CN115148189A - Multifunctional synchronous implementation system and method for driver-defined voice command - Google Patents
Multifunctional synchronous implementation system and method for driver-defined voice commandDownload PDFInfo
- Publication number
- CN115148189A CN115148189ACN202210890758.1ACN202210890758ACN115148189ACN 115148189 ACN115148189 ACN 115148189ACN 202210890758 ACN202210890758 ACN 202210890758ACN 115148189 ACN115148189 ACN 115148189A
- Authority
- CN
- China
- Prior art keywords
- voice
- user
- preset
- semantic
- pinyin
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/005—Language recognition
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
- G10L2015/0631—Creating reference templates; Clustering
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fittings On The Vehicle Exterior For Carrying Loads, And Devices For Holding Or Mounting Articles (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及智能车辆技术领域，具体涉及一种驾驶员自定义语音命令多功能同步实现系统及方法。The invention relates to the technical field of intelligent vehicles, in particular to a system and method for realizing multi-function synchronization of driver-defined voice commands.
背景技术Background technique
随着科技水平的不断提高以及经济社会的不断发展，汽车已经逐渐开始走入了家家户户，成为了每个家庭必备的交通工具。With the continuous improvement of the level of science and technology and the continuous development of the economy and society, automobiles have gradually begun to enter every household and become an essential means of transportation for every family.
在使用车辆的过程中，人们就开始对车辆的智能驾驶体验提出了越来越高的需求，随之诞生的就有通过智能语音的方式来控制车辆内某些功能的使用。目前该功能实现的方式通常都是，一条语音对应控制一个具体的功能，比如打开空调，实际操作的就仅仅是打开空调这一个车机功能，或者一条语音对应控制一个特定的场景，比如打开快速清醒模式，有可能这个快速清醒模式会对应多个小功能，例如打开空调、关闭车窗、关闭天窗等等，但是这种方案通常是车机厂家在车辆出厂时就已经定死了的，无法达到用户通过自己的用车习惯而进行更改的效果。In the process of using the vehicle, people have begun to put forward higher and higher demands for the intelligent driving experience of the vehicle, and the use of certain functions in the vehicle is controlled by means of intelligent voice. At present, this function is usually implemented in a way that a voice corresponds to controlling a specific function, such as turning on the air conditioner, the actual operation is only to turn on the car function of the air conditioner, or a voice corresponds to controlling a specific scene, such as turning on the quick Awake mode, it is possible that this quick wake-up mode will correspond to several small functions, such as turning on the air conditioner, closing the windows, closing the sunroof, etc., but this solution is usually fixed by the car manufacturer when the vehicle leaves the factory, and cannot be used. To achieve the effect of users making changes through their own car habits.
公开号为CN114132143A的发明申请《基于车机语音控制车辆空调的方法、智能汽车和可读介质》中，通过车机接收用户发出的用于控制车辆空调的语音词条，并将语音词条上传至车机云端进行解析，实现了通过多样化的语音输出即可对车辆空调进行控制的效果，通过车机结合语音控制车辆空调交互逻辑判断，简单且快速的实现了语音控制车辆空调进行自动调节的效果，满足了用户的驾驶需求，提高了用户的用车体验。In the invention application with the publication number CN114132143A, "Method for Controlling Vehicle Air Conditioning Based on Vehicle Machine Voice, Smart Car and Readable Medium", the voice entry for controlling the vehicle air conditioner sent by the user is received through the vehicle machine, and the voice entry is uploaded. Go to the car-machine cloud for analysis, and realize the effect of controlling the vehicle air conditioner through a variety of voice outputs. Through the car-machine combined with the voice to control the vehicle air-conditioning interactive logic judgment, the voice-controlled vehicle air-conditioning can be automatically adjusted simply and quickly. It can meet the user's driving needs and improve the user's car experience.
上述方案中即存在着前文所说的问题，由一条语音指令对应一个功能的实现，无法根据用户的实时使用需求以及使用习惯来进行调整，功能性、实用性较差。The above solution has the problem mentioned above. The realization of one function corresponding to one voice command cannot be adjusted according to the user's real-time usage needs and usage habits, and the functionality and practicability are poor.
公开号为CN113851126A的发明申请《车内语音交互方法及系统》中，通过动态地调整语音浮窗的位置和布局，建立起更直观的语音助手UI元素与汽车座舱空间的映射关系，使得用户更便捷地明确当前语音交互的焦点音区，避免交互对象混淆和无效输入，提升与用户的交互体验。In the invention application "In-Vehicle Voice Interaction Method and System" with the publication number CN113851126A, by dynamically adjusting the position and layout of the voice floating window, a more intuitive mapping relationship between the UI elements of the voice assistant and the car cockpit space is established, so that users can be more intuitive. It is convenient to clarify the focus area of the current voice interaction, avoid the confusion of interactive objects and invalid input, and improve the interaction experience with users.
