A voice recognition method and system for a vehicle-mounted terminal

Abstract

The application discloses a visible and speakable vehicle-mounted terminal voice recognition method and system, wherein the method comprises the following steps: judging whether each page control in the vehicle-mounted terminal has a corresponding text mark, and setting a text mark for the page control if the page control does not have the corresponding text mark; judging whether the existing text mark has a conflict or an incomplete information phenomenon if the page control has the corresponding text mark, and correcting the text mark if the conflict or the incomplete information phenomenon exists; obtaining a voice signal to be recognized, removing noise from the voice signal to be recognized, and converting the voice signal after noise removal into text data; labeling all the page controls in sequence from left to right and from top to bottom by using positive integers; recognizing the text data, and controlling the page controls with corresponding labels according to the recognition result.

Description

Technical Field

[0001] This invention relates to the field of speech recognition technology, and in particular to a vehicle-mounted terminal speech recognition method and system that allows users to speak as soon as they see it. Background Technology

[0002] The statements in this section merely refer to the background art related to this invention and do not necessarily constitute prior art.

[0003] Voice communication is the most direct and effective way for people to communicate, and voice recognition technology enables simple and efficient information transmission between humans and machines. Currently, voice recognition technology has penetrated all aspects of our lives, such as mobile phone voice input methods, voice assistants, and voice search applications.

[0004] In the realm of intelligent mobility, voice technology also holds significant value and is a necessity in the automotive sector. From early voice navigation to today's in-vehicle voice control systems, these systems offer a variety of new interaction methods, including vehicle control, social interaction, and entertainment. This frees the driver's attention from complex settings and buttons, enhancing the driving experience and, to some extent, improving driving safety. Unlike traditional in-vehicle systems that rely on buttons or screens for operation, multimodal fusion detection, intelligent voice interaction, and multi-screen interactive gesture controls will become standard features in the next generation of intelligent cockpits. Because the in-vehicle environment is relatively stable, the speech data is less scattered, resulting in high voice recognition rates; therefore, the in-vehicle cabin is an ideal scenario for deploying voice interaction.

[0005] As can be seen and said, it refers to the ability of manually clickable controls within the system to achieve the effect of simulating clicks on controls by speaking, based on the accessibility service capabilities of the Android system and combined with the recognition of semantic hot words, without affecting the original implementation logic of the application APP.

[0006] It is evident that the voice link of the Talkable Voice Platform differs from traditional voice interaction methods. It does not support complex dialogue management, intent recognition, and other capabilities. The scope of semantic understanding is limited to the words defined on the page. Through hot word association, it can be understood as an end-to-end connection method. Its advantage lies in the ability to quickly establish a voice channel and control more elements on the page through voice. The corresponding disadvantage is that it can only manipulate existing elements on the page and does not support complex semantic understanding capabilities.

[0007] The inventors discovered that the existing technology has a technical flaw: if the text, images, switches, buttons, sliders and other controls appearing in the vehicle's infotainment system are not labeled and customized in advance, the driver cannot use voice to control the content played on the page. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a vehicle-mounted terminal voice recognition method and system that allows users to speak as soon as they see the screen. Based on the original voice recognition scheme, the invention customizes and expands the controls such as text, images, switches, buttons, and sliders that appear on the vehicle's infotainment system, enabling voice control functionality that allows users to control the screen as soon as they see it.

[0009] On the one hand, a visible-to-speak in-vehicle terminal voice recognition method is provided, including:

[0010] Determine if each page control in the vehicle terminal has a corresponding text identifier. If not, set a text identifier for the page control. If it does, determine if there are any conflicts or incomplete information in the existing text identifiers. If so, correct the text identifiers for the conflicts or incomplete information.

[0011] The process involves acquiring the speech signal to be recognized, removing noise from the speech signal, and converting the denoised speech signal into text data.

[0012] All page controls are numbered with positive integers in a left-to-right, top-to-bottom order; text data is recognized, and the corresponding page controls are controlled according to the recognition results;

[0013] In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

[0014] On the other hand, a visible and speakable in-vehicle terminal voice recognition system is provided, including:

[0015] The judgment module is configured to: determine whether each page control in the vehicle terminal has a corresponding text identifier; if not, set a text identifier for the page control; if so, determine whether there is a conflict or incomplete information in the existing text identifier; if so, correct the text identifier for the conflict or incomplete information.

