Speech recognition method and device, vehicle and computer readable storage medium

Abstract

The invention is suitable for the technical field of intelligent recognition, and provides a voice recognition method and device, a vehicle and a computer readable storage medium. The voice recognition method comprises the following steps: acquiring voice information to be recognized; extracting keywords contained in the voice information, wherein the keywords comprise a first keyword to be subjected to semantic recognition; obtaining a to-be-recognized pattern corresponding to the first keyword; matching the pattern to be recognized with patterns of icons in a preset icon library; and if the to-be-recognized pattern is successfully matched with the pattern of the target icon in the preset icon library, outputting a corresponding voice recognition result according to the target icon. According to the embodiment of the invention, the effectiveness of speech recognition can be improved.

Description

Technical Field

[0001] The present application belongs to the field of intelligent recognition technology, and in particular, relates to a speech recognition method, device, vehicle and computer-readable storage medium. Background Art

[0002] Speech recognition technology is also known as Automatic Speech Recognition (ASR). Its goal is to convert human speech vocabulary content into computer-readable input, such as keystrokes, binary codes, and strings. At present, speech recognition technology has been applied to various fields to provide different services to users. In the field of automotive technology, speech recognition has become a basic function in vehicles. Speech recognition in vehicles can help users find or activate related functions without touching buttons, thereby improving the safety factor of users while driving.

[0003] Speech recognition usually matches the voice information input by the user with the information in the word library. If the information is not entered in the word library, for example, when the user uses non-professional terms to describe an icon, speech recognition usually fails and the device cannot recognize the specific meaning of the voice information. In this case, the user's voice information will be replied as "I don't understand, please say it again", which is difficult to meet the user's usage needs. Summary of the invention

[0004] The embodiments of the present application provide a speech recognition method, device, vehicle and readable storage medium, which can solve the problem that speech recognition is prone to failure in related technologies.

[0005] A first aspect of an embodiment of the present application provides a speech recognition method, comprising: obtaining speech information to be recognized; extracting keywords contained in the speech information, the keywords including a first keyword to be semantically recognized; obtaining a pattern to be recognized corresponding to the first keyword; matching the pattern to be recognized with a pattern of an icon in a preset icon library; if the pattern to be recognized successfully matches the pattern of a target icon in the preset icon library, outputting a corresponding speech recognition result according to the target icon.

[0006] A second aspect of an embodiment of the present application provides a speech recognition device, comprising: a speech acquisition unit, used to acquire speech information to be recognized; a keyword extraction unit, used to extract keywords contained in the speech information, wherein the keywords include a first keyword to be semantically recognized; a pattern acquisition unit, used to acquire a pattern to be recognized corresponding to the first keyword; a pattern matching unit, used to match the pattern to be recognized with a pattern of an icon in a preset icon library; and a speech recognition unit, used to output a corresponding speech recognition result according to the target icon if the pattern to be recognized successfully matches the pattern of a target icon in the preset icon library.

[0007] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned speech recognition method when executing the computer program.

[0008] A fourth aspect of an embodiment of the present application provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned speech recognition method when executing the computer program.

[0009] A fifth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned speech recognition method are implemented.

[0010] A sixth aspect of an embodiment of the present application provides a computer program product, which, when executed on an electronic device / vehicle, enables the electronic device / vehicle to execute the above-mentioned speech recognition method.

[0011] In an implementation manner of the present application, voice information to be recognized is obtained, keywords contained in the voice information are extracted, and a pattern to be recognized corresponding to the first keyword is obtained. Then, the pattern to be recognized is matched with the pattern of the icon in the preset icon library. If the pattern to be recognized matches the pattern of the target icon in the preset icon library successfully, the corresponding voice recognition result is output according to the target icon. When the user uses non-professional terms to describe an icon, the target icon actually expressed by the user can be recognized in combination with the keyword pattern, thereby outputting the corresponding voice recognition result according to the target icon, thereby improving the effectiveness of voice recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0013] Figure 1 is a schematic diagram of a power reduction warning light provided in an embodiment of the present application;

[0014] Figure 2 This is a schematic diagram of an implementation flow of a speech recognition method provided in an embodiment of the present application;

[0015] Figure 3 is a schematic diagram of a pictographic pattern of a "turtle lamp" provided in an embodiment of the present application;

[0016] Figure 4 It is a schematic diagram of a specific implementation flow of step S203 provided in an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of an icon area provided in an embodiment of the present application;

[0018] Figure 6 is a structural schematic diagram of a speech recognition device provided in an embodiment of the present application;

[0019] Figure 7 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0020] Figure 8 It is a schematic diagram of the structure of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are protected by the present application.

