WO2019101067A1 - 用于数据可视化的信息的处理方法以及装置 - Google Patents

用于数据可视化的信息的处理方法以及装置 Download PDF

Info

Publication number
WO2019101067A1
WO2019101067A1 PCT/CN2018/116415 CN2018116415W WO2019101067A1 WO 2019101067 A1 WO2019101067 A1 WO 2019101067A1 CN 2018116415 W CN2018116415 W CN 2018116415W WO 2019101067 A1 WO2019101067 A1 WO 2019101067A1
Authority
WO
WIPO (PCT)
Prior art keywords
input information
information
recognition result
database
determining
Prior art date
Application number
PCT/CN2018/116415
Other languages
English (en)
French (fr)
Inventor
徐海燕
周宁奕
朱颖华
许天宇
Original Assignee
众安信息技术服务有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 众安信息技术服务有限公司 filed Critical 众安信息技术服务有限公司
Priority to JP2019542091A priority Critical patent/JP6887508B2/ja
Priority to KR1020197023144A priority patent/KR20190107063A/ko
Priority to US16/354,678 priority patent/US20190213998A1/en
Publication of WO2019101067A1 publication Critical patent/WO2019101067A1/zh

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
    • G10L15/04—Segmentation; Word boundary detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/338—Presentation of query results
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • 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
    • G10L15/14—Speech classification or search using statistical models, e.g. Hidden Markov Models [HMMs]
    • G10L15/142—Hidden Markov Models [HMMs]
    • 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
    • G10L15/18—Speech classification or search using natural language modelling
    • G10L15/1815—Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
    • 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
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/225—Feedback of the input speech

Definitions

  • the invention belongs to the field of computer data processing, and in particular relates to a method and a device for processing information for data visualization.
  • Data visualization is a study of the visual representation of data. It is more helpful for people to understand the data from a visual perspective than to obtain information by word-by-line reading of text.
  • the current data positioning interaction mode is mainly through mouse click or touch screen to click on the screen to interact, the learning cost is relatively high, and it is not conducive to long-distance data visualization display operation, which is not convenient and fast.
  • the present invention addresses the above problems, and proposes an interactive manner of natural language processing and information positioning display, which not only improves the efficiency of human-computer interaction during data display, but also effectively enhances the visual display of specific scenes such as large screens. Visual display of the scene.
  • An aspect of the present invention provides a method of processing information for data visualization, comprising: performing identifiability analysis on the received input information; determining whether the input information is correctly identified, when the input information is correctly At the time of recognition, an interaction instruction corresponding to the recognition result is determined based on the recognition result of the input information, thereby executing the interaction instruction.
  • the determining whether the input information is correctly identified comprises: converting the input information that can be correctly identified into media information having a specified presentation form, and determining based on the confirmation information of the media information. Whether the input information is correctly identified, and the confirmation information is used to indicate whether the media information correctly presents the input information.
  • the determining, according to the recognition result of the input information, the interaction instruction corresponding to the recognition result comprises: performing the search matching in the database when the recognition result is stored in the database When the data field corresponding to the recognition result is used, an interaction instruction corresponding to the recognition result is directly determined based on the recognition result.
  • the determining the result of the input information to determine the interaction instruction corresponding to the recognition result comprises: performing the search matching in the database when the recognition result is not present in the database When the data field corresponding to the result is identified, a keyword set is determined based on the recognition result, and an interaction instruction corresponding to the recognition result is determined based on the keyword set.
  • the method further includes: determining whether the input information is successfully received when the input information is received, wherein when the input information is not successfully received, generating Receive the first feedback message that failed.
  • the performing the identifiability analysis on the received input information comprises: analyzing the input information based on a recognition model for identifying the input information, thereby determining identifiable of the input information And wherein, when the input information cannot be recognized, generating second feedback information indicating that the input information cannot be recognized.
  • third feedback information is generated for indicating that the input information identifies an error.
