non-wake-up state. If "open it" is used as a wake-up-free voice command, it may lead to more misrecognitions. Therefore, the voice command "open it" is added to the third voice command set, meaning that "open it" cannot be used for the vehicle's wake-up-free function.

[0226] If the frequency of use of the second voice command is less than the first preset threshold, execute S514.

[0227] For example, when the vehicle is in a wake-up state, in response to the user's voice input "Navigate to People's Hospital," the vehicle counts the usage frequency of the voice command "Navigate to People's Hospital" as 1 time, meaning that the user has only said "Navigate to People's Hospital" once before. The first preset threshold is 10 times, indicating that the voice command "Navigate to People's Hospital" is not currently a frequently used command by the user. Therefore, the usage frequency of "Navigate to People's Hospital" is updated to 2 times, and the usage of this command is continuously counted.

[0228] If the frequency of use of the second voice command is greater than or equal to the first preset threshold, and the first recognition frequency of the second voice command is less than or equal to the second preset threshold, execute S511.

[0229] S511: Recommends the use of the second voice command as a wake-up-free voice command to the user.

[0230] S512: Determine whether the user agrees to use the second voice command as a wake-up-free voice command.

[0231] If the user agrees, proceed to step S513. Otherwise, proceed to step S514.

[0232] S513: Add the second voice command to the second voice command set.

[0233] S514: Add the second voice command to the third voice command set.

[0234] Figures 9-11 illustrate a set of schematic diagrams of adding wake-up-free voice commands provided in the embodiments of this application.

[0235] As shown in Figure 9, in response to the user's voice input "Xiao A Xiao A, navigate to the city library", the vehicle can perform the corresponding navigation operation. As shown in Figure 10, after the vehicle starts navigation, the vehicle can issue a voice prompt to the user: "Do you want to set [navigate to the city library] as a wake-up-free voice command? Answer to confirm to add it, answer to cancel to stop prompting."

[0236] If the user answers "Confirm," it means the user agrees to add a wake-up-free voice command, and "Navigate to the City Library" will be added to the second voice command set. If the user answers "Cancel," it means the user does not intend to use "Navigate to the City Library" as a wake-up-free voice command, and "Navigate to the City Library" will be added to the third voice command set. In the future, the user will no longer be prompted to add a wake-up-free voice command for the "Navigate to the City Library" command.

[0237] For example, during the process of adding a wake-up-free voice command, screen 701 can also display the GUI shown in Figure 11.

[0238] As shown in Figure 11, screen 701 displays "Add wake-up-free voice command prompt" and the specific prompt content "We recommend you add: [Navigate to City Library] as a wake-up-free voice command. This will make your daily operations more convenient. Confirm to add?", a confirmation control, and a cancellation control.

[0239] If the user clicks the confirm control, "Navigate to the city library" will be added to the second set of voice commands; if the user clicks the cancel control, "Navigate to the city library" will be added to the third set of voice commands. In the future, there will be no more "add wake-up-free voice command prompts" for "Navigate to the city library".

[0240] Figure 12 shows a schematic flowchart of a wake-free voice command update method 1200 provided in an embodiment of this application. The flowchart illustrates how this embodiment resolves the conflict between a second voice command set and a third voice command set during the process of adding a new wake-free voice command. As shown in Figure 12, the method 1200 includes:

[0241] S1201: The user confirms that the first voice command has been added to the second voice command set.

[0242] Users can add the first voice command manually or through voice commands.

[0243] For example, in the GUI shown in Figure 11, the user can click the confirmation control to indicate that the user confirms adding the first voice command to the second voice command set, or the user can also say the voice command "Add [Navigate to City Library] to the second voice command set".

[0244] S1202: Determine whether the first voice command is a command in the second voice command set.

[0245] If the first voice command is an instruction from the second set of voice commands, execute S1203. Otherwise, execute S1204.

[0246] S1203: Informs the user that the first voice command already exists and does not need to be added again.

[0247] For example, Figure 13 illustrates a GUI for prompting a user according to an embodiment of this application. As shown in Figure 13, screen 701 displays "Add wake-up-free voice command prompt" and the specific prompt content "Dear user, [Navigate to Library] is already your wake-up-free voice command. There is no need to add it again. You can use it directly." and a confirmation control.

[0248] S1204: Determine whether the first voice command is a command in the third voice command set.

[0249] If the first voice instruction is an instruction from the third set of voice instructions, execute S1206. Otherwise, execute S1205.

[0250] S1205: Add the first voice command to the second voice command set.

[0251] S1206: Prompt the user to first delete the first voice command from the third voice command set.

[0252] S1207: Does the user agree to remove the first voice command from the third voice command set?

[0253] If the user agrees, proceed to step S1208. Otherwise, proceed to step S1209.

[0254] S1208: Remove the first voice command from the third voice command set.

[0255] If S1208 is executed, then S1205 is executed.

[0256] S1209: Process complete.

[0257] For example, Figures 14 and 15 illustrate another GUI for prompting the user provided in an embodiment of this application. As shown in Figure 14, screen 701 displays "Add wake-up-free voice command prompt" and the specific prompt content "Dear user, it has been detected that [Navigate to Library] is a deleted wake-up-free voice command. Please confirm whether you want to set this command as a wake-up-free voice command?", a confirmation setting control 1401, and a cancellation control. When the user clicks the confirmation setting control 1401, the GUI shown in Figure 15 can be displayed on the central control screen.