[0022] Voice recognition usually matches the voice information input by the user with the information in the vocabulary library. If the information is not entered in the vocabulary library, for example, when the user uses non-professional terms to describe an icon, voice recognition will usually fail and the device will not be able to recognize the specific meaning of the voice information, making it difficult to meet the user's usage needs.

[0023] For example, some function icons on the vehicle are highly professional, and users are usually unable to accurately say the corresponding professional terms. Figure 1 The following figure shows an icon in the instrument warning light. The corresponding professional term is "power reduction warning light". That is, when a power reduction warning fault occurs, the vehicle's instrument will light up this warning indicator light. In actual applications, when this warning indicator light is on, the user does not know its professional terminology, so it will figuratively call it a "turtle light". At this time, the user will enter a voice message: "This turtle light is on, what's the situation?" For the device, because the entry library usually does not set non-professional entries such as "turtle light", the device will not be able to recognize the meaning of "turtle light", so it is difficult to recognize the user's semantics. At this time, the user's voice message will be replied as: "I don't understand, please say it again". And the user has no other way to replace this description, and ultimately cannot get the answer he wants.

[0024] In view of this, the present application proposes a speech recognition method that can recognize the pattern of keywords in voice information. When a user uses non-professional terms to describe an icon, the target icon actually expressed by the user can be recognized in combination with the pattern of keywords, thereby recognizing the user's semantics.

[0025] In order to illustrate the technical solution of the present application, a specific embodiment is provided below for illustration.

[0026] Figure 2 A schematic diagram of an implementation flow of a speech recognition method provided in an embodiment of the present application is shown. The method can be applied to a processor and is suitable for situations where the effectiveness of speech recognition needs to be improved.

[0027] In the implementation manner of the present application, the processor may be provided on an electronic device, for example, the electronic device may be a computer, a smart phone, a vehicle-mounted device, etc. The processor may also be integrated into a vehicle, which may be a new energy vehicle, a hybrid vehicle, or other types of vehicles, which is not limited in the present application.

[0028] Specifically, the above-mentioned speech recognition method may include the following steps S201 to S205.

[0029] Step S201, obtaining voice information to be recognized.

[0030] In the implementation manner of the present application, the voice information may carry questions or instructions raised by the user (or the originator of the voice information) through voice. The above processor may control the audio collection component such as the microphone to collect the audio of the user to obtain the voice information to be recognized.

[0031] Preferably, after recognizing the wake-up instruction, the processor can control the audio collection component to collect audio from the user, thereby avoiding the problem of misrecognition. The wake-up instruction can be triggered by the user through touch screen operation, tapping a physical button, etc., or by the user inputting a preset wake-up voice. For example, the user can wake up the voice recognition through a specified wake-up voice such as "Hello, Xiao Wei" or "Hello, Harvard", and the processor controls the voice collection component to collect the user's subsequent voice to obtain the voice information to be recognized.

[0032] Step S202: extracting keywords contained in the voice information.

[0033] In the implementation of the present application, after collecting the voice information, the processor can perform keyword detection to extract the keywords contained in the voice information. For example, the user inputs the voice "What is the situation when this turtle light is on?", and the processor can extract the keywords "this", "turtle light", "on", "is", "what", and "situation" based on the voice information.

[0034] Specifically, the above-mentioned keyword extraction can be achieved by inputting the voice information into a trained voice keyword recognition model, or by analyzing the characteristic parameters of the sound such as pitch, frequency, loudness, etc. in the voice information to determine the keywords in the voice information. This application does not impose any restrictions on this.

[0035] In an implementation manner of the present application, the keywords included in the voice information may include a first keyword to be semantically recognized, and the first keyword may be all or part of the keywords in the voice information, which is not limited in the present application.

[0036] Step S203: obtaining a pattern to be recognized corresponding to the first keyword.

[0037] The pattern to be identified corresponding to the first keyword refers to the pattern described by the first keyword and is used to identify the semantics of the first keyword, for example, it may refer to the pictographic pattern of the first keyword.