  • determining the keyword set based on the recognition result of the input information comprises: identifying the input information as semantic text, extracting the keyword set from the semantic text, wherein the key The set of words includes at least one field.
  • the determining, according to the set of keywords, the interaction instruction corresponding to the recognition result comprises: comparing the data field in the database based on the keyword set; when the key When the field in the word set matches the data field in the database, the interaction instruction is determined based on the matching result; when the field in the keyword set does not match the data field in the database, the fourth feedback is generated.
  • the input information comprises at least one of the following: a voice, a touch, or a body motion.
  • the method further comprises determining whether the input information is successfully received when the input information is received, wherein the input information comprises a voice, wherein determining whether the input information is successful Receiving includes determining whether the voice is successfully received based on the first threshold.
  • the first threshold comprises one or more of the following: a speech length threshold, a speech intensity threshold, a speech audio domain threshold.
  • the media information includes at least one of the following: video, audio, picture, or text.
  • Another aspect of the present invention is directed to a computer readable storage medium having computer readable program instructions stored thereon that, when executed, implement information for data visualization as described above The steps of the processing method.
  • Another aspect of the present invention also provides an information processing apparatus for data visualization, comprising: a processor; a memory for storing instructions, when the instructions are executed, the processor performs the aforementioned The steps of the processing method for the information to be visualized.
  • the interaction between the user and the data display can be improved in the data visualization scenario, and the unity of the current data visualization interaction mode is improved.
  • FIG. 1 illustrates an information processing method for data visualization according to an embodiment of the present invention
  • FIG. 2 is an information processing method for data visualization based on voice recognition according to an embodiment of the present invention.
  • FIG. 1 illustrates an information processing method for data visualization in accordance with an embodiment of the present invention.
  • the method includes:
  • Step S101 Perform identifiability analysis on the input information.
  • the identifiability of the input information is analyzed, so that the recognition model identifies the identifiable input information.
  • the input information of the user may be, but not limited to, indicative information such as voice, touch or limb movement.
  • the voice is recognized by a voice recognition model.
  • the gesture is recognized by the gesture recognition model.
  • the recognition model can obtain the recognition result of the input information.
  • Step S102 Convert the identified input information into media information to generate confirmation information.
  • the recognition result of the input information or the input information obtained in the previous step is converted into the medium information having the specified presentation form.
  • the user can confirm whether the input information is correctly identified, and then generate corresponding confirmation information.
  • the media information herein may include an image that is visible to the user, a text, or a voice audible to the user, and the media information may have a different form from the input information. In this way, the user can know the recognition result in a variety of ways.
  • Step S103 Determine whether the media information correctly presents the input information based on the confirmation information.
  • the user can judge whether the input information is correctly recognized based on the media information. If the input information is not correctly recognized, feedback information is generated (step S106) to prompt the user to re-enter because the current input information is not correctly recognized.
  • step S104 is performed, that is, the keyword set is determined based on the identified input information, and then the matching is found in the database.
  • the input information is not limited to indicative information such as voice, limb motion, and touch. Therefore, after the recognition system recognizes the input information, the keyword set corresponding to the input information can be determined based on the recognition result.
  • the recognition result is a semantic text corresponding to the input information
  • the keyword set may include at least one field extracted from the semantic text and capable of reflecting the intention of the input information.
  • the keyword set After determining the keyword set, it can be searched in the database based on the fields included in the keyword set to see if there is a corresponding data field. When there is a data field corresponding to the keyword set in the database, the matching between the keyword and the data field in the database can be implemented, thereby determining the interactive instruction corresponding to the keyword. Obviously, by extracting the keyword set, the intent of the input information to be expressed can be determined.
  • Step S105 Determine an interaction instruction according to the matching result, and then perform a corresponding operation.
  • the interaction instruction corresponding to the keyword can be determined.
  • the system will execute the interactive instruction to generate an operation corresponding to the user's input information.
  • the response to various forms of input information of the user in the data visualization scenario can be realized, thereby simplifying the operation and facilitating better display.
  • the following takes the input information as the voice information as an example for description.