[0258] As shown in Figure 15, this interface displays a notification indicating successful addition of a wake-up-free voice command. Screen 701 shows "Successful Addition" and the specific message: "Dear user, [Navigate to Library] has been successfully added as your wake-up-free voice command. Would you like to add other wake-up-free voice commands?", along with options to add or remove controls. If the user clicks "Add Control," they can continue adding wake-up-free voice commands; if the user clicks "Cancel," the notification box closes.

[0259] Optionally, before updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command, method 300 further includes: sending the first voice command and the text recognition result of the first voice command to the cloud; receiving a verification result sent by the cloud, the verification result being used to indicate whether the text recognition result is accurate and whether the first voice command is an instruction issued by the user to the vehicle; updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command, including: obtaining a second recognition frequency according to the first recognition frequency and the verification result; and updating the wake-free voice command set based on the usage frequency and the second recognition frequency.

[0260] Figure 16 illustrates a method 1600 using a verification model provided in an embodiment of this application. As shown in Figure 16, the method 1600 includes:

[0261] S1601: The first voice command was successfully decoded when the vehicle was in a non-wake-up state.

[0262] S1602: Send the audio of the first voice command and the vehicle-side text recognition result of the first voice command to the cloud.

[0263] S1603: Analyze using a large ASR model in the cloud.

[0264] S1604: Based on the analysis, return the verification result.

[0265] S1605: Obtain the second recognition frequency based on the verification result.

[0266] The content of the cloud-based ASR large model analysis may include at least one of the following:

[0267] (1) Determine whether the text recognition results on the vehicle are accurate.

[0268] The ASR (Automatic Speech Recognition) large-scale model can more accurately recognize user speech, especially in context-sensitive, noisy environments, and situations involving a large number of entities and proper nouns. Therefore, the ASR large-scale model can perform a second speech recognition on the audio of the first voice command to obtain the cloud-based text recognition result for the first voice command. If the vehicle-side text recognition result and the cloud-based text recognition result are inconsistent, the vehicle-side text recognition result is considered incorrect, meaning the first voice command was incorrectly recognized.

[0269] For example, within a month, when the vehicle was in a non-wake-up state, the command "go pick up the child from school" was successfully decoded 18 times, which is the first recognition frequency of 18 times / month. Among them, there were 10 instances where the cloud-based text recognition results and the vehicle-side text recognition results were inconsistent, which means that these 2 instances of incorrect recognition occurred, i.e., the second recognition frequency of 10 times / month.

[0270] (2) Determine the context of the sound.

[0271] The ASR model can also determine whether the audio of the first voice command belongs to human-to-human interaction, human-to-vehicle interaction, or electronic device sound. If the audio belongs to human-to-human interaction or electronic device sound, the first voice command is considered to have been incorrectly recognized.

[0272] For example, within a week, when the vehicle was in a non-wake-up state, the command "Activate Joy Mode" was successfully decoded 7 times. Among these, 2 times were the sound of electronic devices, 2 times were the content of conversations between passengers, and 3 times were the user's attempt to give instructions to the vehicle. Therefore, "Activate Joy Mode" was incorrectly recognized 4 times, that is, the second recognition frequency was 4 times / week.

[0273] This application does not limit the model for achieving the above two judgments.

[0274] For example, in some implementations, the ASR big model can directly complete the analysis in (1) and (2) above.

[0275] For example, in some implementations, the ASR model can include a first model and a second model. The first model can be used to determine whether the text recognition result on the vehicle is accurate, and the second model can be used to determine the audio context. For instance, the first model can be a speech recognition model, and the second model can be an audio model.

[0276] The solution provided in this application embodiment can accurately obtain the probability that the first voice command is misrecognized.

[0277] It should be noted that if the information processed in this application involves users' personal information, the processing of users' personal information will be based on legality, and users will be fully informed and authorized, in accordance with the relevant laws and regulations on personal information protection in the country or region where the application is made.

[0278] Optionally, method 300 further includes: if the first voice command is an instruction in the wake-free voice command set, updating the wake-free voice command set based on the cancellation frequency of the first voice command.

[0279] The cancellation frequency of the first voice command refers to the frequency at which, in response to a user's wake-up-free voice command, the vehicle begins to execute the corresponding operation, but within a subsequent first time period, the operation is actively cancelled by the user through voice or gesture. The first time period can be 3 seconds or 5 seconds; this embodiment does not limit the duration of the first time period.

[0280] Among them, canceling the corresponding operation by voice can be: the user cancels the wake-up-free voice command A by voice command B. Voice command B can be "cancel command A" or a command that is the opposite of the action corresponding to command A.

[0281] For example, at the first moment, in response to the voice input "open the windows," the vehicle begins to open the front and rear windows. Three seconds later, at the second moment, the user issues the command "close the windows" or "cancel opening the sunroof," and the vehicle begins to close the front and rear windows. The command "close the windows" or "cancel opening the sunroof" can be considered a cancellation of the command "open the windows."

[0282] Among them, canceling the corresponding operation by action can be: the user cancels the wake-up-free voice command A by action C, and action C can be the opposite of the action corresponding to command A.

[0283] For example, at the first moment, in response to the user's voice input "turn on the air conditioning," the vehicle's air conditioning is turned on. At the second moment, 5 seconds later, the user manually turns off the air conditioning switch. The user's action of turning off the air conditioning switch can be considered a cancellation of the "turn on the air conditioning" command.