[0038] It should be noted that the present application does not limit the method for obtaining the pattern to be identified. For example, the processor may search for the first keyword in a search engine to obtain the pattern to be identified corresponding to the first keyword. For another example, the processor may be pre-installed with a pattern library, each pattern in the pattern library is annotated with corresponding keyword information, and based on the keyword information, the pattern corresponding to the first keyword can be searched from the pattern library.

[0039] Moreover, for the first keyword, the processor can further decompose the keyword semantically, and then use the decomposed sub-keywords to obtain the corresponding pattern to be identified. For example, the aforementioned "turtle lamp" can be decomposed into two sub-keywords, "turtle" and "lamp". Based on the sub-keyword "turtle", the search can obtain Figure 3 The "turtle" pattern shown is used as a pattern to be identified corresponding to the first keyword "turtle lamp".

[0040] Step S204, matching the pattern to be identified with the patterns of icons in the preset icon library.

[0041] The preset icon library contains one or more icons, each of which is composed of a corresponding pattern. The so-called "pattern" may include a background pattern and a foreground pattern. In an embodiment of the present application, the processor may match the pattern to be identified with the pattern of the icon in the preset icon library to confirm which icon in the preset icon library the first keyword in the voice information refers to.

[0042] In some embodiments, the processor may perform a similarity calculation between the pattern to be identified and the pattern of the icons in the preset icon library to confirm whether the pattern to be identified matches the pattern of the icons in the preset icon library. For example, if the contour similarity between the two patterns is greater than a preset similarity threshold, the processor may confirm that the two patterns match successfully. In other embodiments, the processor may also match the pattern to be identified with the pattern of the icons in the preset icon library based on pattern information such as grayscale values ​​and pixel coordinates. It should be understood that other image / pattern matching methods are also applicable to the present application, and the present application does not limit this.

[0043] Step S205: if the pattern to be recognized successfully matches the target icon in the preset icon library, the corresponding speech recognition result is output according to the target icon.

[0044] In an embodiment of the present application, if the pattern to be recognized successfully matches the target icon in the preset icon library, it means that the target icon is the icon referred to by the first keyword. At this time, the processor can output the corresponding voice recognition result according to the target icon.

[0045] For example, by Figure 3 The "turtle light" pattern shown is matched with the pattern of the icon in the preset icon library to confirm Figure 1 The icon "power reduction warning light" shown is similar to Figure 3 If the "turtle light" pattern shown is matched successfully, it can be confirmed that the semantics of the "turtle light" in the voice information is recognized as "power reduction warning light", and the voice recognition result is output.

[0046] Correspondingly, based on the speech recognition result, the processor can execute the control strategy corresponding to the speech information.

[0047] Specifically, in some implementations, if the voice information of the user is identified as a voice question according to the voice recognition result, the interpretation text of the corresponding entry can be searched in the preset entry library based on the voice recognition result, and the interpretation text can be output through voice. For example, if the user inputs the voice "What is the situation when this turtle light is on?", the voice recognition will call out the interpretation text of "power reduction warning light" in the preset entry library, and broadcast the interpretation text in the form of voice broadcast to provide a voice prompt to the user.

[0048] In other implementations, when the user's voice information is identified as a voice command according to the voice recognition result, the corresponding language command can be executed. For example, if the user inputs the voice "turn off the turtle light", the processor can execute the language command to turn off the "power reduction warning light".

[0049] Of course, if the pattern to be identified fails to match the icons in the preset icon library, the processor can output a preset prompt voice, such as "I don't understand, please say it again", to remind the user to re-enter the voice.

[0050] In an implementation manner of the present application, voice information to be recognized is obtained, keywords contained in the voice information are extracted, and a pattern to be recognized corresponding to the first keyword is obtained. Then, the pattern to be recognized is matched with the pattern of the icon in the preset icon library. If the pattern to be recognized matches the pattern of the target icon in the preset icon library successfully, the corresponding voice recognition result is output according to the target icon. When the user uses non-professional terms to describe an icon, the target icon actually expressed by the user can be recognized in combination with the keyword pattern, thereby outputting the corresponding voice recognition result according to the target icon, thereby improving the effectiveness of voice recognition.