  • the method in FIG. 2 takes voice information as an example, the method in FIG. 2 is equally applicable to other forms of input information, including but not limited to limb movements, touches, and the like.
  • the method includes:
  • Step S201 Receive voice input information.
  • an instruction issued by the user will be received through the terminal device, where the terminal device may be a mobile phone, a microphone, or the like that has been matched with the display content.
  • the terminal device is a voice receiving device having the capability of further processing (e.g., identification) of the voice input information
  • the terminal device can process the voice input information according to the settings. If the terminal device is a voice receiving device such as a microphone, the terminal device will transmit the received voice input information to the designated processing device.
  • Step S202 Determine whether the voice input information is successfully received based on the first threshold.
  • the terminal device based on the first threshold, it is determined whether the terminal device successfully receives the voice input information. Due to environmental influences or the working status of the terminal device itself, the terminal device may not be able to receive or receive voice input information completely.
  • the voice length threshold may be set at the terminal device, and when the length of the received voice input information is less than the voice length threshold, it may be determined that the voice input information is invalid information.
  • the voice intensity threshold may also be set. When the strength of the received voice input information is less than the voice intensity threshold, it may be determined that the voice input information is invalid information. It can be understood that, according to the needs of the application, a corresponding threshold can be set for judging, for example, a voice domain threshold.
  • the first threshold may include, but is not limited to, a speech length threshold, a speech intensity threshold, or a speech domain threshold, a combination of the above types of thresholds, and the like.
  • step S204 is performed to send the first feedback information to the user.
  • the first feedback information herein can be any form of information that can be perceived by the user.
  • step S203 is performed to identify the voice input information according to the system model.
  • the system model in this embodiment can adopt any existing speech recognition model, such as a hidden Markov model. Similarly, the system model can also be trained through artificial neural networks.
  • Step S205 It is judged whether the voice input information can be recognized.
  • the identifiability of the received voice input information is judged. For some irregular, unclear or other speech that exceeds the recognition ability of the speech recognition model, speech recognition cannot be achieved even if the speech is successfully received. Therefore, by performing this step, the identifiability of the voice input information can be judged.
  • the step S207 is executed to issue the second feedback information to the user.
  • the second feedback information herein can be any form of information that can be perceived by the user.
  • step S206 is performed to convert the voice input information into media information.
  • the media information herein may include an image that is visible to the user, a text, or a user audible voice. In this way, the user can know the recognition result in a variety of ways.
  • Step S208 determining whether the recognition result of the voice input information is correct?
  • the recognition result of the voice input information is judged.
  • the voice input information is converted into media information, it is possible to determine whether the recognition result is correct in response to the user's confirmation information, wherein the recognition result is a semantic text corresponding to the input information.
  • the system does not need further confirmation by the user, and may choose to judge whether the identification information is correct or not, and thus, step S206 may optionally not be performed.
  • step S207 is executed to issue third feedback information to the user.
  • the third feedback information herein can be any form of information that can be perceived by the user.
  • step S210 or S214 is performed.
  • the following description will be made by taking the recognition result as "I really want to go to Beijing" as an example.
  • Steps S210-S213 are first explained.
  • the recognition result corresponding to the voice input information When the recognition result corresponding to the voice input information is correct, the recognition result may be analyzed (for example, split), and then the keyword associated with the recognition result is determined, for example, according to a specific field or a semantic algorithm. Extract keywords from the recognition results. By extracting the recognition result "I really want to go to Beijing”, I can extract the keywords "I", “Want to go”, "Beijing”. After the above keywords are determined, a lookup match will be made in the database (for example, corpus).
  • the database for example, corpus
  • Step S211 Determine whether the keyword can match the word in the database.
  • step S212 is executed to issue the fourth feedback information to the user.
  • the fourth feedback information herein can be any form of information that can be perceived by the user.
  • step S213 is performed, that is, a corresponding operation is generated according to the result of the matching.
  • the corresponding actions will be triggered based on the keywords "I", “Want to go", “Beijing".