[0284] The embodiments of this application do not limit the statistical method for the revocation frequency.

[0285] For example, the total number of cancellations can be used as the cancellation frequency of the first voice command.

[0286] For example, if the total number of undo commands for the wake-up-free voice command "Navigate to the company" is 15, then the undo frequency of this command is 15.

[0287] For example, the total number of wake-up calls and the total number of cancellations of the first voice command can be counted, and the ratio of the total cancellations to the total wake-up calls can be used as the cancellation frequency of the first voice command. The total number of wake-up calls can be the number of times the first voice command was successfully recognized when the vehicle is in a wake-up-free state.

[0288] For example, the total number of wake-up calls for the wake-up-free voice command "Turn on seat massage" is 50, and the total number of cancellations is 10. According to the above calculation method, the cancellation frequency of "Turn on seat massage" is 10 / 50 = 20%.

[0289] In some implementations, a time decay mechanism can be used to calculate the withdrawal frequency. Time decay is a statistical concept that assigns different weights to data at different points in time, allowing more recent data to have a greater impact on the statistical results, while data from more distant periods have a smaller impact.

[0290] For example, the following formula (1) can express a method of time decay: F t =f t +α×F t-1 (1)

[0291] In formula (1), t represents the time window number, t≥1, α represents the aging factor, and f t F represents the frequency of canceling wake-up-free voice commands within the t-th time window. t-1 F represents the cancellation frequency of wake-up-free voice commands up to the (t-1)th time window. t This indicates the cancellation frequency of wake-up-free voice commands up to the t-th time window.

[0292] For example, if the time window is one week long, the aging factor α is 0.9, the cancellation frequency of the wake-up-free voice command A up to the end of last week is 20, and the cancellation frequency of command A within this week is 10, then the cancellation frequency up to the end of this week is 10 + 20 × 0.9 = 28.

[0293] For example, if the time window is one month long, the aging factor is 0.4, the cancellation frequency of the wake-up-free voice command B up to the end of last month is 50%, and the cancellation frequency of command B this month is 60%, then the cancellation frequency up to the end of this month is 60% + 50% × 0.4 = 80%.

[0294] For example, the following formula (2) can also express a method of time decay:

[0295] In formula (2), F xThis represents the cancellation frequency up to the x-th time window, where x is the time window number, x≥1. x1 represents the first time window within the statistical time range. n Let represent the nth time window within the statistical time range. C(x) represents the withdrawal frequency within the xth time window, and α(x) represents the aging factor within the xth time window. α(x) can be expressed in different forms depending on different statistical needs. For example, if each time window is one month long and the statistical time range is five months, α(x) can be expressed as:

[0296] In formula (3), if x represents the fifth month (the most recent month), the aging factor is 1; if x represents the first to fourth months (the past four months), the aging factor is 0. Therefore, formula (2) can be simplified to: F x =C(x5)×α(x5)=C(x5) (4)

[0297] Formula (4) represents the cancellation frequency F x The decision is based solely on the cancellation frequency in the fifth month, without considering the cancellation frequency in the first four months.

[0298] In this embodiment of the application, the purpose of statistically analyzing the revocation frequency is to evaluate whether the instructions in the set of wake-up-free voice instructions are those that frequently cause the voice assistant to be incorrectly woken up.

[0299] The embodiments of this application update the voice command set based on the cancellation frequency of the first voice command, which helps to reduce false wake-ups and improve the user's driving experience.

[0300] Optionally, the wake-free voice command set is updated based on the cancellation frequency of the first voice command, including: deleting the first voice command from the wake-free voice command set when the cancellation frequency is greater than or equal to a third preset threshold.

[0301] The third preset threshold can be a pre-set cancellation frequency threshold. If the cancellation frequency of the first voice command is greater than or equal to the third preset threshold, it means that the first voice command is frequently cancelled during use.

[0302] Specifically, if the cancellation of the first voice command occurs frequently, it indicates that the vehicle may be frequently performing operations that the user does not need. Therefore, the cancellation frequency of the first voice command can reflect the probability of false wake-ups caused by it. A higher cancellation frequency indicates a higher probability of false wake-ups, which may seriously affect the user experience. If the first voice command continues to be used as a command in the set of wake-up-free voice commands, users may have a poor experience due to frequent false wake-ups, and may even face more safety hazards while driving. Therefore, the first voice command can be removed from the set of wake-up-free voice commands.

[0303] Optionally, the wake-free voice command set is updated based on the cancellation frequency of the first voice command, including: when the cancellation frequency is greater than or equal to a third preset threshold, the control prompt device prompts the user to delete the first voice command; in response to receiving input from the user confirming the deletion of the first voice command, the first voice command is deleted from the wake-free voice command set.

[0304] In some embodiments, the user's opinion needs to be sought before removing wake-up-free voice commands.

[0305] For example, the vehicle might ask the user, "Do you want to remove '[Open the fragrance]' from the set of wake-up-free voice commands? Answer 'confirm' to remove it, answer 'cancel' to stop the prompt." In response to the user's voice input "confirm," "Open the fragrance" will be removed from the set of wake-up-free voice commands.

[0306] In this way, the set of wake-up-free voice commands can be updated based on the user's actual wishes, making the set of wake-up-free voice commands more in line with the user's usage preferences.