[0051] In order to more accurately obtain the pattern to be identified corresponding to the first keyword, in some implementations, such as Figure 4 As shown, the above step S203 may include the following steps S401 to S403.

[0052] Step S401, obtaining the gaze point position of the originator of the voice information.

[0053] Specifically, the processor can obtain an image of the initiator of the voice information captured by the in-vehicle camera, and based on the image, identify the position of the initiator's gaze point. Considering that there may be multiple drivers and passengers in the vehicle, each driver and passenger can pre-record a preset test voice, and the processor will bind the sound characteristics of the test voice (such as speaking speed, intonation, etc.) with the driver's avatar. When the voice information to be recognized is obtained, the processor can compare the sound characteristics corresponding to the voice information with the sound characteristics of the test voice, retrieve the corresponding avatar, and then use the avatar to search for the location of the initiator of the voice information in the car, and use the in-vehicle camera to shoot the initiator at that location, thereby avoiding the occurrence of misidentification problems.

[0054] In some implementations, in response to the wake-up instruction, the processor can control the microphone to collect the voice of the initiator to obtain voice information, and control the camera to collect images of the initiator. Accordingly, when obtaining the gaze point position, the processor can determine the initiator's eye movement information based on the image collected by the camera, and determine the initiator's gaze point position based on the eye movement information.

[0055] Among them, the wake-up instruction can refer to the previous description, and this application will not go into details. Eye movement information is information used for eye tracking. Based on the image captured by the in-vehicle camera, the processor can identify the initiator's eye features (such as pupil position) and determine the user's line of sight based on the eye features to obtain the initiator's gaze point position.

[0056] Step S402: If the gaze point is located on a target component of the vehicle, a regional image of the icon area on the target component is obtained.

[0057] The vehicle may be provided with several components, such as an instrument, a head-up display, a multimedia display, etc., each of which may be provided with one or more icons. The target component is any component provided on the vehicle.

[0058] In an embodiment of the present application, if the gaze point is located on a target component of the vehicle, it means that the initiator is gazing at the target component while initiating a voice command, that is, the initiator is outputting voice in response to an icon on the target component. At this time, the processor can obtain an area image of the icon area on the target component.

[0059] Specifically, the icon area is the area where the icon appears on the target component. For example, when the target component is a head-up display or a multimedia display, the icon area may refer to the entire display area, and the processor may obtain the regional image of the icon area by taking a screenshot; when the target component is an instrument, the icon area may refer to the area composed of all instruments, and the processor may obtain the regional image of the icon area by photographing the instruments in the car through a camera.

[0060] Step S403: Based on the regional image, a pattern to be identified corresponding to the first keyword is determined from among the patterns of icons in the icon region.

[0061] For example, the processor may compare the pattern of the icon in the icon area with the pictographic pattern of the first keyword, and use the pattern with the highest similarity to the pictographic pattern as the pattern to be identified. For another example, the processor may perform contour recognition based on the pattern of the icon in the icon area, match the corresponding description text based on the contour of the pattern, and then compare the first keyword with the description text. If the first keyword exists in the description text, the pattern corresponding to the description text may be used as the pattern to be identified corresponding to the first keyword. This application does not limit this.

[0062] In this way, the pattern to be identified can be made to fit the pattern of the icon actually seen by the user, and then when the pattern to be identified is matched with the pattern of the icon in the preset icon library, the target icon can be matched more accurately.

[0063] In order to improve the search efficiency, the icon area can be divided into multiple sub-areas, each of which contains at least one icon. Figure 5 The icon area shown contains 15 icons, and the icon area is divided into 3*3 sub-areas, each of which contains at least one icon.

[0064] Accordingly, the processor can determine the sub-region where the gaze point is located. If the gaze point is located in the target sub-region, then based on the regional image, a pattern to be identified corresponding to the first keyword is determined in the pattern of icons in the target sub-region. The target sub-region is any one of the multiple sub-regions.

[0065] For example, after receiving the wake-up command, the camera can capture an image, generate an eye movement track (i.e., the track of the gaze point changes) based on the image, project the gaze point position onto a display screen divided into a nine-square grid, determine the specific square (i.e., the target sub-area) in the nine-square grid where the gaze point position is located, and search for the pattern to be identified corresponding to the first keyword in the pattern of icons in the square. Figure 5 When the sub-area of ​​the third column of the first row is shown, the pattern to be recognized corresponding to the first keyword can be determined from the pattern of the fifth icon and the pattern of the sixth icon. In this way, the processor does not need to match all icon patterns in the icon area, which helps to improve the efficiency of speech recognition.