该方案中，解决了二维的中控车机UI界面与三维的车舱内的空间位置，导致视觉上无法体现出语音助手与车内说话人的空间位置关系的问题，而在语音功能实现的方面则也没有相关的内容涉及。In this solution, the two-dimensional central control car UI interface and the three-dimensional spatial position in the cabin are solved, resulting in the problem that the spatial position relationship between the voice assistant and the speaker in the car cannot be visually reflected. There is no related content involved.
发明内容SUMMARY OF THE INVENTION
针对现有技术中存在的不足，本发明目的是提供一种能够帮助用户按照自己的用车习惯实现定制化的车机功能的驾驶员自定义语音命令多功能同步实现系统。Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide a multi-functional synchronization implementation system for driver's self-defined voice commands that can help users implement customized car-machine functions according to their own car habits.
为解决上述技术问题，本发明提供的技术方案是：所述的驾驶员自定义语音命令多功能同步实现系统，包括场景定义模块、语音接收模块、语义分析模块以及MCU；In order to solve the above-mentioned technical problems, the technical solution provided by the present invention is as follows: the multi-function synchronization implementation system of the driver's self-defined voice command includes a scene definition module, a voice receiving module, a semantic analysis module and an MCU;
场景定义模块位于用户的移动智能终端，场景定义模块根据用户的预设功能对应生成json文件，并上传至云端储存解析；The scene definition module is located in the user's mobile smart terminal, and the scene definition module generates a corresponding json file according to the user's preset function, and uploads it to the cloud for storage and analysis;
语音接收模块位于车端，用于接收用户在车内时发出的语音控制指令并上传至云端；The voice receiving module is located at the end of the car, and is used to receive the voice control commands issued by the user when they are in the car and upload them to the cloud;
语义分析模块位于云端，语义分析模块对场景定义模块上传的json文件进行语义识别，并对语音接收模块上传的语音控制指令同样进行语义识别，在将两者语义识别为符合后向MCU发送对应的操作执行指令；The semantic analysis module is located in the cloud. The semantic analysis module performs semantic recognition on the json file uploaded by the scene definition module, and also performs semantic recognition on the voice control instructions uploaded by the voice receiving module. operation execution instructions;
MCU位于车端，MCU在接收到来自语义分析模块发送的操作执行指令后，控制车机TBox执行相应的功能。The MCU is located at the vehicle end. After the MCU receives the operation execution instruction sent from the semantic analysis module, it controls the vehicle machine TBox to perform the corresponding function.
在上述技术方案中，用户可首先通过移动智能终端对特定场景下的车机功能执行需求进行预设，比如用户可以预设在“打开空调”模式下，同步控制车内车窗关闭、天窗关闭等等，随后通过云端的语义分析模块对用户预设的内容进行储存。在用户通过车端的语音接收模块发出语音控制指令后，云端的语义分析模块可以对该指令进行语义分析，以确认用户的实际功能需求，随后对储存的预设内容进行比对，根据比对结果向车端发送相对应的操作执行指令，最后在车端完成用户预设的功能需求。In the above technical solution, the user can first preset the vehicle function execution requirements in a specific scenario through the mobile smart terminal. For example, the user can preset the "turn on the air conditioner" mode to synchronously control the closing of the car window and the sunroof in the car. And so on, and then store the content preset by the user through the semantic analysis module in the cloud. After the user sends a voice control command through the voice receiving module on the vehicle end, the semantic analysis module in the cloud can perform semantic analysis on the command to confirm the user's actual functional requirements, and then compare the stored preset contents, according to the comparison results. The corresponding operation execution instructions are sent to the vehicle end, and finally the user's preset functional requirements are completed at the vehicle end.
由此实现了整车功能的定制化实现，提高了用户体验，节省了车机操作繁琐度与时间。As a result, the customized realization of the functions of the whole vehicle is realized, the user experience is improved, and the tedious operation and time of the vehicle and the machine are saved.
优选的，场景定义模块包括功能区以及预设区，功能区包括可通过车机TBox执行的所有功能单元，用户可通过拖动的方式将功能单元从功能区拖动到预设区，预设区内存留的功能单元即为用户的预设功能并形成为json文件。Preferably, the scene definition module includes a functional area and a preset area, the functional area includes all functional units that can be executed through the TBox of the vehicle, and the user can drag the functional units from the functional area to the preset area by dragging. The functional units remaining in the area are the user's preset functions and are formed into json files.