[0307] Optionally, deleting the first voice command from the wake-free voice command set includes: setting the first voice command as a command in the non-wake-free voice command set.

[0308] Optionally, the wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. Before setting the first voice command as an instruction in a non-wake-free voice command set, the method further includes: deleting the first voice command from the second voice command set.

[0309] The first, second, and third voice command sets can be referred to in the previous descriptions and will not be repeated here.

[0310] Figure 17 shows a schematic flowchart of a wake-free voice command update method 1700 provided in an embodiment of this application. The flowchart illustrates how this embodiment updates the wake-free voice command set according to the cancellation frequency of a first voice command. As shown in Figure 17, the method 1700 includes:

[0311] S1701: Obtain the first voice command when the vehicle is in wake-free mode.

[0312] S1702: The first voice command successfully matched the command in the wake-up-free voice command set.

[0313] S1703: Determine whether the first voice command is a command in the second voice command set.

[0314] If the first voice command is an instruction from the second set of voice commands, execute S1704. Otherwise, execute S1705.

[0315] S1704: Process terminated.

[0316] S1705: Determine whether the first voice command is a command in the third voice command set.

[0317] If the first voice instruction is an instruction from the third set of voice instructions, execute S1704. Otherwise, execute S1706.

[0318] It is understandable that if the first voice command is a command from the second or third set of voice commands, it can be considered that the user has already made a choice about whether to keep or discard the first voice command, so the process can be terminated directly.

[0319] If the first voice command is neither a command in the second voice command set nor a command in the third voice command set, it means that the wake-up-free voice command is a command in the first voice command set, and the user has not made a clear choice regarding the retention or removal of the first voice command. Then, according to the following steps S1706-S1712, it can be considered whether the first voice command should be added to the second or third voice command set.

[0320] S1706: Determine if there is a record of the first voice command being revoked.

[0321] If the first voice command has a record of being revoked, execute S1707. Otherwise, execute S1704.

[0322] S1707: Obtain the cancellation frequency of the first voice command.

[0323] S1708: Determine whether the cancellation frequency is greater than or equal to the third preset threshold.

[0324] If the cancellation frequency is greater than or equal to the third preset threshold, execute S1709. Otherwise, execute S1704.

[0325] The statistical method for calculating the frequency of cancellations has been introduced earlier and will not be repeated here for the sake of brevity.

[0326] S1709: Recommend to the user that the first voice command be removed from the set of wake-up-free voice commands.

[0327] S1710: Determine whether the user agrees to remove the first voice command from the set of wake-up-free voice commands.

[0328] If the user agrees to remove the first voice command from the set of wake-up-free voice commands, execute S1711. Otherwise, execute S1712.

[0329] S1711: Add the first voice command to the third voice command set.

[0330] S1712: Add the first voice command to the second voice command set.

[0331] Figures 18-21 illustrate a set of schematic diagrams for deleting wake-up-free voice commands provided in an embodiment of this application.

[0332] As shown in Figure 18, at the first moment, the vehicle recognizes the command input "turn on the massage chair", begins to execute the corresponding operation, and issues a voice message to the user: "I heard you say [turn on the massage chair], now we will start your cabin massage time."

[0333] A second moment after the vehicle performs the operation shown in Figure 18, the user says the voice command "Xiao A Xiao A, turn off the massage chair" as shown in Figure 19, in order to retract the wake-free voice command "turn on the massage chair". In other implementations, the user can also retract the wake-free voice command "turn on the massage chair" by manually turning off the massage chair's switch.

[0334] In response to the user's voice input "Xiao A, Xiao A, turn off the massage chair" in Figure 19, the massage chair in the cabin stops operating. After the massage chair stops operating, the vehicle can issue a voice message to the user: "We have detected that you have repeatedly canceled the wake-up-free voice command [turn on the massage chair]. Do you want to delete this command?" If the user answers "yes," the first voice command is added to the third voice command set. If the user answers "no," the first voice command is added to the second voice command set.

[0335] For example, during the process of deleting wake-up-free voice commands, screen 701 can also display the GUI shown in Figure 21.

[0336] As shown in Figure 21, screen 701 displays a "Delete Wake-Free Voice Command Prompt" and the specific prompt content: "Detected that you have repeatedly revoked the wake-free voice command: [Open the massage chair]. Do you want to cancel this wake-free voice command?", a "Required Control" and a "Cancel Control". If the user clicks the "Required Control", "Open the massage chair" will be added to the third voice command set; if the user clicks the "Cancel Control", "Navigate to the city library" will be added to the second voice command set.

[0337] Figure 22 shows a schematic flowchart of a wake-free voice command update method 2200 provided in an embodiment of this application. The flowchart illustrates how this embodiment resolves the conflict between a second voice command set and a third voice command set during the deletion of wake-free voice commands. As shown in Figure 22, the method 2200 includes:

[0338] S2201: The user confirms that the first voice command has been added to the third voice command set.

[0339] The user can add the first voice command to the third voice command set via voice or action.

[0340] For example, in the scenario shown in Figure 20, a user can issue the voice command "Confirm deletion of this command", or, in the GUI shown in Figure 21, the user can click the required control to indicate that the user needs to add the first voice command to the third voice command set.

[0341] S2202: Determine whether the first voice command is a command in the third voice command set.

[0342] S2203: The user is prompted that the first voice command has been deleted and there is no need to repeat the operation.