[0066] In actual application scenarios, users may not input voice for the current scene. For example, the user may recall that the "power reduction warning light" was on, and suddenly want to know about the warning indicator light and ask: "What happens when the turtle light is on?" At this time, the user will not look at the components in the car. In order to avoid such scenarios affecting the accuracy of voice recognition, the processor can confirm whether there is a second keyword with regional directionality in the keywords contained in the voice information after extracting the keywords contained in the voice information. Among them, the second keyword is used to point to a certain area in the current scene.

[0067] If there is a second keyword with regional directivity among the keywords included in the voice information, it means that the user is asking a question for the current scene, and the processor can execute steps S401 to S403 to find the icon described by the user in the current scene.

[0068] Specifically, the second keyword can be classified into two categories: one is demonstrative pronouns, such as "this" and "that"; the other is component names, such as "on the display screen" and "on the instrument". If demonstrative pronouns exist in the keywords contained in the voice information, the processor can determine the pattern to be recognized corresponding to the first keyword in the current scene according to the method of steps S401 to S403. If the keyword contained in the voice information contains a component name, the processor can directly use the corresponding component as the target component, obtain the regional image of the icon area on the target component, and based on the regional image, determine the pattern to be recognized corresponding to the first keyword in the pattern of the icon in the icon area.

[0069] In this way, when the user asks a question regarding the current scene, the processor can quickly determine the pattern of the icon described by the user in the current scene, and obtain the pattern to be identified corresponding to the first keyword.

[0070] If the keyword included in the voice information does not contain the second keyword, or the user's gaze point is not located on the vehicle component, the processor can use the pictographic pattern of the first keyword as the pattern to be recognized. Although the pictographic pattern may be different from the pattern of the icon that the user actually wants to express, due to a certain similarity, it can still be used as a basis for matching with the pattern of the target icon in the preset icon library.

[0071] Considering that the computational complexity of pattern matching is higher than that of text matching, in some embodiments of the present application, before obtaining the pattern to be recognized corresponding to the first keyword, the processor may also match the keyword contained in the voice information with each entry in the preset entry library. The preset entry library may be stored in a local processor or in the cloud, which is not limited by the present application.

[0072] If the keyword included in the voice information successfully matches the target entry in the preset entry library, the processor can output the corresponding voice recognition result according to the target entry without the need for pattern matching.

[0073] If the keyword included in the voice information fails to match the target word in the preset word library, the processor can execute the aforementioned steps S203 to S205 to output the corresponding voice recognition result.

[0074] For ease of understanding, the solution provided in this application is explained below through an example. After the user inputs the wake-up command, the processor controls the camera and microphone in the car to collect images and audio of the user respectively. The user inputs the voice "What's the situation when this turtle light is on", and the processor extracts the keyword "turtle light" based on the voice information, and fails to match in the preset entry library. Since the voice information contains the keyword "this", the processor determines the user's gaze point position based on the image captured by the camera in the car to obtain the pattern to be recognized corresponding to the first keyword. When matching the pattern to be recognized with the pattern of the icon in the preset icon library, since the pattern to be recognized successfully matches the pattern of the icon "power reduction alarm light", the processor can confirm that the "turtle light" actually refers to the "power reduction alarm light" and output the corresponding voice recognition result.

[0075] It should be noted that, for the sake of simplicity of description, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders.

[0076] like Figure 6 FIG. 6 is a schematic diagram of the structure of a speech recognition device 600 provided in an embodiment of the present application, wherein the speech recognition device 600 is configured on a processor.

[0077] Specifically, the speech recognition device 600 may include:

[0078] The voice acquisition unit 601 is used to acquire the voice information to be recognized;

[0079] A keyword extraction unit 602, configured to extract keywords contained in the voice information, wherein the keywords include a first keyword to be semantically recognized;

[0080] A pattern acquisition unit 603 is used to acquire a pattern to be identified corresponding to the first keyword;

[0081] A pattern matching unit 604, used to match the pattern to be identified with the pattern of the icon in the preset icon library;

[0082] The speech recognition unit 605 is configured to output a corresponding speech recognition result according to the target icon if the pattern to be recognized successfully matches the pattern of the target icon in the preset icon library.