  • the current user may be provided with a route to Beijing, or a flight to a Beijing, a train, and the like, and the availability of alternative vehicles.
  • the user can directly speak the pre-configured device receivable field when performing the data visualization on-site demonstration.
  • the terminal device receives the instruction, it can directly compare with the background data to quickly display the required data on the display terminal. That is to say, if there is already a data field corresponding to the voice "I really want to go to Beijing" at the terminal device or the processing device, then there is no need to perform keyword extraction on the voice at this time, and the data field can be directly executed. Corresponding operation (step S214).
  • speech recognition and natural language processing are realized in the data visualization scenario, which can improve the interaction between the user and the data display, and improve the unity of the current data visualization interaction mode.
  • Users can complete operations through natural language transmission, reduce the complexity of data visualization and interaction, and improve display efficiency, especially suitable for use in large-screen data display scenarios.
  • the above embodiment employs voice input information as an embodiment, those skilled in the art can understand that indicative information such as limb movements, touches, and the like are equally applicable to the above method.
  • the video component in the terminal device captures the action of the user's hands clasped, the action will be identified by the corresponding motion recognition model.
  • the action of the two-handedness can be associated with the "shutdown" function through training, whereby when the motion recognition model correctly recognizes the action, the "shutdown" function is triggered.
  • the flow of the information processing method of Figures 1, 2 also represents machine readable instructions including programs executed by a processor.
  • the program can be embodied in software stored on a tangible computer readable medium such as a CD-ROM, floppy disk, hard disk, digital versatile disk (DVD), Blu-ray disk or other form of memory.
  • a tangible computer readable medium such as a CD-ROM, floppy disk, hard disk, digital versatile disk (DVD), Blu-ray disk or other form of memory.
  • some or all of the example methods in Figures 1, 2 may utilize an application specific integrated circuit (ASIC), programmable logic device (PLD), field programmable logic device (EPLD), discrete logic, hardware, firmware Any combination of the like is implemented.
  • ASIC application specific integrated circuit
  • PLD programmable logic device
  • EPLD field programmable logic device
  • FIGS. 1 and 2 describes the data processing method, the steps in the processing method may be modified, deleted, or merged.
  • the example processes of Figures 1, 2 can be implemented using coded instructions (such as computer readable instructions) stored on a tangible computer readable medium, such as a hard disk, flash memory, read only memory (ROM), optical disk. (CD), digital versatile disc (DVD), cache, random access memory (RAM), and/or any other storage medium on which information can be stored for any time (eg, long, permanent, transient) Situation, temporary buffering, and/or caching of information).
  • a tangible computer readable medium such as a hard disk, flash memory, read only memory (ROM), optical disk. (CD), digital versatile disc (DVD), cache, random access memory (RAM), and/or any other storage medium on which information can be stored for any time (eg, long, permanent, transient) Situation, temporary buffering, and/or caching of information).
  • a tangible computer readable medium such as a hard disk, flash memory, read only memory (ROM), optical disk. (CD), digital versatile disc (DVD), cache, random access memory (RAM), and/or any other storage medium on which information
  • the example processes of Figures 1, 2 may be implemented with encoded instructions (such as computer readable instructions) stored on a non-transitory computer readable medium, such as a hard disk, flash memory, read only memory, optical disk, Digital versatile disc, cache, random access memory and/or any other storage medium in which information can be stored at any time (eg, long time, permanently, transient, temporary buffering, and/or caching of information). ).
  • a non-transitory computer readable medium such as a hard disk, flash memory, read only memory, optical disk, Digital versatile disc, cache, random access memory and/or any other storage medium in which information can be stored at any time (eg, long time, permanently, transient, temporary buffering, and/or caching of information).