[0343] For example, Figure 23 illustrates a GUI for prompting a user according to an embodiment of this application. As shown in Figure 13, screen 701 displays "Delete wake-up-free voice command prompt" and the specific prompt content "Dear user, [Navigate to Library] has been deleted from your wake-up-free voice command set. There is no need to delete it again. Please be aware." and a confirmation control.

[0344] S2204: Determine whether the first voice command is a command in the second voice command set.

[0345] S2205: Add the first voice command to the third voice command set.

[0346] S2206: Prompt the user to first delete the first voice command from the second voice command set.

[0347] S2207: Does the user agree to remove the first voice command from the second voice command set?

[0348] S2208: Remove the first voice command from the second voice command set.

[0349] S2209: Process complete.

[0350] During the process of deleting wake-up-free voice commands, the GUI for resolving conflicts between the second and third voice command sets is similar to the GUIs in Figures 14 and 15, and will not be described in detail here for the sake of simplicity.

[0351] For example, Table 1 shows the correspondence between instructions in different instruction sets and the operation of updating wake-up-free voice commands.

[0352] Table 1

[0353] Table 1 clearly shows what actions should be taken when updating the wake-up-free voice command when the first voice command belongs to a different set of voice commands.

[0354] For example, if the first voice command is in the first voice command set but not in the second or third voice command set, the voice command can be directly added to the second or third voice command set.

[0355] For example, if the first voice command is not in the first voice command set but is in the second voice command set, adding the first voice command set to the second voice command set can prompt the user that the command already exists; if adding the first voice command set to the third voice command set, the first voice command needs to be deleted from the second voice command set first.

[0356] Optionally, method 300 further includes: displaying a management interface for a set of wake-free voice commands, the management interface displaying an updated set of wake-free voice commands. Figures 24-34 illustrate a set of GUIs for managing wake-free voice commands provided in embodiments of this application.

[0357] As shown in Figure 24, this interface is the intelligent voice assistant settings interface, including a voice wake-up control, an intelligent wake-up-free control, a wake-up-free management control 2401, and a dialect setting control. When the user clicks the wake-up-free voice command management control 2401, the GUI shown in Figure 25 can be displayed on the central control screen.

[0358] As shown in Figure 25, this interface is a wake-free management interface, including: a wake-free voice command list management control 2501, a reset control, a new command management control 2502, and a delete command management control 2503. When user input is detected by clicking the wake-free voice command list management control 2501, the GUI shown in Figure 26 can be displayed on the central control screen; when user input is detected by clicking the new command management control 2502, the GUI shown in Figure 30 can be displayed on the central control screen; when user input is detected by clicking the delete command management control 2503, the GUI shown in Figure 32 can be displayed on the central control screen.

[0359] As shown in Figure 26, this interface is the management interface for the wake-up-free voice command list. Newly added wake-up-free voice commands can be displayed on this interface through wake-up-free voice command cards. This interface includes a new control 2601, a search box, a search control 2602, wake-up-free voice command cards, and a page number control. The wake-up-free voice command cards include a delete control. The new control 2601 can be used to add new wake-up-free voice commands. The search control 2602 can search for commands related to the search terms in the search box, including already added wake-up-free voice commands, already deleted wake-up-free voice commands, and wake-up-free voice commands that the user may want to use. The delete control in the wake-up-free voice command card is used to delete the wake-up-free voice command displayed on that card. The page number control is used to display the content of different pages.

[0360] When the user clicks on the new control 2601, the GUI shown in Figure 27 can be displayed on the central control screen. As shown in Figure 27, this interface is for manually adding wake-up-free voice commands. This interface includes: keyboard controls, add command controls, and complete controls. The user can click the add command control to activate the keyboard controls, input the wake-up-free voice command to be added using the keyboard controls, and finally click the complete control to add the wake-up-free voice command.

[0361] As shown in Figure 28, when the search term "library" is entered in the search box of the wake-free voice command list management interface, and the user clicks on the search control 2602, the GUI shown in Figure 29 can be displayed on the central control screen. Figure 29 shows the relevant search results prompts for "library" and specific wake-free voice command cards. The relevant search results can include at least one or more of the following commands:

[0362] (1) This instruction is in the first set of voice instructions;

[0363] (2) This instruction is in the second set of voice instructions;

[0364] (3) This instruction is in the third set of voice instructions;

[0365] (4) The instruction is not in the first set of voice instructions, nor in the second set of voice instructions, nor in the third set of voice instructions, but the user has used the instruction before.

[0366] (5) The instruction is not in the first set of voice instructions, nor in the second set of voice instructions, nor in the third set of voice instructions, but the instruction is recommended for use.

[0367] Each wake-up-free voice command card contains a "+" control 2901 and a "-" control 2902. The "+" control is used to indicate the wake-up-free voice command shown on the newly added card, and the "-" control is used to indicate the wake-up-free voice command shown on the deleted card.

[0368] For example, when a user clicks the "+" control 2901 on the wake-free voice command card "Navigate to Library", it means that the user wants to add a wake-free voice command for "Navigate to Library". In response to the user's input of clicking the "+" control 2901, the command "Navigate to Library" can be added to the second set of voice commands.

[0369] If the command "Navigate to the library" is already in the second set of voice commands, the GUI shown in Figure 13 can be displayed on the central control screen to remind the user that there is no need to add a new wake-up-free voice command. If the command "Navigate to the library" is already in the third set of voice commands, the GUI shown in Figure 14 can be displayed on the central control screen to remind the user that the command has been deleted and ask the user to confirm whether they want to add a new wake-up-free voice command.