[0083] In some embodiments of the present application, the pattern acquisition unit 603 can be specifically used to: obtain the gaze point position of the initiator of the voice information; if the gaze point position is located on a target component of the vehicle, obtain an area image of the icon area on the target component, where the target component is any component set on the vehicle; based on the area image, determine the pattern to be identified corresponding to the first keyword in the pattern of icons in the icon area.

[0084] In some embodiments of the present application, the above-mentioned icon area is divided into multiple sub-areas, each of which contains at least one icon; the above-mentioned pattern acquisition unit 603 can be specifically used to: if the gaze point position is located in the target sub-area, then based on the area image, determine the pattern to be identified corresponding to the first keyword in the pattern of icons in the target sub-area, wherein the target sub-area is any one of the multiple sub-areas.

[0085] In some embodiments of the present application, the above-mentioned voice acquisition unit 601 can be specifically used to control the microphone to collect the voice of the initiator in response to the wake-up instruction to obtain the voice information, and control the camera to collect images of the initiator; the above-mentioned pattern acquisition unit 603 can be specifically used to: determine the eye movement information of the initiator based on the image collected by the camera; determine the gaze point position of the initiator based on the eye movement information.

[0086] In some embodiments of the present application, the pattern acquisition unit 603 may be specifically configured to: if the keywords included in the voice information include a second keyword with regional directivity, then acquire the gaze point position.

[0087] In some embodiments of the present application, the pattern acquisition unit 603 may be specifically configured to: if the second keyword with regional directivity does not exist in the keywords included in the voice information, use the pictographic pattern of the first keyword as the pattern to be identified.

[0088] In some embodiments of the present application, the above-mentioned speech recognition device 600 may also include a term matching unit, which is used to match the keywords contained in the speech information with each term in a preset term library; if the keywords contained in the speech information successfully match the target terms in the preset term library, the corresponding speech recognition results are output according to the target terms.

[0089] It should be noted that, for the convenience and simplicity of description, the specific working process of the above-mentioned speech recognition device 600 can be referred to Figures 1 to 5 The corresponding process of the method will not be repeated here.

[0090] like Figure 7 , which is a schematic diagram of an electronic device provided in an embodiment of the present application. Specifically, the electronic device 7 may include: a processor 70, a memory 71, and a computer program 72 stored in the memory 71 and executable on the processor 70, such as a speech recognition program. When the processor 70 executes the computer program 72, the steps in the above-mentioned speech recognition method embodiments are implemented, such as Figure 2 Alternatively, when the processor 70 executes the computer program 72, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Figure 6 The functions of the speech acquisition unit 601, the keyword extraction unit 602, the pattern acquisition unit 603, the pattern matching unit 604 and the speech recognition unit 605 are shown.

[0091] The computer program may be divided into one or more modules / units, which are stored in the memory 71 and executed by the processor 70 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0092] For example, the computer program can be divided into: a voice acquisition unit, a keyword extraction unit, a pattern acquisition unit, a pattern matching unit, and a voice recognition unit. The specific functions of each unit are as follows: a voice acquisition unit, used to acquire voice information to be recognized; a keyword extraction unit, used to extract keywords contained in the voice information, wherein the keywords include a first keyword to be semantically recognized; a pattern acquisition unit, used to acquire a pattern to be recognized corresponding to the first keyword; a pattern matching unit, used to match the pattern to be recognized with the pattern of an icon in a preset icon library; a voice recognition unit, used to output a corresponding voice recognition result according to the target icon if the pattern to be recognized successfully matches the pattern of a target icon in the preset icon library.

[0093] The electronic device may include, but is not limited to, a processor 70 and a memory 71. Those skilled in the art will appreciate that Figure 7 These are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0094] The processor 70 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0095] The memory 71 may be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device. The memory 71 may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Furthermore, the memory 71 may also include both an internal storage unit of the electronic device and an external storage device. The memory 71 is used to store the computer program and other programs and data required by the electronic device. The memory 71 may also be used to temporarily store data that has been output or is to be output.

[0096] It should be noted that, for the convenience and brevity of description, the structure of the above electronic device can also refer to the specific description of the structure in the method embodiment, which will not be repeated here.

