A vehicle-mounted voice automatic control method and system simulating manual interaction

Through the automatic vehicle voice control method that simulates manual interaction, the use of voice commands to generate executable node flows, the problems of high cost of vehicle voice control and difficulty in updating in the existing technology are solved, full function coverage and continuous updates are achieved, and the product is improved Personalization and intelligence.

CN113553021BActive Publication Date: 2024-11-12HUIZHOU DESAY SV AUTOMOTIVE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110659613.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-06-15
Publication Date
2024-11-12
Estimated Expiration
2041-06-15

AI Technical Summary

Technical Problem

The existing vehicle voice control system relies on open or customized APIs for various ecological applications, resulting in cost surges and difficulty in updating functions, and it is impossible to achieve differentiated experience and timely updates.

Method used

Using the automatic vehicle voice control method that simulates manual interaction, the user input voice commands can obtain responsive voice skills, generate executable node flow, and execute according to preset interaction strategies, get rid of the customization of third-party ecological application interfaces, and realize sub-functions in each application. Automatic control and update.

Benefits of technology

It reduces the cost of using ecological applications, shortens the development cycle of product voice functions, realizes full function coverage and continuous update of on-board voice control, and improves the personalized experience and intelligence of the product.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113553021B_ABST
    Figure CN113553021B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of voice control technology, and in particular to a vehicle-mounted voice automatic control method and system for simulating manual interaction. An arbitration module, a script parsing module, a node flow generation module, and an execution module are arranged to form a simulated manual interaction mechanism. In order to be compatible with different ecological applications, the most basic script is used as the starting point. After obtaining the user's voice command, when it is further judged that the corresponding response voice skill is a voice action skill, the vehicle-mounted voice automatic control process is converted into the execution of the executable node flow simulating manual interaction by generating an executable node flow by obtaining the corresponding voice action script and executing it according to a preset interaction strategy. This not only makes the vehicle-mounted voice control free from the limitations of third-party ecological application interface customization, but also realizes the coverage of automatic control of any sub-function in each application, greatly reducing the use cost of ecological applications and shortening the development cycle of product voice functions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of voice control technology, and in particular to a vehicle-mounted voice automatic control method and system simulating manual interaction. Background Art

[0002] The in-vehicle voice control systems that have been launched on the market have the following problems:

[0003] First, the existing in-vehicle voice control system must rely on the corresponding control interface provided by each application to realize the control of the application. With the enrichment and openness of the application ecosystem of the Internet of Vehicles, a large number of Internet of Vehicles ecological applications (such as APPs, quick applications, mini programs, etc.) have been installed in the car, but not every application has a ready-made or customized voice control interface, so the in-vehicle voice cannot realize voice control of all applications;

[0004] Second, even if ecosystem application vendors are willing to customize voice interfaces, since the application includes a large number of sub-functions, it will take a lot of time and increase production costs to customize the interfaces of all sub-functions in the application. That is, voice customized interfaces will bring high costs to terminal manufacturers, and there is also a long interface customization cycle, which will lead to a significant increase in product time and material costs;

[0005] 3. Terminal manufacturers have different system application configuration requirements and different ecological requirements for voice control. If voice control is only implemented by relying on the common interfaces opened by various ecological applications, it will be difficult to achieve a differentiated experience of in-vehicle voice.

[0006] 4. The update of open interfaces of ecological applications is slow and cannot meet the demand for in-vehicle voice update functions in a short period of time.

[0007] Therefore, the existing vehicle voice control system is in urgent need of improvement. Summary of the invention

[0008] The present invention provides a method and system for automatic vehicle voice control that simulates manual interaction, which solves the technical problem that the current vehicle voice control relies on the open or customized APIs of various ecological applications to realize the voice control of corresponding functions, but the customized API will lead to a surge in the cost of related products, and the differences among various ecological applications lead to the inability to update and cover the voice function in a timely manner.

[0009] In order to solve the above technical problems, the present invention provides a vehicle-mounted voice automatic control method simulating manual interaction, comprising the steps of:

[0010] S1. Obtaining a response voice skill according to the voice command input by the user;

[0011] S2. When it is determined that the response voice skill is a voice action skill, obtaining a corresponding voice action script according to user preference settings;

[0012] S3, generating an executable node flow according to the voice action script;

[0013] S4. Execute the executable node flow according to a preset interaction strategy.