[0370] For example, when a user clicks the "-" control 2902 of the wake-free voice command card "Navigate to Library", it indicates that the user wants to delete the wake-free voice command "Navigate to Library". In response to the user's input of clicking the "-" control 2902, the command "Navigate to Library" can be added to the third voice command set.

[0371] If the command is already in the third voice command set, the GUI shown in Figure 23 can be displayed on the central control screen to remind the user that there is no need to delete the wake-up-free voice command repeatedly; if the command "navigate to library" is already in the second voice command set, it can be displayed with reference to Figure 14 and the corresponding description, which will not be elaborated here.

[0372] As shown in Figure 30, this interface is the management interface for adding commands, including: a manual input control 3001, a search box, a search control 3002 for adding commands, a wake-up-free voice command card, and a page number control. The wake-up-free voice command card includes a delete control. The descriptions of the wake-up-free voice command card, delete control, and page number control can be found in the corresponding descriptions in Figure 26, and will not be repeated here. When the user clicks the manual input control 3001, the GUI shown in Figure 27 can be displayed on the central control screen; when the search term "map" is entered in the search box, and the user clicks the search control 3002 for adding commands, the GUI shown in Figure 31 can be displayed on the central control screen.

[0373] Figure 31 shows the search results pop-up for new commands related to "Map" and the specific wake-up-free voice command cards. These cards correspond to newly added wake-up-free voice commands. Each wake-up-free voice command card includes a "-" control, which can be used to delete newly added wake-up-free voice commands.

[0374] Besides adding commands manually, another implementation method is to add commands via voice input. For example, a user can say, "Hey A, set [Report Financial News] as a wake-up-free voice command," thus adding a wake-up-free voice command for "Report Financial News."

[0375] As shown in Figure 32, this interface is the management interface for deletion commands. The latest deleted wake-up-free voice commands can be displayed on the current interface through wake-up-free voice command cards. This interface includes: a manual input deletion control 3201, a search box, a deletion command search control 3202, wake-up-free voice command cards, and a page number control. The wake-up-free voice command cards include a restore control, which is used to add deleted wake-up-free voice commands to the wake-up-free voice command set. For example, the command "Turn on the heater" is in the third voice command set. In response to the user clicking the restore control, the command "Turn on the heater" can be deleted from the third voice command set and added to the second voice command set.

[0376] When the user clicks the manual input delete control 3201, the GUI shown in Figure 33 can be displayed on the central control screen. The content shown in Figure 33 can be referred to the description in Figure 27, and will not be repeated here.

[0377] When the search term "steering wheel" is entered into the search box in Figure 32, and the user clicks the search control 3202 for the deletion command, the GUI shown in Figure 34 can be displayed on the central control screen. Figure 34 shows the search result prompt box for the deletion command related to "steering wheel" and the specific wake-up-free voice command cards. The commands corresponding to these cards are all deleted wake-up-free voice commands. Each wake-up-free voice command card contains a "+" control, which can be used to restore the deleted wake-up-free voice command, that is, to add the deleted command to the wake-up-free voice command set.

[0378] It should be noted that the method for sorting the wake-up-free voice command cards in this application embodiment is not limited. For example, the wake-up-free voice command cards can be sorted according to the frequency of use of different commands, or according to the order of modification time.

[0379] Figure 35 shows a schematic block diagram of a voice interaction device 3500 provided in an embodiment of this application. The device includes: an acquisition unit 3510, configured to: acquire a first voice command; and an update unit 3520, configured to: update the wake-up-free voice command set based on the usage frequency and a first recognition frequency of the first voice command when the first voice command is not in the wake-up-free voice command set, wherein the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state.

[0380] Optionally, the update unit 3520 is further configured to: update the wake-up-free voice command set based on the cancellation frequency of the first voice command when the first voice command is a command in the wake-up-free voice command set.

[0381] Optionally, the update unit 3520 is further configured to: set the first voice command as a command in the wake-up-free voice command set when the usage frequency is greater than or equal to the first preset threshold and the first recognition frequency is less than or equal to the second preset threshold.

[0382] Optionally, the device further includes a control unit 3530, which is configured to: when the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold, control the prompting device to prompt the user to add the first voice command to the wake-up-free voice command set; the updating unit 3520 is further configured to: in response to receiving input from the user confirming that the first voice command has been added to the wake-up-free voice command set, set the first voice command as a command in the wake-up-free voice command set.

[0383] Optionally, the wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. The updating unit is further configured to: set the first voice command as an instruction in the second voice command set.

[0384] Optionally, the update unit is also used to: delete the first voice command from the set of non-wake-up-free voice commands.

[0385] Optionally, the device further includes a sending unit 3540 and a receiving unit 3550. The sending unit 3540 is configured to: send a first voice command and a text recognition result of the first voice command to the cloud; the receiving unit 3550 is configured to: receive a verification result sent from the cloud, the verification result indicating whether the text recognition result is accurate and whether the first voice command is a command issued by the user to the vehicle; the acquisition unit 3510 is configured to: acquire a second recognition frequency based on the first recognition frequency and the verification result; and the updating unit 3520 is further configured to: update the wake-up-free voice command set based on the usage frequency and the second recognition frequency.

[0386] Optionally, the update unit 3520 is further configured to: delete the first voice command from the wake-up-free voice command set when the cancellation frequency is greater than or equal to a third preset threshold.