[0016] The acquisition module is configured to: acquire the speech signal to be recognized, remove noise from the speech signal to be recognized, and convert the denoised speech signal into text data.

[0017] The recognition control module is configured to: number all page controls using positive integers in a left-to-right, top-to-bottom order; recognize text data and control the corresponding numbered page controls according to the recognition results;

[0018] In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

[0019] Furthermore, an electronic device is also provided, including:

[0020] Memory, used for non-transitory storage of computer-readable instructions; and

[0021] Processor, for executing the computer-readable instructions,

[0022] When the computer-readable instructions are executed by the processor, they perform the method described in the first aspect above.

[0023] In another aspect, a storage medium is also provided for non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method described in the first aspect are executed.

[0024] In another aspect, a computer program product is also provided, including a computer program that, when run on one or more processors, is used to implement the method described in the first aspect above.

[0025] One of the above technical solutions has the following advantages or beneficial effects:

[0026] By marking and locating the various apps displayed on the in-vehicle infotainment system, users can control the playback interface, select the previous or next song, navigate to the next level, or return to the main page via voice commands. This eliminates the need for manual operation, improving convenience and enhancing the user experience.

[0027] By controlling the vehicle's infotainment system through voice recognition, users can access functions visible to the naked eye, such as setting up and closing multimedia systems, clicking buttons, and swiping left, right, and up, thereby improving the convenience of cockpit control, enhancing user experience, and increasing customer satisfaction. Attached Figure Description

[0028] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0029] Figure 1 This is a flowchart of the method in Example 1. Detailed Implementation

[0030] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0031] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0032] All data acquisition in this embodiment is carried out in accordance with laws and regulations and with user consent, and the data is used legally.

[0033] Speech recognition technology is a comprehensive discipline that integrates various knowledge bases. In recent years, speech recognition technology has made significant progress, evolving from the recognition of small vocabularies and isolated words in the laboratory to the recognition of large vocabularies and speaker-independent continuous speech.

[0034] Voice recognition has a wide range of applications, the most common being voice input systems. Compared to keyboard input, it is more in line with people's daily habits and is more natural and efficient. Voice control of devices is also faster and more convenient than manual control.

[0035] Example 1

[0036] This embodiment provides a vehicle-mounted terminal voice recognition method that is both visible and speakable;

[0037] like Figure 1 As shown, a visible-to-speak in-vehicle terminal voice recognition method includes:

[0038] S101: Determine whether each page control in the vehicle terminal has a corresponding text identifier. If not, set a text identifier for the page control. If it does, determine whether there is a conflict or incomplete information in the existing text identifier. If so, correct the text identifier for the conflict or incomplete information.

[0039] S102: Acquire the speech signal to be recognized, remove noise from the speech signal to be recognized, and convert the denoised speech signal into text data;

[0040] S103: Number all page controls using positive integers in a left-to-right, top-to-bottom order; recognize text data and control the corresponding numbered page controls according to the recognition results;

[0041] In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

[0042] Furthermore, the page controls include: text controls, switch controls, button controls, and slider controls.

[0043] Furthermore, the text identifiers include: switches, titles, drag bars, and sliding components.

[0044] Furthermore, text labels are set for page controls, specifically including: setting text labels for switch buttons, such as "Bluetooth switch" or "WIFI switch", which are not visible to the user.

[0045] Furthermore, the determination of whether existing text labels are conflicting or incomplete refers to the following: conflicting refers to the same label being applied to two controls with different functions on the same page; incomplete information refers to the absence of text labels for controls or incorrect text labels.

[0046] Furthermore, the text labeling correction for conflicts or incomplete information specifically includes: modifying the conflicting labels to ensure that each switch button on the page has a unique label.

[0047] Further, S102: acquiring the speech signal to be recognized and removing noise from the speech signal to be recognized, specifically includes: removing noise from the speech signal to be recognized using an echo cancellation algorithm.

[0048] Furthermore, the step of converting the denoised speech signal into text data specifically includes: converting the recognized speech commands into a text data stream and transmitting it to the CPU for processing.