[0014] This basic solution is compatible with different ecological applications, so it starts with the most basic script. After obtaining the user's voice command, it further determines that the corresponding response voice skill is a voice action skill. By obtaining the corresponding voice action script to generate an executable node flow and executing it according to the preset interaction strategy, the vehicle-mounted voice automatic control process is converted into the execution of an executable node flow that simulates manual interaction. This not only makes the in-vehicle voice control free from the limitations of third-party ecological application interface customization, but also realizes the coverage of automatic control of any sub-functions in each application, greatly reducing the cost of using ecological applications and shortening the development cycle of product voice functions.

[0015] In a further embodiment, step S4 comprises the steps of:

[0016] S41, sequentially loading each executable node in the executable node stream;

[0017] S42, obtaining the current node type, determining whether the currently executed executable node is a dialog node, if so, proceeding to the next step, otherwise executing the executable node;

[0018] S43, entering a voice multi-round dialogue mode according to the executable node, and obtaining a user's selection instruction;

[0019] S44. Update the node information and node type of the currently executed executable node according to the selection instruction, and return to step S42 until the executable node flow is executed.

[0020] This solution takes the preset interaction strategy as the core, and executes each executable node in the executable node flow sequentially. When it is determined that the executable node requires voice interaction, it automatically enters the voice multi-round dialogue mode to obtain the user's selection instructions, so as to simulate the user's manual selection of each sub-function of each ecological application. In this way, while fully considering the user's intentions, it covers the sub-functions of all ecological applications and eliminates the interface customization of each sub-function of each ecological application, so that terminal manufacturers have a wider range and more flexibility in choosing voice ecology, which greatly saves third-party customization costs and shortens the adaptation cycle of third-party ecological applications.

[0021] In a further embodiment, step S1 comprises the steps of:

[0022] S11, obtaining a voice command input by a user and recognizing and obtaining text information;

[0023] S12: Determine the user intention according to the text information, and then match the voice skill library according to the user intention to obtain a response voice skill.

[0024] In a further embodiment, the present invention further comprises the steps of:

[0025] S01. When a new ecological application is detected, edit the corresponding voice action script according to the sub-function of the new ecological application and create a new version of script data;

[0026] S02: Obtain the new version script data and update the local script library.

[0027] This solution coordinates the overall situation through the background. When a new ecological application is detected, the corresponding voice action script is edited according to the sub-functions of the new ecological application and a new version of the script data is produced. It is supplied to each application terminal as update data. The application terminal only needs to update the new version of the script data to be compatible with the new ecological application, thereby reducing the production cost of third-party applications (reducing the cost of interface customization) and the limitations of application terminal use, and realizing a full-function coverage closed-loop experience of voice control of the vehicle system.

[0028] In a further embodiment, the present invention further comprises the steps of:

[0029] S5. When the executable node flow is executed abnormally, the corresponding new version script data is obtained and the local script library is updated.

[0030] When an exception occurs during the execution of an executable node flow, this solution can autonomously update the data by obtaining the corresponding new version of the script data and updating the local script library, so that the same response voice skill can be executed smoothly next time (the executable node flow can be re-executed).

[0031] The present invention provides a vehicle-mounted voice automatic control system simulating manual interaction, comprising a voice recognition module and a voice action engine; the voice action engine comprises an arbitration module, a script parsing module, a node flow generation module, and an execution module;

[0032] The speech recognition module is used to obtain the voice command input by the user and recognize the user's intention, and obtain the response voice skill according to the user's intention;

[0033] The arbitration module is used to determine whether the response voice skill is a voice action skill, and obtain the corresponding voice action script from the local script library according to the user preference setting;

[0034] The script parsing module is used to parse the script sequence of the voice action skill;

[0035] The node flow generation module is used to generate an executable node flow according to the script sequence;

[0036] The execution module is used to execute the executable node flow according to a preset interaction strategy.

[0037] This basic solution sets up an arbitration module, a script parsing module, a node flow generation module, and an execution module to form a simulated manual interaction mechanism, which provides a physical basis for the above-mentioned vehicle-mounted voice automatic control method that simulates manual interaction. The script parsing module and node flow generation module of the voice control system cooperate with the arbitration module to replace the customized interface of each ecological application. The recognized voice action skills are converted into voice action scripts and executable node flows in turn, and finally the executable nodes are executed to simulate the user's manual selection of each sub-function of each ecological application, thereby getting rid of the limitations of third-party ecological application interface customization, and realizing the coverage of automatic control of any sub-function in each application, greatly reducing the cost of using ecological applications and shortening the development cycle of product voice functions.