以听歌场景为例，用户通过通过场景定义模块，即手机APP中的功能，新增自定义场景，场景名称为“听歌场景”，然后在功能区内可以看到所有可实现的车机功能，用户将需要的功能从功能区拖动到预设区内，比如“关闭车窗”、“关闭天窗”、“打开在线音乐”、“打开空调”、“风量设为一级”、“温度设为23℃”等，随后点击确定，即可生成该场景的语义命令，每一个小功能生产一个子json串：Taking the listening song scene as an example, the user adds a custom scene through the scene definition module, that is, the function in the mobile APP, the scene name is "listening song scene", and then in the function area, all the vehicles that can be realized can be seen. function, the user drags the desired function from the functional area to the preset area, such as "close the window", "close the sunroof", "open online music", "turn on the air conditioner", "set the air volume to the first level", " Set the temperature to 23°C", etc., and then click OK to generate the semantic command of the scene, and each small function produces a sub-json string:
关闭车窗：Close the windows:
{"action":"execute","domain":"win_control","intention":"instructing","type":"close_win"}；{"action":"execute","domain":"win_control","intention":"instructing","type":"close_win"};
关闭天窗：To close the sunroof:
{"action":"execute","domain":"sunroof_control","intention":"instructing","type":"close_sunroof"}；{"action":"execute","domain":"sunroof_control","intention":"instructing","type":"close_sunroof"};
打开在线音乐：Open online music:
{"action":"execute","domain":"media","intention":"instructing","type":"open_net_music"}；{"action":"execute","domain":"media","intention":"instructing","type":"open_net_music"};
打开空调：on the aircon:
{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac"}；{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac"};
风量设为一级：The air volume is set to one level:
"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{speed:1}"}]；"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{speed:1}"}];
温度设为23℃：Set the temperature to 23°C:
{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{temperature:23}"}；{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{temperature:23}"};
上述全部json内容参照【听歌场景.json】文件。For all the above json content, please refer to the [Listening to Song Scene.json] file.
优选的，功能单元在预设区内由上至下排列，功能单元的上下顺序可调节，MCU根据功能单元的排序控制车机TBox依次执行相应的功能，这样可以方便用户通过简单的顺序调换即可达到控制车机功能工作顺序的控制需求，同时操作简便，极大地丰富了系统的功能性，提高了用户的使用体验。Preferably, the functional units are arranged from top to bottom in the preset area, and the upper and lower order of the functional units can be adjusted. It can meet the control requirements of controlling the working sequence of vehicle functions, and at the same time, it is easy to operate, which greatly enriches the functionality of the system and improves the user experience.
优选的，语音接收模块包括设置在驾驶位的多个麦克风阵列，由此可通过多个麦克风阵列实现对驾驶员的语音指令收集。Preferably, the voice receiving module includes a plurality of microphone arrays arranged at the driver's seat, so that the collection of the driver's voice instructions can be realized through the plurality of microphone arrays.
优选的，语义分析模块包括获取单元和识别单元，获取单元获取语音信息并根据语音信息提取语音状态，识别单元将提取得到的语音状态输入目标语义识别模型中，目标语义识别模型根据语音状态获得拼音特征、或者拼音特征和文字特征，得到与语音信息相对应的语义信息，由此通过语义分析模块实现了对获取得到的驾驶员发出的语音指令的语音信息分析，从而确定驾驶员此时需要实现什么样的车机功能。Preferably, the semantic analysis module includes an acquisition unit and a recognition unit, the acquisition unit acquires voice information and extracts the voice state according to the voice information, the recognition unit inputs the extracted voice state into the target semantic recognition model, and the target semantic recognition model obtains pinyin according to the voice state features, or pinyin features and text features, to obtain semantic information corresponding to the voice information, thereby realizing the voice information analysis of the obtained voice commands issued by the driver through the semantic analysis module, so as to determine that the driver needs to implement What kind of car function.
优选的，识别单元可获取训练数据集，并将该训练数据集输入至初始语义识别模型中，初始语义识别模型包括拼音转换支路和匹配支路，拼音转换支路根据语音状态获得的拼音特征、或者拼音特征和文字特征，结合匹配支路根据拼音特征得到的对应语义信息，得到目标语义识别模型。Preferably, the recognition unit can acquire a training data set, and input the training data set into an initial semantic recognition model, the initial semantic recognition model includes a pinyin conversion branch and a matching branch, and the pinyin conversion branch obtains pinyin features according to the speech state , or pinyin features and text features, combined with the corresponding semantic information obtained by the matching branch according to the pinyin features, to obtain a target semantic recognition model.
优选的，根据语音状态获得的拼音特征、或者拼音特征和文字特征包括：Preferably, the pinyin features, or the pinyin features and text features obtained according to the voice state include:
根据多个依次排列的语音状态，依次得到每个语音状态对应的文字特征，并根据文字特征依次得到对应的拼音特征；According to a plurality of sequentially arranged voice states, the text features corresponding to each voice state are sequentially obtained, and the corresponding pinyin features are obtained in turn according to the text features;
或者根据多个依次排列的语音状态得到对应的文字特征，其中包括第一语音状态对应的文字特征，并从第一语音状态对应的文字特征开始向前端和后端的文字特征依次得到对应的拼音特征，直到获得所有文字特征对应的拼音特征；Or obtain corresponding text features according to a plurality of sequentially arranged voice states, including text features corresponding to the first voice state, and sequentially obtain corresponding pinyin features from the text features corresponding to the first voice state to the text features at the front end and the back end , until the pinyin features corresponding to all text features are obtained;
该拼音特征还标注有音调特征，音调特征用于结合拼音特征得到对应的语义信息。The pinyin feature is also marked with a pitch feature, and the pitch feature is used to obtain corresponding semantic information in combination with the pinyin feature.