  • the computer readable instructions can also be stored in a web server and on a cloud platform for user convenience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种用于数据可视化的信息的处理方法，包括：对输入信息进行可识别性分析（S101）；将经识别的输入信息转换为媒介信息以生成确认信息（S102）；确定媒介信息是否正确呈现了所述输入信息（S103），当所述输入信息被正确识别时，基于经识别的输入信息确定关键字集，以在数据库中进行查找匹配（S104），并根据匹配结果生成交互指令并执行（S105）；当所述输入信息未被正确识别时，生成反馈信息（S106）。该方法在数据可视化场景中能够提升用户与数据展示间的互动性，改善当前数据可视化交互方式的单一性。

Description

用于数据可视化的信息的处理方法以及装置
本申请要求2017年11月21日提交的申请号为No.201711166559.1的中国申请的优先权，通过引用将其全部内容并入本文。
技术领域
本发明属于计算机数据处理领域，尤其涉及一种用于数据可视化的信息的处理方法以及装置。
发明背景
数据可视化是关于数据之视觉表现形式的研究，比起逐字逐行的阅读文字等方式获取信息，更有助于人们从视觉的角度去理解数据。当前的数据定位交互方式，主要是通过鼠标点击或触屏点击屏幕的方式进行交互，学习成本相对较高，也不利于远距离的数据可视化展示操作，不够便捷快速。
因此，亟需一种能够应用在数据可视化场景中快速交互的方法与装置。
发明内容
本发明针对上述问题，提出一种自然语言处理及信息定位展示的交互方式，这种方式不仅能提高数据展示时人机交互的效率，而且在大屏等特定场景进行可视化展示时，能有效提升现场的视觉展示效果。
本发明的一方面提出了一种用于数据可视化的信息的处理方法，包括：对所接收的输入信息进行可识别性分析；确定所述输入信息是否被正确识别，当所述输入信息被正确识别时，基于所述输入信息的识别结果来确定与所述识别结果相对应的交互指令，进而执行所述交互指令。
在一种实施例中，所述确定所述输入信息是否被正确识别包括：将能够正确识别的所述输入信息转换为具有指定呈现形式的媒介信息，并基于所述媒介信息的确认信息来确定所述输入信息是否被正确识别，所述确认信息用于指示所述媒 介信息是否正确地呈现了所述输入信息。
在一种实施例中，所述基于所述输入信息的识别结果来确定与所述识别结果相对应的交互指令包括：将所述识别结果在数据库中进行查找匹配，当所述数据库中存有与所述识别结果相对应的数据字段时，基于所述识别结果直接确定与所述识别结果相对应的交互指令。
在一种实施例中，所述输入信息的识别结果来确定与所述识别结果相对应的交互指令包括：将所述识别结果在数据库中进行查找匹配，当所述数据库中不存在与所述识别结果相对应的数据字段时，基于所述识别结果来确定关键字集，基于所述关键字集来确定与所述识别结果相对应的交互指令。
在一种实施方式中，所述方法还包括：当对所述输入信息进行接收时，判断所述输入信息是否被成功接收，其中，当所述输入信息未被成功接收，则生成用于指示接收失败的第一反馈信息。
在一种实施方式中，所述对接收的输入信息进行可识别性分析包括：基于用于识别所述输入信息的识别模型来对所述输入信息进行分析，进而确定所述输入信息的可识别性，其中，当所述输入信息无法被识别时，生成用于指示所述输入信息无法被识别的第二反馈信息。
在一种实施方式中，当所述输入信息未被正确识别时，产生用于指示所述输入信息识别错误的第三反馈信息。
在一种实施方式中，基于所述输入信息的识别结果来确定关键字集包括：将所述输入信息识别为语义文本，从所述语义文本中抽取所述关键字集，其中，所述关键字集包括至少一个字段。