[0038] In a further implementation scheme, the present invention also includes a background management platform, which is used to edit the corresponding voice action script according to the sub-functions of the new ecological application when a new ecological application is detected, and to create a new version of the script data; it is also used to send push information to each of the voice action engines according to the new version of the script data.

[0039] In a further implementation scheme, the voice action engine also includes an update module, and the update module is used to download the new version script data after obtaining the push message from the background management platform and the user's update opinion.

[0040] This solution converts the pressure of customized interfaces of vehicle-mounted terminal ecological applications into updates of new version script data of the background management platform. It can not only significantly reduce the adaptation cycle of ecological applications and vehicle-mounted terminals, but also realize the continuous update and upgrade of the voice control function of vehicle-mounted terminals to enhance the personalized experience and intelligence of the product, while also having strong practicality.

[0041] In a further embodiment, the voice action engine further comprises an exception reporting module, wherein the exception reporting module is used to record exception information and report to the background management platform when the executable node flow executes abnormally;

[0042] In a further implementation scheme, the background management platform is also used to match the corresponding new version script data according to the abnormal information, and send push information to the corresponding voice action engine.

[0043] This solution sets up two methods: update push of the background management platform and autonomous discovery and update of the vehicle side (update module, exception reporting module). It solves the problem of automatic improvement and repair of exceptions sent during system application software upgrades or user use, thereby helping users to automatically upgrade the in-vehicle voice function experience and ensure the normal use of in-vehicle voice. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a vehicle-mounted voice automatic control method simulating manual interaction provided in Example 1 of the present invention;

[0045] Figure 2 This is a workflow diagram of a vehicle-mounted voice automatic control system simulating manual interaction provided in Example 1 of the present invention;

[0046] Figure 3 This is a system framework diagram of an in-vehicle voice automatic control system simulating manual interaction provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0047] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0048] Example 1

[0049] An embodiment of the present invention provides a vehicle-mounted voice automatic control method simulating manual interaction, such as Figure 1 , Figure 2 As shown, in this embodiment, steps S01, S02, S1-S5 are included:

[0050] S01. When a new ecological application is detected, the corresponding voice action script is edited according to the sub-function of the new ecological application and a new version of the script data is produced;

[0051] S02. Obtain new version script data and update the local script library.

[0052] This embodiment coordinates the overall situation through the background. When a new ecological application is detected, the corresponding voice action script is edited according to the sub-functions of the new ecological application and a new version of the script data is produced. The script data is supplied to each application terminal as update data. The application terminal only needs to update the new version of the script data to be compatible with the new ecological application, thereby reducing the production cost of third-party applications (reducing the cost of interface customization) and the limitations of application terminal use, and realizing a full-function coverage closed-loop experience of voice control of the vehicle system.

[0053] S1. Acquire a response voice skill according to a voice command input by a user, including steps S11 to S12:

[0054] S11, obtaining a voice command input by a user and recognizing and obtaining text information;

[0055] S12. Determine the user intention based on the text information, and then match the voice skill library based on the user intention to obtain a response voice skill.

[0056] S2. When it is determined that the response voice skill is a voice action skill, the corresponding voice action script is obtained according to the user preference setting; otherwise, the response voice skill is directly executed according to the general execution method (that is, when the voice skill is a common skill);

[0057] Among them, the user preference settings can be manually input by the user or the user habit preferences obtained based on system big data statistics.

[0058] S3, generating an executable node flow according to the voice action script;

[0059] S4, executing the executable node flow according to the preset interaction strategy, including steps S41 to S44:

[0060] S41, sequentially loading each executable node in the executable node stream;

[0061] S42, obtaining the current node type, determining whether the currently executed executable node is a dialog node, if so, proceeding to the next step, otherwise executing the executable node;

[0062] S43, entering a voice multi-round dialogue mode according to the executable node, and obtaining a user's selection instruction;

[0063] S44, updating the node information and node type of the currently executed executable node according to the selection instruction, and returning to step S42 until the executable node flow is executed. For example, after the current dialogue ends, after obtaining the selection instruction, the node type of the currently executed executable node is updated to a non-dialogue node.