所述的驾驶员自定义语音命令多功能同步实现方法，包括以下步骤：The multi-function synchronization implementation method of the driver's self-defined voice command includes the following steps:
(1)用户通过移动智能终端预设需要的执行功能需求，移动智能终端将用户预设的执行功能需求对应生成json文件并上传至云端储存解析；(1) The user presets the required execution function requirements through the mobile intelligent terminal, and the mobile intelligent terminal generates a json file corresponding to the user's preset execution function requirements and uploads it to the cloud for storage and analysis;
(2)云端对接收得到的json文件进行语义识别得到用户的语义信息结果并进行储存；(2) The cloud performs semantic recognition on the received json file to obtain the user's semantic information result and stores it;
(3)用户在车端发出语音控制指令，由车端接收后将该语音控制指令上传至云端；(3) The user sends a voice control command at the car end, and the voice control command is uploaded to the cloud after being received by the car end;
(4)云端对接收得到的语音控制指令进行语义识别得到新的语义信息结果，云端将该新的语义信息结果与储存的用户预设的语义信息结果进行比对，在比对得到符合的结果信息后，将对应的预设功能操作执行指令发送回车端；(4) The cloud performs semantic recognition on the received voice control instructions to obtain a new semantic information result, and the cloud compares the new semantic information result with the stored semantic information result preset by the user, and obtains a consistent result after the comparison. After receiving the information, send the corresponding preset function operation execution instruction to the carriage return terminal;
(5)车端在接收到云端发送的操作执行指令后，控制车机TBox执行相应的功能。(5) After receiving the operation execution instruction sent by the cloud, the vehicle terminal controls the vehicle machine TBox to execute the corresponding function.
通过上述技术方案即可实现用户根据自身需求对车机功能进行场景应用预设，从而在用户选择某一场景时，车机能够根据用户预设的内容自动执行相应的功能，有效提高了用户的使用体验。Through the above technical solutions, the user can preset the functions of the vehicle and the vehicle according to their own needs, so that when the user selects a certain scene, the vehicle can automatically execute the corresponding function according to the content preset by the user, which effectively improves the user's experience. Use experience.
优选的，在步骤(1)中，用户在通过移动智能终端预设需要的执行功能需求时，移动智能终端向用户给出提示，该提示包括三部分：Preferably, in step (1), when the user presets the required execution function requirements through the mobile intelligent terminal, the mobile intelligent terminal gives a prompt to the user, and the prompt includes three parts:
Part.1：车端所有可执行功能的提示；Part.1: Tips for all executable functions on the vehicle end;
Part.2：用户历史预设执行功能的提示；Part.2: Prompt for user history preset execution function;
Part.3：与用户已预设执行功能相关的执行功能的提示。Part.3: Prompts of executive functions related to the user's preset executive functions.
上述方案都可以极大地简化用户的使用复杂度，通过提示的方式来帮助用户选择合适的预设功能。The above solutions can greatly simplify the user's use complexity, and help the user select an appropriate preset function by means of prompts.
优选的，在步骤(5)中，车端根据用户在移动智能终端预设的执行功能需求顺序控制车机TBox执行相应的功能，用户在移动智能终端可对预设的执行功能需求的顺序进行调整。Preferably, in step (5), the vehicle terminal sequentially controls the vehicle-machine TBox to execute corresponding functions according to the user's preset execution function requirements in the mobile intelligent terminal, and the user can perform the preset execution function requirements in the mobile intelligent terminal. Adjustment.
与现有技术相比，本方案具备的显著优点有：Compared with the prior art, the significant advantages of this solution are:
1、用户可通过移动终端进行功能预设操作，使用方便快捷；1. The user can perform function preset operation through the mobile terminal, which is convenient and quick to use;
2、用户可对车机功能进行场景预设，提高了使用体验以及功能多样性；2. Users can preset scenes for the functions of the car and machine, which improves the user experience and functional diversity;
3、整体定制化的服务可有效提高用户的使用体验，并帮助用户完成多项功能操作。3. The overall customized service can effectively improve the user experience and help users complete multiple functional operations.