在一种实施方式中，所述基于所述关键字集来确定与所述识别结果相对应的交互指令包括：基于所述关键字集来与数据库中的数据字段进行比对；当所述关键字集中的字段与所述数据库中的数据字段相匹配时，基于匹配结果来确定所述交互指令；当所述关键字集中的字段与所述数据库中的数据字段不匹配时，生成第四反馈信息，其中，所述第四反馈信息用于指示所述关键字集中的字段与所述数据库中的数据字段无法匹配。
在一种实施例中，所述输入信息包括以下项中的至少一项：语音、触摸或者肢体动作。
在一种实施例中，所述方法还包括当对所述输入信息进行接收时，判断所述输入信息是否被成功接收，其中所述输入信息包括语音，其中，判断所述输入信息是否被成功接收包括：基于第一阈值来判断所述语音是否被成功接收。
在进一步的实施例中，第一阈值包括以下项中的一种或者多种组合：语音长度阈值、语音强度阈值、语音频域阈值。
在一种实施方式中，所述媒介信息包括以下项中的至少一项：视频、音频、图片或文字。
本发明的另一方面提出了一种计算机可读存储介质，具有存储在其上的计算机可读程序指令，所述计算机可读程序指令执行时，实现如前述所述的用于数据可视化的信息的处理方法的步骤。
本发明的另一方面还提出了一种用于数据可视化的信息处理装置，包括：处理器；存储器，其用于存储指令，当所述指令在执行时，所述处理器执行前述所述的用于可视化的信息的处理方法的步骤。
通过实施本发明的技术方案，在数据可视化场景中能够提升用户与数据展示间的互动性，改善当前数据可视化交互方式的单一性。
附图简要说明
参考附图示出并阐明实施例。这些附图用于阐明基本原理，从而仅仅示出了对于理解基本原理必要的方面。这些附图不是按比例的。在附图中，相同的附图标记表示相似的特征。
图1示出了依据本发明实施例的用于数据可视化的信息处理方法；
图2为依据本发明实施例的基于语音识别的数据可视化的信息处理方法。
实施本发明的方式
在以下优选的实施例的具体描述中，将参考构成本发明一部分的所附的附图。 所附的附图通过示例的方式示出了能够实现本发明的特定的实施例。示例的实施例并不旨在穷尽根据本发明的所有实施例。可以理解，在不偏离本发明的范围的前提下，可以利用其他实施例，也可以进行结构性或者逻辑性的修改。因此，以下的具体描述并非限制性的，且本发明的范围由所附的权利要求所限定。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为说明书的一部分。对于附图中的各单元之间的连线，仅仅是为了便于说明，其表示至少连线两端的单元是相互通信的，并非旨在限制未连线的单元之间无法通信。
下面结合附图对本发明基于数据可视化场景，进行自然语言处理及信息定位展示的交互方式作进一步详细描述。
图1示出了依据本发明实施例的用于数据可视化的信息处理方法。该方法包括：
步骤S101：对输入信息进行可识别性分析。
在该步骤中，将对输入信息的可识别性进行分析，进而使得识别模型对可识别的输入信息进行识别。可以理解的，用户的输入信息可以是但不限于语音、触摸或肢体动作等具有指示性的信息。譬如，当用户输入语音时，将通过语音识别模型对语音进行识别。同样，当用户输入手势时，将通过手势识别模型对手势进行识别。通过执行本步骤，识别模型可以获得该输入信息的识别结果。
步骤S102：将经识别的输入信息转换为媒介信息，以生成确认信息。
在该步骤中，将前一步骤所获得的输入信息或输入信息的识别结果转换为具有指定呈现形式的媒介信息。通过执行该步骤，可以让用户进行确认该输入信息是否被正确识别，进而生成相应的确认信息。可以理解的，此处的媒介信息可以包括用户可视的图像、文字或是用户可听的语音等，媒介信息可以与输入信息具有不同的形式。如此，用户可以通过多种方式来得知该识别结果。
步骤S103：基于确认信息确定媒介信息是否正确呈现输入信息。