[0064] This embodiment takes the preset interaction strategy as the core, and executes each executable node in the executable node flow in sequence. When it is determined that the executable node requires voice interaction, it automatically enters the voice multi-round dialogue mode to obtain the user's selection instructions, so as to simulate the user's manual selection of each sub-function of each ecological application. In this way, while fully considering the user's intentions, it covers the sub-functions of all ecological applications and eliminates the interface customization of each sub-function of each ecological application, so that terminal manufacturers have a wider range and more flexibility in choosing voice ecology, which greatly saves third-party customization costs and shortens the adaptation cycle of third-party ecological applications.

[0065] S5. When the executable node flow is executed abnormally, the corresponding new version script data is obtained and the local script library is updated.

[0066] In this embodiment, when an exception occurs during the execution of an executable node flow, the corresponding new version script data is obtained and the local script library is updated, and the data is updated autonomously, so that the same response voice skill can be executed smoothly next time (the executable node flow is re-executed).

[0067] In order to be compatible with different ecological applications, the embodiment of the present invention starts with the most basic script. After obtaining the user's voice command, when it is further determined that the corresponding response voice skill is a voice action skill, the corresponding voice action script is obtained to generate an executable node flow and execute it according to a preset interaction strategy. The vehicle-mounted voice automatic control process is converted into the execution of an executable node flow that simulates manual interaction. This not only frees the vehicle-mounted voice control from the limitations of third-party ecological application interface customization, but also realizes the coverage of automatic control of any sub-functions within each application, greatly reducing the cost of using ecological applications and shortening the development cycle of product voice functions.

[0068] Example 2

[0069] In the embodiments of the present invention, Figure 3 The reference numerals that appear include: speech recognition module 1; speech action engine 2, arbitration module 21, script parsing module 22, node flow generation module 23, execution module 24, update module 25, exception reporting module 26; background management platform 3.

[0070] The present invention provides a vehicle-mounted voice automatic control system simulating manual interaction, see Figure 3 , including speech recognition module, speech action engine and background management platform; speech action engine includes arbitration module, script parsing module, node flow generation module and execution module;

[0071] The speech recognition module is used to obtain the voice command input by the user and recognize the user's intention, and obtain the response voice skill according to the user's intention;

[0072] The arbitration module is used to determine whether the response voice skill is a voice action skill, and obtain the corresponding voice action script from the local script library according to the user's preference settings;

[0073] The script parsing module is used to parse the script sequence of voice action skills;

[0074] The node flow generation module is used to generate executable node flows according to script sequences;

[0075] The execution module is used to execute the executable node flow according to the preset interaction strategy.

[0076] In a further implementation scheme, the background management platform is used to edit the corresponding voice action script according to the sub-functions of the new ecological application when a new ecological application is detected, and to create a new version of the script data; it is also used to send push information to each voice action engine based on the new version of the script data.

[0077] In a further implementation scheme, the voice action engine also includes an update module, which is used to download the new version of script data after receiving a push message from the background management platform and obtaining user update opinions.

[0078] Specifically, after the backend management platform confirms the push range, it sends update push messages in batches. After receiving the update notification, the update module of the vehicle terminal confirms whether the local version is the latest version. If it is, it will be ignored. If not, it will prompt the user to find the new version script data and ask whether it needs to be updated. If the user confirms the update, the update module will download and update the version (new version script data) according to the new version information obtained in the previous step, and actively find that the update process is complete; if the user does not agree to the update, the update process will be canceled.

[0079] This embodiment converts the pressure of the customized interface of the vehicle terminal ecological application into the update of the new version of the script data of the background management platform, which can not only greatly reduce the adaptation cycle of the ecological application and the vehicle terminal, but also realize the continuous update and upgrade of the voice control function of the vehicle terminal to enhance the personalized experience and intelligence of the product, while having strong practicality.

[0080] In a further embodiment, the voice action engine further includes an exception reporting module, which is used to record exception information and report it to the background management platform when the executable node flow executes abnormally;

[0081] For example, if the vehicle terminal is connected to the Internet, the vehicle terminal will upload the above execution status and error information (abnormal information) to the background management platform in real time; if the vehicle terminal is offline, it will record (such as abnormal log) the above execution status and error information (abnormal information) and upload it after connecting to the Internet.

[0082] In a further implementation scheme, the backend management platform is also used to match the corresponding new version script data according to the exception information, and send push information to the corresponding voice action engine. Specifically, the backend management platform is also used to check whether there is a new version that can fix the current exception according to the exception information. If so, the new version and push information are sent to the update module to enter the data update process.