附图说明Description of drawings
附图用来提供对本发明的进一步理解，并且构成说明书的一部分，与本发明的实施例一起用于解释本发明，并不构成对本发明的限制。在附图中：The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the attached image:
图1为本发明中同步实现系统的模块结构图；Fig. 1 is the module structure diagram of synchronous realization system in the present invention;
图2为本发明中同步实现方法的步骤流程图。FIG. 2 is a flow chart of the steps of the synchronization implementation method in the present invention.
具体实施方式Detailed ways
以下结合附图对本发明的优选实例进行说明，应当理解，此处所描述的优选实例仅用于说明和解释本发明，并不用于限定本发明。The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, but not to limit the present invention.
如图1所示，本发明所述的驾驶员自定义语音命令多功能同步实现系统，包括场景定义模块、语音接收模块、语义分析模块以及MCU；As shown in Figure 1, the driver's self-defined voice command multi-function synchronization implementation system of the present invention includes a scene definition module, a voice receiving module, a semantic analysis module and an MCU;
场景定义模块位于用户的移动智能终端，场景定义模块根据用户的预设功能对应生成json文件，并上传至云端储存解析。The scene definition module is located in the user's mobile smart terminal, and the scene definition module generates a corresponding json file according to the user's preset function, and uploads it to the cloud for storage and analysis.
具体的，该模块包括功能区以及预设区，功能区包括可通过车机TBox执行的所有功能单元，用户可通过拖动的方式将功能单元从功能区拖动到预设区，预设区内存留的功能单元即为用户的预设功能并形成为json文件。Specifically, the module includes a functional area and a preset area. The functional area includes all functional units that can be executed through the TBox of the vehicle. The user can drag the functional units from the functional area to the preset area by dragging. The functional unit remaining in the memory is the user's preset function and is formed into a json file.
以听歌场景为例，用户通过通过场景定义模块，即手机APP中的功能，新增自定义场景，场景名称为“听歌场景”，然后在功能区内可以看到所有可实现的车机功能，用户将需要的功能从功能区拖动到预设区内，比如“关闭车窗”、“关闭天窗”、“打开在线音乐”、“打开空调”、“风量设为一级”、“温度设为23℃”等，随后点击确定，即可生成该场景的语义命令，每一个小功能生产一个子json串：Taking the listening song scene as an example, the user adds a custom scene through the scene definition module, that is, the function in the mobile APP, the scene name is "listening song scene", and then in the function area, all the vehicles that can be realized can be seen. function, the user drags the desired function from the functional area to the preset area, such as "close the window", "close the sunroof", "open online music", "turn on the air conditioner", "set the air volume to the first level", " Set the temperature to 23°C", etc., and then click OK to generate the semantic command of the scene, and each small function produces a sub-json string:
关闭车窗：Close the windows:
{"action":"execute","domain":"win_control","intention":"instructing","type":"close_win"}；{"action":"execute","domain":"win_control","intention":"instructing","type":"close_win"};
关闭天窗：To close the sunroof:
{"action":"execute","domain":"sunroof_control","intention":"instructing","type":"close_sunroof"}；{"action":"execute","domain":"sunroof_control","intention":"instructing","type":"close_sunroof"};
打开在线音乐：Open online music:
{"action":"execute","domain":"media","intention":"instructing","type":"open_net_music"}；{"action":"execute","domain":"media","intention":"instructing","type":"open_net_music"};
打开空调：on the aircon:
{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac"}；{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac"};
风量设为一级：The air volume is set to one level:
"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{speed:1}"}]；"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{speed:1}"}];
温度设为23℃：Set the temperature to 23°C:
{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{temperature:23}"}；{"action":"execute","domain":"ac_control","intention":"instructing","type":"open_ac","semantic":"{temperature:23}"};
上述全部json内容参照【听歌场景.json】文件。For all the above json content, please refer to the [Listening to Song Scene.json] file.
功能单元在预设区内由上至下排列，功能单元的上下顺序可调节，MCU根据功能单元的排序控制车机TBox依次执行相应的功能，这样可以方便用户通过简单的顺序调换即可达到控制车机功能工作顺序的控制需求，同时操作简便，极大地丰富了系统的功能性，提高了用户的使用体验The functional units are arranged from top to bottom in the preset area, and the upper and lower order of the functional units can be adjusted. The MCU controls the TBox of the vehicle to execute the corresponding functions in sequence according to the sequence of the functional units, which is convenient for users to achieve control by simply changing the sequence. The control requirements of the working sequence of the functions of the vehicle and the machine are easy to operate, which greatly enriches the functionality of the system and improves the user experience.
语音接收模块位于车端，用于接收用户在车内时发出的语音控制指令并上传至云端，该模块包括设置在驾驶位的多个麦克风阵列，由此可通过多个麦克风阵列实现对驾驶员的语音指令收集。The voice receiving module is located at the end of the car, and is used to receive the voice control commands issued by the user when they are in the car and upload them to the cloud. collection of voice commands.