[0097] like Figure 8 FIG. 8 is a schematic diagram of a vehicle provided in an embodiment of the present application. Specifically, the vehicle 8 may be integrated with a Figure 1 A processor 80 of the illustrated method.

[0098] Specifically, the vehicle 8 may include: a processor 80, a memory 81, and a computer program 82 stored in the memory 81 and executable on the processor 80, such as a parking control program. When the processor 80 executes the computer program 82, the steps in the above-mentioned various speech recognition method embodiments are implemented, such as Figure 2 Alternatively, when the processor 80 executes the computer program 82, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Figure 6 The functions of the speech acquisition unit 601, keyword extraction unit 602, pattern acquisition unit 603, pattern matching unit 604 and speech recognition unit 605 are shown.

[0099] It should be noted that, for the convenience and simplicity of description, the structure of the above-mentioned vehicle can also refer to the description in the electronic device embodiment and the specific description of the structure in the method embodiment, which will not be repeated here.

[0100] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0101] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0102] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0103] In the embodiments provided in the present application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0105] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0106] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0107] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A speech recognition method, characterized in that: include: Acquire voice information to be recognized; Extracting keywords contained in the voice information, wherein the keywords include a first keyword to be semantically recognized; Acquire a pattern to be identified corresponding to the first keyword; Matching the pattern to be identified with patterns of icons in a preset icon library; If the pattern to be recognized successfully matches the pattern of the target icon in the preset icon library, the corresponding voice recognition result is output according to the target icon.

2. The speech recognition method according to claim 1, characterized in that: The acquiring the pattern to be recognized corresponding to the first keyword includes: Obtaining the gaze point position of the originator of the voice information; If the gaze point is located on a target component of the vehicle, a regional image of an icon region on the target component is acquired, where the target component is any component provided on the vehicle; Based on the area image, a pattern to be identified corresponding to the first keyword is determined from among the patterns of icons in the icon area.

3. The speech recognition method according to claim 2, characterized in that: The icon area is divided into a plurality of sub-areas, and each of the sub-areas contains at least one icon; The step of determining, based on the regional image, a pattern to be identified corresponding to the first keyword from among the patterns of icons in the icon region, comprises: If the gaze point is located in a target sub-region, a pattern to be identified corresponding to the first keyword is determined in the pattern of icons in the target sub-region based on the region image, wherein the target sub-region is any one of the multiple sub-regions.

4. The speech recognition method according to claim 2, characterized in that: The acquiring of the voice information to be recognized includes: In response to the wake-up instruction, controlling the microphone to collect the voice of the initiator to obtain the voice information, and controlling the camera to collect an image of the initiator; The obtaining of the gaze point position of the originator of the voice information comprises: Determining the eye movement information of the initiator according to the image collected by the camera; The gaze point position of the initiator is determined according to the eye movement information.

5. The speech recognition method according to claim 2, characterized in that: The obtaining of the gaze point position of the originator of the voice information comprises: If there is a second keyword with regional directivity among the keywords included in the voice information, the gaze point position is obtained.

6. The speech recognition method according to claim 5, characterized in that: If the keywords included in the voice information do not include the second keyword, the voice recognition method further includes: The pictographic pattern of the first keyword is used as the pattern to be recognized.

7. The speech recognition method according to any one of claims 1 to 6, characterized in that: Before obtaining the pattern to be recognized corresponding to the first keyword, the method further includes: Matching the keywords contained in the voice information with each entry in a preset entry library; If the keyword contained in the voice information successfully matches the target entry in the preset entry library, the corresponding voice recognition result is output according to the target entry.

8. A speech recognition device, characterized in that: include: A voice acquisition unit, used to acquire voice information to be recognized; A keyword extraction unit, used to extract keywords contained in the voice information, wherein the keywords include a first keyword to be semantically recognized; A pattern acquisition unit, used to acquire a pattern to be identified corresponding to the first keyword; A pattern matching unit, used to match the pattern to be identified with the pattern of the icon in the preset icon library; The speech recognition unit is used to output a corresponding speech recognition result according to the target icon if the pattern to be recognized successfully matches the pattern of the target icon in the preset icon library.

9. A vehicle comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the speech recognition method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the speech recognition method according to any one of claims 1 to 7 are implemented.

Images