[0387] Optionally, the device further includes a control unit 3530, which is configured to: prompt the user to delete the first voice command when the cancellation frequency is greater than or equal to a third preset threshold; the update unit 3520 is further configured to: delete the first voice command from the wake-up-free voice command set in response to receiving input from the user confirming the deletion of the first voice command.

[0388] Optionally, the update unit 3520 is further configured to: set the first voice command as an instruction in the set of non-wake-up-free voice commands.

[0389] Optionally, the wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set is not allowed to be updated, while the second voice command set is allowed to be updated. The update unit 3520 is also used to delete the first voice command from the second voice command set.

[0390] Optionally, the device further includes a processing unit 3560, which is used to: perform standardization processing on the first voice command, the standardization processing including at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

[0391] Optionally, the device further includes a display unit 3570, which is used to display a management interface for a set of wake-up-free voice commands, the management interface displaying an updated set of wake-up-free voice commands.

[0392] It should be understood that the division of units in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units in the device can be implemented by a processor calling software; for example, the device includes a processor connected to memory, which stores instructions. The processor calls the instructions stored in memory to implement any of the above methods or to implement the functions of each unit in the device. The processor can be, for example, a general-purpose processor, such as a CPU or microprocessor, and the memory can be internal or external to the device. Alternatively, the units in the device can be implemented as hardware circuits. The functions of some or all units can be implemented through the design of the hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all units are implemented through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a PLD, such as an FPGA, which can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby implementing the functions of some or all units. All units of the above devices can be implemented entirely through processor calling software, or entirely through hardware circuits, or partially through processor calling software with the remaining parts implemented through hardware circuits.

[0393] In this application embodiment, a processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction reading and execution capabilities, such as a CPU, microprocessor, GPU, or DSP. In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented as an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the processor loading instructions to implement the functions of some or all of the above units. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, or DPU.

[0394] As can be seen, each unit in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0395] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together as a System-on-a-Chip (SoC). The SoC may include at least one processor for implementing any of the above methods or implementing the functions of the units in the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and AI processor, CPU and GPU, etc.

[0396] This application also provides a voice interaction device, which includes a memory for storing a computer program and a processor for executing the computer program stored in the memory, so that the device performs the methods or steps described in the above embodiments.

[0397] Optionally, if the voice interaction device is located in a vehicle, the processor may be the processor 121-12n shown in Figure 1.

[0398] This application also provides a voice interaction system, which may include a microphone and a computing platform, the computing platform including the aforementioned voice interaction device 3500.

[0399] This application also provides a vehicle that may include the above-described voice interaction device 3500 or the above-described voice interaction system.

[0400] This application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to perform the methods described in the above embodiments.

[0401] This application also provides a chip, which includes a circuit for performing the methods described in the above embodiments.

[0402] In implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software. The method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or by a combination of hardware and software modules within the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, power-on erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.

[0403] It should be understood that in the embodiments of this application, the memory may include read-only memory and random access memory, and provides instructions and data to the processor.

[0404] It should also be understood that, in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0405] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0406] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0407] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0408] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0409] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0410] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0411] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be covered. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A voice interaction method, characterized in that, The method includes: Obtain the first voice command; If the first voice command is not a command in the wake-free voice command set, the wake-free voice command set is updated based on the usage frequency of the first voice command and the first recognition frequency, where the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state.

2. The method according to claim 1, characterized in that, The method further includes: If the first voice command is an instruction in the set of wake-up-free voice commands, the set of wake-up-free voice commands is updated based on the cancellation frequency of the first voice command.

3. The method according to claim 1, characterized in that, The step of updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command includes: When the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold, the first voice command is set as a command in the set of wake-up-free voice commands.

4. The method according to claim 3, characterized in that, The step of updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command includes: When the usage frequency is greater than or equal to the first preset threshold and the first recognition frequency is less than or equal to the second preset threshold, the control prompting device prompts the user to add the first voice command to the wake-up-free voice command set; In response to receiving user confirmation to add the first voice command to the wake-up-free voice command set, the first voice command is set as a command in the wake-up-free voice command set.

5. The method according to claim 3 or 4, characterized in that, The wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. Setting the first voice command as an instruction in the wake-free voice command set includes: Set the first voice command as a command in the second set of voice commands.

6. The method according to claim 5, characterized in that, Before setting the first voice command as an instruction in the second voice command set, the method further includes: Remove the first voice command from the set of non-wake-up-free voice commands.

7. The method according to any one of claims 1, 3-6, characterized in that, Before updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command, the method further includes: Send the first voice command and the text recognition result of the first voice command to the cloud; The system receives a verification result sent from the cloud, which is used to indicate whether the text recognition result is accurate and whether the first voice command is a command issued by the user to the vehicle. The step of updating the wake-free voice command set based on the usage frequency and recognition frequency of the first voice command includes: Based on the first identification frequency and the verification result, a second identification frequency is obtained; The set of wake-up-free voice commands is updated based on the usage frequency and the second recognition frequency.

8. The method according to claim 2, characterized in that, The step of updating the wake-up-free voice command set based on the cancellation frequency of the first voice command includes: If the cancellation frequency is greater than or equal to a third preset threshold, the first voice command is deleted from the set of wake-up-free voice commands.