语义分析模块位于云端，语义分析模块对场景定义模块上传的json文件进行语义识别，并对语音接收模块上传的语音控制指令同样进行语义识别，在将两者语义识别为符合后向MCU发送对应的操作执行指令；具体的，该模块包括获取单元和识别单元，获取单元获取语音信息并根据语音信息提取语音状态，识别单元将提取得到的语音状态输入目标语义识别模型中，目标语义识别模型根据语音状态获得拼音特征、或者拼音特征和文字特征，得到与语音信息相对应的语义信息，由此通过语义分析模块实现了对获取得到的驾驶员发出的语音指令的语音信息分析，从而确定驾驶员此时需要实现什么样的车机功能。The semantic analysis module is located in the cloud. The semantic analysis module performs semantic recognition on the json file uploaded by the scene definition module, and also performs semantic recognition on the voice control instructions uploaded by the voice receiving module. operation execution instruction; specifically, the module includes an acquisition unit and a recognition unit, the acquisition unit acquires voice information and extracts the voice state according to the voice information, the recognition unit inputs the extracted voice state into the target semantic recognition model, and the target semantic recognition model according to the voice The state obtains pinyin features, or pinyin features and text features, and obtains semantic information corresponding to the voice information, thereby realizing the voice information analysis of the obtained voice commands issued by the driver through the semantic analysis module, so as to determine the driver's voice information. What kind of car function needs to be realized.
识别单元可获取训练数据集，并将该训练数据集输入至初始语义识别模型中，初始语义识别模型包括拼音转换支路和匹配支路，拼音转换支路根据语音状态获得的拼音特征、或者拼音特征和文字特征，结合匹配支路根据拼音特征得到的对应语义信息，得到目标语义识别模型。The recognition unit can obtain the training data set, and input the training data set into the initial semantic recognition model, the initial semantic recognition model includes a pinyin conversion branch and a matching branch, and the pinyin conversion branch obtains the pinyin feature according to the phonetic state, or the pinyin Features and text features, combined with the corresponding semantic information obtained by the matching branch according to the pinyin features, to obtain the target semantic recognition model.
此外，根据语音状态获得的拼音特征、或者拼音特征和文字特征包括：In addition, the pinyin features, or pinyin features and text features obtained according to the voice state include:
根据多个依次排列的语音状态，依次得到每个语音状态对应的文字特征，并根据文字特征依次得到对应的拼音特征；According to a plurality of sequentially arranged voice states, the text features corresponding to each voice state are sequentially obtained, and the corresponding pinyin features are obtained in turn according to the text features;
或者根据多个依次排列的语音状态得到对应的文字特征，其中包括第一语音状态对应的文字特征，并从第一语音状态对应的文字特征开始向前端和后端的文字特征依次得到对应的拼音特征，直到获得所有文字特征对应的拼音特征；Or obtain corresponding text features according to a plurality of sequentially arranged voice states, including text features corresponding to the first voice state, and sequentially obtain corresponding pinyin features from the text features corresponding to the first voice state to the text features at the front end and the back end , until the pinyin features corresponding to all text features are obtained;
该拼音特征还标注有音调特征，音调特征用于结合拼音特征得到对应的语义信息。The pinyin feature is also marked with a pitch feature, and the pitch feature is used to obtain corresponding semantic information in combination with the pinyin feature.
MCU位于车端，MCU在接收到来自语义分析模块发送的操作执行指令后，控制车机TBox执行相应的功能。The MCU is located at the vehicle end. After the MCU receives the operation execution instruction sent from the semantic analysis module, it controls the vehicle machine TBox to perform the corresponding function.
在上述技术方案中，用户可首先通过移动智能终端对特定场景下的车机功能执行需求进行预设，比如用户可以预设在“打开空调”模式下，同步控制车内车窗关闭、天窗关闭等等，随后通过云端的语义分析模块对用户预设的内容进行储存。在用户通过车端的语音接收模块发出语音控制指令后，云端的语义分析模块可以对该指令进行语义分析，以确认用户的实际功能需求，随后对储存的预设内容进行比对，根据比对结果向车端发送相对应的操作执行指令，最后在车端完成用户预设的功能需求。In the above technical solution, the user can first preset the vehicle function execution requirements in a specific scenario through the mobile smart terminal. For example, the user can preset the "turn on the air conditioner" mode to synchronously control the closing of the car window and the sunroof in the car. And so on, and then store the content preset by the user through the semantic analysis module in the cloud. After the user sends a voice control command through the voice receiving module on the vehicle end, the semantic analysis module in the cloud can perform semantic analysis on the command to confirm the user's actual functional requirements, and then compare the stored preset content, according to the comparison result. The corresponding operation execution instructions are sent to the vehicle end, and finally the user's preset functional requirements are completed at the vehicle end.