在该步骤中，用户可以根据媒介信息来判断输入信息是否被正确识别。若输入信息并未被正确识别，则生成反馈信息(步骤S106)，以提示用户由于当前的 输入信息并未被正确识别，可以进行重新输入。
若输入信息被正确识别，则执行步骤S104，即基于经识别的输入信息确定关键字集，进而在数据库中查找匹配。
由前述可知，输入信息不限于语音、肢体动作、触摸等具有指示性的信息。因此，识别系统识别输入信息之后，可以基于识别结果来确定对应于输入信息的关键字集。在本实施例中，识别结果则是与该输入信息相对应的语义文本，关键字集则可以包括从该语义文本中抽取并且能够反应该输入信息的意图的至少一个字段。
确定关键字集后，可以基于关键字集中所包括的字段在数据库中进行查找是否存在与之对应的数据字段。当数据库中存在与关键字集对应的数据字段时，便可以实现关键字与数据库中数据字段之间的匹配，进而确定该关键字所对应的交互指令。显然，通过对关键字集的提取，可以确定该输入信息所要表达的意图。
步骤S105：根据匹配结果，确定交互指令，进而执行相应的操作。
由前一步骤可知，当关键字能够与数据库中的数据字段进行匹配时，能够确定该关键字所对应的交互指令。当该交互指令被确定时，系统将执行该交互指令，以产生与用户的输入信息相对应的操作。
通过执行图1中的信息处理方法，可以实现在数据可视化场景下的对用户多种形式的输入信息的响应，从而简化了操作，便于更好地展示。
为了进一步描述本实施例，下面结合图2，以输入信息为语音信息为例进行阐述。本领域技术人员可以理解的是，虽然图2中的方法以语音信息为例，但图2中的方法也同样适用于其它形式的输入信息，包括但不限于肢体动作、触摸等等。
图2为依据本发明实施例的基于语音识别的数据可视化的信息处理方法。该方法包括：
步骤S201：接收语音输入信息。
在该步骤中，将通过终端设备接收用户所发出的指令，这里的终端设备可以是与展示内容已匹配过的手机、麦克风等。当终端设备是具有能够对该语音输入 信息进行进一步处理(譬如，识别)的能力的语音接收设备时，该终端设备可以根据设置对该语音输入信息进行处理。若该终端设备是麦克风之类的语音接收设备时，该终端设备将把所接收的语音输入信息传送至指定的处理设备处。
步骤S202：基于第一阈值判断是否成功接收语音输入信息。
在该步骤中，将基于第一阈值，对终端设备是否成功接收该语音输入信息进行判断。由于环境影响或是终端设备自身的工作状态的影响，终端设备可能无法接收或是无法完全接收语音输入信息。譬如，可以在终端设备处设置语音长度阈值，当接收到的语音输入信息的长度小于该语音长度阈值时，可以判断为该语音输入信息是无效信息。同样，还可以设置语音强度阈值，当接收到的语音输入信息的强度小于该语音强度阈值时，可以判断为该语音输入信息是无效信息。可以理解的，根据应用的需要，可以设置相应的阈值进行判断，譬如，语音频域阈值。本实施例无需对所有可能实现的方式仅仅枚举。经过执行此步骤，可以对语音输入信息的接收进行判断。由上可知，第一阈值可以包括但不限于语音长度阈值、语音强度阈值或语音频域阈值，也可以是上述类型阈值的组合等等。
当步骤S202的判断结果为否时，即此时并未成功接收语音输入信息，则执行步骤S204，向用户发出第一反馈信息。可以理解的，这里的第一反馈信息可以是任何形式的能够让用户感知的信息。
当步骤S202的判断结果为是时，即此时成功地接收了语音输入信息，则执行步骤S203，根据系统模型对该语音输入信息进行识别。本实施例中的系统模型可以采用现有的任意一种语音识别模型，譬如，隐马尔可夫模型。同样，该系统模型还可以是通过人工神经网络进行训练所得。
步骤S205：判断是否能够识别该语音输入信息。
在该步骤中，将对该接收的语音输入信息的可识别性进行判断。对于一些不规则的、不清楚的或是其它超出了语音识别模型的识别能力的语音，即使该些语音被成功接收，也无法实现语音识别。因此，经过执行此步骤，可以对语音输入信息的可识别性进行判断。