[0049] Further, in S103: determine whether the text data exceeds the set length. If so, use a string fuzzy matching algorithm to match the text. Specifically, the CPU compares the text data with the pre-marked text identifier. If the matching rate reaches the set threshold, it is considered that the user has issued a voice command.

[0050] Furthermore, the method also includes:

[0051] When a page redirection command is encountered, click the corresponding control to enter the next level interface and play the corresponding program.

[0052] When the system receives commands such as previous track, next track, pause, play, or stop, it recognizes the clicked button and performs the function.

[0053] Furthermore, the method also includes:

[0054] When encountering a composite control, first determine the primary and secondary functions of the composite control, execute the primary function first, and then execute the secondary function, with the primary function having higher priority than the secondary function.

[0055] For example, to adjust the volume, first enter the volume adjustment function and then adjust the volume to the target value.

[0056] Combined components include: components that have different functions when clicked at different locations, or components that have the same function when clicked at different locations.

[0057] Furthermore, the method also includes:

[0058] Decorative controls are hidden and not numbered.

[0059] Since the voice assistant needs to decide which control to respond to after scanning all controls on the page, the controls need to support simulated clicks. This requires the application layer to embed text annotations. Also, since the accessibility service scans all controls, decorative controls that cannot be clicked (such as brand logos) need to be made invisible so that they can be spoken.

[0060] The definition is as clear as it is visible: Some controls are represented by icons without text labels. The product team needs to define the description of the controls. If the existing text labels conflict or the information is incomplete, these controls also need to be defined.

[0061] Block visible but speakable controls: Since the accessibility service will scan all controls, decorative controls that cannot be clicked should be excluded from the visible and speakable range to avoid accidental clicks. For example, logos or decorative images with text that are not clickable or have no function should be avoided. Voice function clicks should also be avoided to prevent a poor user experience.

[0062] Distinguishing between semantic and visible / speakable commands: Because visible / speakable commands do not support overly complex intent recognition (for example, if a button control is not on this page and a voice command to navigate to a certain place is entered on a music playback interface, the navigation map cannot be opened via the visible / speakable command and the traditional voice link must be used), it is not suitable to use the visible / speakable command when the control supports a large number of intents. The traditional voice link implementation is more flexible.

[0063] Support for generalized expressions: To provide users with more grammatical options, and considering that users in driving scenarios are unlikely to be able to produce fully matching statements for longer texts and titles (such as program names in media services or point-of-interest names in navigation services), the semantic recognition needs to have fuzzy matching capabilities and be compatible with commonly used prefixes and suffixes across various services, such as "on / on-on / turn on Bluetooth," or using suffixes like "Bluetooth on / on-on / turn on," to improve semantic recognition accuracy. The product team needs to compile a list of prefix and suffix requirements from other applications and submit a semantic customization request.

[0064] Since the system has many pages, simple simulated click rules are difficult to predict and accommodate many conflicting situations. Therefore, it is necessary to identify conflicting cases, make appropriate feedback designs, and continuously improve the overall solution.

[0065] Numbering rules for controls: Numbers are assigned from left to right and from top to bottom, increasing in ascending order of positive integers;

[0066] Within an entire application page, there should be no repeated numbers. For example, on a media homepage, the numbers should run from the top of the page to the very bottom of the page after scrolling down, rather than only within the currently visible part of the page.

[0067] The label appears when the voice assistant is activated, and disappears when the voice assistant is deactivated.

[0068] Preset numbers: First, second... Users can select the corresponding number by saying first, second...

[0069] The original text on the control is visible to the user. Supplementary descriptions are embedded in the control but are not visible to the user. For example, the name of the icon is embedded. These labels can be scanned by the accessibility service. To improve the accuracy of clicks that are visible and descriptive, the invisible parts need to be responded to first because they are more comprehensive.

[0070] Example 2

[0071] This embodiment provides a visible and speakable in-vehicle terminal voice recognition system, including:

[0072] The judgment module is configured to: determine whether each page control in the vehicle terminal has a corresponding text identifier; if not, set a text identifier for the page control; if so, determine whether there is a conflict or incomplete information in the existing text identifier; if so, correct the text identifier for the conflict or incomplete information.