如图2所示，本发明所述的驾驶员自定义语音命令多功能同步实现方法，包括以下步骤：As shown in Figure 2, the multi-function synchronization implementation method of the driver's self-defined voice command according to the present invention comprises the following steps:
(1)用户通过移动智能终端预设需要的执行功能需求，移动智能终端将用户预设的执行功能需求对应生成json文件并上传至云端储存解析；(1) The user presets the required execution function requirements through the mobile intelligent terminal, and the mobile intelligent terminal generates a json file corresponding to the user's preset execution function requirements and uploads it to the cloud for storage and analysis;
用户在通过移动智能终端预设需要的执行功能需求时，移动智能终端向用户给出提示，该提示包括三部分：When the user presets the required execution function requirements through the mobile intelligent terminal, the mobile intelligent terminal gives a prompt to the user, and the prompt includes three parts:
Part.1：车端所有可执行功能的提示；Part.1: Tips for all executable functions on the vehicle end;
Part.2：用户历史预设执行功能的提示；Part.2: Prompt for user history preset execution function;
Part.3：与用户已预设执行功能相关的执行功能的提示。Part.3: Prompts of executive functions related to the user's preset executive functions.
上述方案都可以极大地简化用户的使用复杂度，通过提示的方式来帮助用户选择合适的预设功能；All of the above solutions can greatly simplify the user's use complexity, and help users select appropriate preset functions through prompts;
(2)云端对接收得到的json文件进行语义识别得到用户的语义信息结果并进行储存；(2) The cloud performs semantic recognition on the received json file to obtain the user's semantic information result and stores it;
(3)用户在车端发出语音控制指令，由车端接收后将该语音控制指令上传至云端；(3) The user sends a voice control command at the car end, and the voice control command is uploaded to the cloud after being received by the car end;
(4)云端对接收得到的语音控制指令进行语义识别得到新的语义信息结果，云端将该新的语义信息结果与储存的用户预设的语义信息结果进行比对，在比对得到符合的结果信息后，将对应的预设功能操作执行指令发送回车端；(4) The cloud performs semantic recognition on the received voice control instructions to obtain a new semantic information result, and the cloud compares the new semantic information result with the stored semantic information result preset by the user, and obtains a consistent result after the comparison. After receiving the information, send the corresponding preset function operation execution instruction to the carriage return terminal;
(5)车端在接收到云端发送的操作执行指令后，控制车机TBox执行相应的功能，车端根据用户在移动智能终端预设的执行功能需求顺序控制车机TBox执行相应的功能，用户在移动智能终端可对预设的执行功能需求的顺序进行调整。(5) After receiving the operation execution instruction sent by the cloud, the vehicle terminal controls the vehicle TBox to execute the corresponding function. The preset order of executing functional requirements can be adjusted on the mobile smart terminal.
通过上述技术方案即可实现用户根据自身需求对车机功能进行场景应用预设，从而在用户选择某一场景时，车机能够根据用户预设的内容自动执行相应的功能，有效提高了用户的使用体验。Through the above technical solutions, the user can preset the functions of the vehicle and the vehicle according to their own needs, so that when the user selects a certain scene, the vehicle can automatically execute the corresponding function according to the content preset by the user, which effectively improves the user's experience. Use experience.
最后应说明的是：以上所述仅为本发明的优选实例而已，并不用于限制本发明，尽管参照前述实施例对本发明进行了详细的说明，对于本领域的技术人员来说，其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换。凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。Finally, it should be noted that the above descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features therein. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210890758.1ACN115148189A (en) | 2022-07-27 | 2022-07-27 | Multifunctional synchronous implementation system and method for driver-defined voice command |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210890758.1ACN115148189A (en) | 2022-07-27 | 2022-07-27 | Multifunctional synchronous implementation system and method for driver-defined voice command |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115148189Atrue CN115148189A (en) | 2022-10-04 |
Family
ID=83414662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210890758.1APendingCN115148189A (en) | 2022-07-27 | 2022-07-27 | Multifunctional synchronous implementation system and method for driver-defined voice command |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115148189A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118798618A (en)* | 2024-09-11 | 2024-10-18 | 浙江吉利控股集团有限公司 | A method, device, medium and program product for generating vehicle workflow |
| CN119296538A (en)* | 2024-11-27 | 2025-01-10 | 深圳市特星球信息科技有限公司 | A vehicle-mounted aftermarket voice control interaction method |
| CN119763552A (en)* | 2024-12-20 | 2025-04-04 | 广州小鹏汽车科技有限公司 | Model training method, voice interaction method, server and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104965596A (en)* | 2015-07-24 | 2015-10-07 | 上海宝宏软件有限公司 | Voice control system |
| US20170358300A1 (en)* | 2016-06-10 | 2017-12-14 | Apple Inc. | Digital assistant providing automated status report |