9. The method according to claim 8, characterized in that, The step of updating the wake-up-free voice command set based on the cancellation frequency of the first voice command includes: If the cancellation frequency is greater than or equal to the third preset threshold, the control prompting device prompts the user to delete the first voice command; In response to receiving user confirmation to delete the first voice command, the first voice command is deleted from the set of wake-up-free voice commands.

10. The method according to claim 8 or 9, characterized in that, Deleting the first voice command from the set of wake-up-free voice commands includes: Set the first voice command to a command from the set of non-wake-up-free voice commands.

11. The method according to claim 10, characterized in that, The wake-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. Before setting the first voice command as an instruction not in the wake-free voice command set, the method further includes: Delete the first voice command from the second set of voice commands.

12. The method according to any one of claims 1-11, characterized in that, The method further includes: The first voice command is standardized, and the standardization process includes at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

13. The method according to any one of claims 1-12, characterized in that, The method further includes: The management interface of the wake-up-free voice command set is displayed, and the management interface displays the updated wake-up-free voice command set.

14. A voice interaction device, characterized in that, include: Acquisition unit, used for: acquiring the first voice command; The updating unit is configured to: update the wake-up-free voice command set based on the usage frequency and the first recognition frequency of the first voice command when the first voice command is not a command in the wake-up-free voice command set, wherein the first recognition frequency is the frequency at which the vehicle recognizes the first voice command in a non-wake-up state.

15. The apparatus according to claim 14, characterized in that, The updating unit is further configured to: update the wake-up-free voice command set based on the cancellation frequency of the first voice command when the first voice command is an instruction in the wake-up-free voice command set.

16. The apparatus according to claim 14, characterized in that, The updating unit is further configured to: set the first voice instruction as an instruction in the set of wake-up-free voice instructions when the usage frequency is greater than or equal to a first preset threshold and the first recognition frequency is less than or equal to a second preset threshold.

17. The apparatus according to claim 16, characterized in that, The device also includes a control unit. The control unit is configured to: when the usage frequency is greater than or equal to the first preset threshold and the first recognition frequency is less than or equal to the second preset threshold, control the prompting device to prompt the user to add the first voice command to the wake-up-free voice command; The updating unit is further configured to: in response to receiving input from the user confirming the addition of the first voice command to the wake-up-free voice command set, set the first voice command as a command in the wake-up-free voice command set.

18. The apparatus according to claim 16 or 17, characterized in that, The wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. The updating unit is further configured to: set the first voice instruction as an instruction in the second voice instruction set.

19. The apparatus according to claim 18, characterized in that, The updating unit is further configured to: delete the first voice command from the set of non-wake-up-free voice commands.

20. The apparatus according to any one of claims 14, 16-19, characterized in that, The device further includes a transmitting unit and a receiving unit. The sending unit is configured to: send the first voice command and the text recognition result of the first voice command to the cloud; The receiving unit is configured to: receive a verification result sent from the cloud, the verification result being used to indicate whether the text recognition result is accurate and whether the first voice command is a command issued by the user to the vehicle; The acquisition unit is configured to: acquire a second identification frequency based on the first identification frequency and the verification result; The updating unit is further configured to: update the wake-up-free voice command set based on the usage frequency and the second recognition frequency.

21. The apparatus according to claim 15, characterized in that, The update unit is further configured to: delete the first voice command from the wake-up-free voice command set when the cancellation frequency is greater than or equal to a third preset threshold.

22. The apparatus according to claim 21, characterized in that, The device also includes a control unit. The control unit is configured to: when the cancellation frequency is greater than or equal to the third preset threshold, control the prompting device to prompt the user to delete the first voice command; The updating unit is further configured to: delete the first voice command from the wake-free voice command set in response to receiving input from the user confirming the deletion of the first voice command.

23. The apparatus according to claim 21 or 22, characterized in that, The updating unit is further configured to: set the first voice instruction as an instruction in the set of non-wake-up-free voice instructions.

24. The apparatus according to claim 23, characterized in that, The wake-up-free voice command set includes a first voice command set and a second voice command set. The first voice command set cannot be updated, while the second voice command set can be updated. The updating unit is further configured to: delete the first voice instruction from the second voice instruction set.

25. The apparatus according to any one of claims 14-24, characterized in that, The device also includes a processing unit. The processing unit is configured to: perform standardization processing on the first voice command, wherein the standardization processing includes at least one of the following: removing interjections, removing redundant information, correcting typos, and adjusting word order.

26. The apparatus according to any one of claims 14-25, characterized in that, The device also includes a display unit. The display unit is used to: display the management interface of the wake-up-free voice command set, wherein the management interface displays the updated wake-up-free voice command set.

27. A voice interaction device, characterized in that, include: Memory, used to store computer programs; A processor for executing a computer program stored in the memory to cause the apparatus to perform the method as described in any one of claims 1 to 13.

28. A voice interaction system, characterized in that, It includes a microphone and a computing platform, the computing platform including the means as described in any one of claims 14 to 27.

29. A vehicle, characterized in that, Includes the apparatus as described in any one of claims 14 to 27, or the system as described in claim 28.

30. A computer-readable storage medium, characterized in that, It stores instructions that, when executed by a processor, cause the processor to implement the method as described in any one of claims 1 to 13.

31. A computer program product, characterized in that, The computer program product includes computer program code that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 13.

32. A chip, characterized in that, The chip includes circuitry for performing the method as described in any one of claims 1 to 13.

Images