[0073] The acquisition module is configured to: acquire the speech signal to be recognized, remove noise from the speech signal to be recognized, and convert the denoised speech signal into text data.

[0074] The recognition control module is configured to: number all page controls using positive integers in a left-to-right, top-to-bottom order; recognize text data and control the corresponding numbered page controls according to the recognition results;

[0075] In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

[0076] It should be noted that the aforementioned judgment module, acquisition module, and identification control module correspond to steps S101 to S103 in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should also be noted that these modules, as part of the system, can be executed in a computer system, such as a set of computer-executable instructions.

[0077] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0078] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0079] Example 3

[0080] This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory to cause the electronic device to perform the method described in Embodiment 1.

[0081] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0082] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0083] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0084] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0085] Those skilled in the art will recognize that the units and algorithm steps described in connection with the various examples of this embodiment 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 invention.

[0086] Example 4

[0087] This embodiment also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the method described in Embodiment 1.

[0088] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A visible-to-speak vehicle terminal voice recognition method, characterized in that, include: Determine if each page control in the vehicle terminal has a corresponding text identifier. If not, set a text identifier for the page control. If it does, determine if there are any conflicts or incomplete information in the existing text identifiers. If so, correct the conflicting or incomplete text identifiers. The determination of whether existing text labels have conflicts or incomplete information, wherein a conflict refers to: on the same page, two controls with different functions are labeled with the same label; incomplete information refers to: the text label is incorrect. The text identification correction for conflicts or incomplete information specifically includes: modifying the conflict identifier to ensure that each page control has a unique identifier; The process involves acquiring the speech signal to be recognized, removing noise from the speech signal, and converting the denoised speech signal into text data. All page controls are numbered with positive integers in a left-to-right, top-to-bottom order; text data is recognized, and the corresponding page controls are controlled according to the recognition results; In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

2. The in-vehicle terminal voice recognition method as described in claim 1, characterized in that, The process involves acquiring the speech signal to be recognized and removing noise from it, specifically by using an echo cancellation algorithm to remove noise from the speech signal to be recognized.

3. The in-vehicle terminal voice recognition method as described in claim 1, characterized in that, If the text data exceeds the set length, a fuzzy string matching algorithm is used to match the text. Specifically, the CPU compares the text data with pre-labeled text identifiers. If the matching degree reaches the set threshold, it is considered that the user has issued a voice command.

4. The in-vehicle terminal voice recognition method as described in claim 1, characterized in that, The method further includes: when a page jump instruction is encountered, clicking the corresponding control to enter the next level interface and playing the corresponding program.

5. The in-vehicle terminal voice recognition method as described in claim 1, characterized in that, The method further includes: when encountering a composite control, first identifying the primary and secondary functions of the composite control, executing the primary function first, and then executing the secondary function, wherein the primary function has a higher priority than the secondary function.

6. A visible and speakable in-vehicle terminal voice recognition system, characterized in that, Implementing a vehicle-mounted terminal voice recognition method as described in any one of claims 1-5, comprising: The judgment module is configured to: determine whether each page control in the vehicle terminal has a corresponding text identifier; if not, set a text identifier for the page control; if so, determine whether there is a conflict or incomplete information in the existing text identifier; if so, correct the text identifier for the conflict or incomplete information. The acquisition module is configured to: acquire the speech signal to be recognized, remove noise from the speech signal to be recognized, and convert the denoised speech signal into text data. The recognition control module is configured to: number all page controls using positive integers in a left-to-right, top-to-bottom order; recognize text data and control the corresponding numbered page controls according to the recognition results; In the process of recognizing text data, it is determined whether the text data exceeds the set length. If it does, a string fuzzy matching algorithm is used to match the text, and the similarity between the text data and the text identifier is calculated. The text identifier corresponding to the maximum similarity is used as the filtered text identifier. The corresponding page controls are controlled to perform actions according to the filtered text identifier.

7. An electronic device, characterized in that it comprises: Memory is used to store computer-readable instructions in a non-transitory manner. as well as Processor, for executing the computer-readable instructions, When the computer-readable instructions are executed by the processor, they perform the method described in any one of claims 1-5.

8. A storage medium, characterized in that, The computer-readable instructions are stored non-transitory, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method according to any one of claims 1-5 are executed.

Images