| CN111192572A (en)* | 2019-12-31 | 2020-05-22 | 斑马网络技术有限公司 | Semantic recognition method, device and system |
| CN111883118A (en)* | 2020-07-09 | 2020-11-03 | 浙江吉利汽车研究院有限公司 | A vehicle control method, device and storage medium based on personalized voice |
| CN113589938A (en)* | 2021-08-04 | 2021-11-02 | 安徽江淮汽车集团股份有限公司 | Vehicle-mounted terminal information sharing system with bullet screen function |
| CN114326484A (en)* | 2021-12-10 | 2022-04-12 | 重庆长安汽车股份有限公司 | Scene type remote vehicle control system and method based on mobile phone voice |
| EP4030423A1 (en)* | 2021-01-15 | 2022-07-20 | Volvo Car Corporation | Vehicle, voice control device and method |
- 2022
- 2022-07-27CNCN202210890758.1Apatent/CN115148189A/enactivePending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104965596A (en)* | 2015-07-24 | 2015-10-07 | 上海宝宏软件有限公司 | Voice control system |
| US20170358300A1 (en)* | 2016-06-10 | 2017-12-14 | Apple Inc. | Digital assistant providing automated status report |
| CN111192572A (en)* | 2019-12-31 | 2020-05-22 | 斑马网络技术有限公司 | Semantic recognition method, device and system |
| CN111883118A (en)* | 2020-07-09 | 2020-11-03 | 浙江吉利汽车研究院有限公司 | A vehicle control method, device and storage medium based on personalized voice |
| EP4030423A1 (en)* | 2021-01-15 | 2022-07-20 | Volvo Car Corporation | Vehicle, voice control device and method |
| CN113589938A (en)* | 2021-08-04 | 2021-11-02 | 安徽江淮汽车集团股份有限公司 | Vehicle-mounted terminal information sharing system with bullet screen function |
| CN114326484A (en)* | 2021-12-10 | 2022-04-12 | 重庆长安汽车股份有限公司 | Scene type remote vehicle control system and method based on mobile phone voice |
Non-Patent Citations (1)
| Title |
|---|
| 陈功等: "《Creo Parametric2.0三维造型及应用》", 31 July 2014, 东南大学出版社, pages: 29 - 30* |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118798618A (en)* | 2024-09-11 | 2024-10-18 | 浙江吉利控股集团有限公司 | A method, device, medium and program product for generating vehicle workflow |
| CN119296538A (en)* | 2024-11-27 | 2025-01-10 | 深圳市特星球信息科技有限公司 | A vehicle-mounted aftermarket voice control interaction method |
| CN119763552A (en)* | 2024-12-20 | 2025-04-04 | 广州小鹏汽车科技有限公司 | Model training method, voice interaction method, server and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115148189A (en) | Multifunctional synchronous implementation system and method for driver-defined voice command | |
| CN105989841B (en) | Vehicle-mounted voice control method and device | |
| CN111885547B (en) | A vehicle-mounted human-computer interaction system | |
| KR102249392B1 (en) | Apparatus and method for controlling device of vehicle for user customized service | |
| CN110435660A (en) | A kind of autocontrol method and device of vehicle drive contextual model | |
| CN106379310A (en) | System and method for controlling cruising of vehicles by voice | |
| US10198696B2 (en) | Apparatus and methods for converting user input accurately to a particular system function | |
| CN112721565A (en) | Vehicle-mounted air conditioner control method, vehicle and medium | |
| CN208827749U (en) | Vehicle-mounted voice control device and system | |
| CN114299947A (en) | Generation method, vehicle control method, terminal device and storage medium | |
| CN118314894A (en) | Reasoning dialogue type voice interaction system and method based on large model | |
| CN116567895A (en) | Vehicle ambient light control method, device, electronic device and vehicle | |
| CN111506229A (en) | Interaction method and device and vehicle | |
| CN112519639B (en) | Vehicle control method, system, device, vehicle and storage medium | |
| CN106114518A (en) | Cruise system that a kind of phonetic entry is controlled and vehicle | |
| CN118782044A (en) | Multimodal interaction method, device, electronic device and storage medium | |
| CN112576137B (en) | Vehicle window control method and device, vehicle and storage medium | |
| CN113160808B (en) | A voice control method and system, and a voice control device | |
| CN114581096A (en) | Cabin control method and device of network car, cloud server and storage medium | |
| CN204619374U (en) | Utilize voice-operated remote control Multi-axis aircraft toy | |
| CN114987367B (en) | In-vehicle karaoke system, method and storage medium with in-vehicle and out-of-vehicle singing function | |
| JP2020060730A (en) | Voice recognition device | |
| CN215069236U (en) | Voice-controlled cosmetic mirror | |
| CN115904204A (en) | Control method of programmable gestures of new energy automobile, storage medium and automobile | |
| DE102022213306A1 (en) | VOICE RECOGNITION SYSTEM AND METHOD OF PROVIDING A VOICE RECOGNITION SERVICE |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20221004 | |
| RJ01 | Rejection of invention patent application after publication |