[0083] This embodiment provides two update push methods: the background management platform and the vehicle-side (update module, exception reporting module) autonomous discovery and update. It solves the problem of automatic improvement and repair of exceptions sent during system application software upgrades or user use, thereby helping users to automatically upgrade the in-vehicle voice function experience and ensure the normal use of in-vehicle voice.

[0084] The embodiment of the present invention sets an arbitration module, a script parsing module, a node flow generation module, and an execution module to form a simulated manual interaction mechanism, which provides a physical basis for the above-mentioned vehicle-mounted voice automatic control method simulating manual interaction, and uses the script parsing module and the node flow generation module of the voice control system to cooperate with the arbitration module to replace the customized interface of each ecological application. By converting the recognized voice action skills into voice action scripts and executable node flows in turn, and finally executing each executable node, the user's manual selection of each sub-function of each ecological application is simulated, thereby getting rid of the limitations of third-party ecological application interface customization, and realizing the coverage of automatic control of any sub-function in each application, greatly reducing the cost of using ecological applications and shortening the development cycle of product voice functions.

[0085] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A vehicle-mounted voice automatic control method simulating manual interaction, characterized in that: Includes steps: S1. Obtaining a response voice skill according to the voice command input by the user; S2. When it is determined that the response voice skill is a voice action skill, obtaining a corresponding voice action script according to user preference settings; S3, generating an executable node flow according to the voice action script; S4, executing the executable node flow according to a preset interaction strategy; Also includes the steps: S01. When a new ecological application is detected, edit the corresponding voice action script according to the sub-function of the new ecological application and create a new version of script data; S02: Obtain the new version script data and update the local script library.

2. The vehicle-mounted voice automatic control method simulating manual interaction as claimed in claim 1, characterized in that: The step S4 comprises the steps of: S41, sequentially loading each executable node in the executable node stream; S42, obtaining the current node type, determining whether the currently executed executable node is a dialog node, if so, proceeding to the next step, otherwise executing the executable node; S43, entering a voice multi-round dialogue mode according to the executable node, and obtaining a user's selection instruction; S44. Update the node information and node type of the currently executed executable node according to the selection instruction, and return to step S42 until the executable node flow is executed.

3. The vehicle-mounted voice automatic control method simulating manual interaction as claimed in claim 2, characterized in that: The step S1 comprises the steps of: S11, obtaining a voice command input by a user and recognizing and obtaining text information; S12: Determine the user intention according to the text information, and then match the voice skill library according to the user intention to obtain a response voice skill.

4. The vehicle-mounted voice automatic control method simulating manual interaction as claimed in claim 1, characterized in that: The method also includes the following steps: S5: when the executable node flow is executed abnormally, the corresponding new version script data is obtained and the local script library is updated.

5. A vehicle-mounted voice automatic control system simulating manual interaction, used to implement a vehicle-mounted voice automatic control method simulating manual interaction as claimed in any one of claims 1 to 4, characterized in that: It includes a speech recognition module and a speech action engine; the speech action engine includes an arbitration module, a script parsing module, a node flow generation module, and an execution module; The speech recognition module is used to obtain the voice command input by the user and recognize the user's intention, and obtain the response voice skill according to the user's intention; The arbitration module is used to determine whether the response voice skill is a voice action skill, and obtain the corresponding voice action script from the local script library according to the user preference setting; The script parsing module is used to parse the script sequence of the voice action skill; The node flow generation module is used to generate an executable node flow according to the script sequence; The execution module is used to execute the executable node flow according to a preset interaction strategy.

6. The vehicle-mounted voice automatic control system simulating manual interaction as claimed in claim 5, characterized in that: It also includes a background management platform, which is used to edit the corresponding voice action script according to the sub-functions of the newly added ecological application and create a new version of script data when a new ecological application is detected; It is also used to send push information to each of the voice action engines based on the new version script data.

7. The vehicle-mounted voice automatic control system simulating manual interaction according to claim 6, characterized in that: The voice action engine also includes an update module, which is used to download the new version script data after obtaining a push message from the background management platform and obtaining user update opinions.

8. The vehicle-mounted voice automatic control system simulating manual interaction as claimed in claim 6, characterized in that: The voice action engine further includes an exception reporting module, which is used to record exception information and report it to the background management platform when the executable node flow executes abnormally; The background management platform is also used to match the corresponding new version script data according to the abnormal information, and send push information to the corresponding voice action engine.