当步骤S205的判断结果为否时，即此时无法对语音输入信息进行识别，则执 行步骤S207，向用户发出第二反馈信息。可以理解的，这里的第二反馈信息可以是任何形式的能够让用户感知的信息。
当步骤S205的判断结果为是时，即此时成功地能够对该语音输入信息进行识别，则执行步骤S206，将该语音输入信息转换为媒介信息。可以理解的，此处的媒介信息可以包括用户可视的图像、文字或是用户可听的语音等。如此，用户可以通过多种方式来得知该识别结果。
步骤S208：判断该语音输入信息的识别结果是否正确？
在该步骤中，将对该语音输入信息的识别结果进行判断。在本实施例中，由于该语音输入信息被转换为媒介信息，因此，可以响应于用户的确认信息来判断识别结果是否正确，其中，识别结果是与该输入信息相对应的语义文本。
可以理解的，在其它实施例中，系统无需用户的进一步确认，可以选择自行判断识别信息是否正确，如此，步骤S206可以选择性地无需执行。
当步骤S208的判断结果为否时，即此时对应于该语音输入信息的识别结果是错误的，则执行步骤S207，向用户发出第三反馈信息。可以理解的，这里的第三反馈信息可以是任何形式的能够让用户感知的信息。
当步骤S208的判断结果为是时，即此时对应于该语音输入信息的识别结果是正确的，则执行步骤S210或S214。为了更好地阐述本实施例，下面以识别结果为“我非常想去北京”为例进行描述。
首先对步骤S210-S213进行阐述。
当对应于该语音输入信息的识别结果是正确时，可以对该识别结果进行分析(譬如，拆分)，然后确定与该识别结果相关联的关键字，譬如，根据特定的字段或是语义算法从识别结果中抽取关键字。通过对识别结果“我非常想去北京”进行抽取，可以抽取到关键字“我”、“想去”、“北京”。待确定好上述关键字后，将在数据库(譬如，语料库)中进行查找匹配。
步骤S211：判断关键字能否与数据库中的字条进行匹配？
在该步骤中，将对关键字与数据库中的数据字段的匹配情况进行判断。
当步骤S211的判断结果为否时，即此时数据库中没有与当前的关键字相匹配 的数据字段，如此，则执行步骤S212，向用户发出第四反馈信息。可以理解的，这里的第四反馈信息可以是任何形式的能够让用户感知的信息。
当步骤S211的判断结果为是时，即此时数据库中存在与当前的关键字相匹配的数据字段，则执行步骤S213，即根据匹配的结果产生相应的操作。换而言之，将基于关键字“我”、“想去”、“北京”来触发相应的操作。当在数据可视化场景中，可以向当前的用户提供去北京的路线，或是去北京的航班、车次等等可供选择的交通工具的可用情况。
另外，当系统中直接配置好固定的可接收字段，用户在进行数据可视化现场展示讲解时，可直接说出预先配置好的设备可接收字段。展示的过程中，当终端设备在接收到指令后，可直接与后台数据进行比对，快速在展示端上展示所需数据。也就是说，若终端设备或处理设备处已经存有与语音“我非常想去北京”相对应的数据字段，则此时无需再对该语音进行关键字抽取，可以直接执行与该数据字段相对应的操作(步骤S214)。
通过上述方法，在数据可视化场景中实现了基于语音识别及自然语言处理，能够提升用户与数据展示间的互动性，改善当前数据可视化交互方式的单一性。用户通过自然语言传输即可完成操作，降低数据可视化交互操作的复杂性，提升展示效率，尤其适合在数据大屏展示场景中使用。
虽然上述实施例采用了语音输入信息作为实施例，本领域技术人员能够理解的是，肢体动作、触摸等具有指示性的信息同样适用于上述方法。譬如，当终端设备中的视频组件捕捉到用户双手合十的动作时，将通过相应的动作识别模型对该动作进行识别。譬如，可以通过训练，将该双手合十的动作与“关机”功能相关联，由此，当动作识别模型对该动作进行正确识别后，将触发“关机”功能。
图1、2中的信息处理方法的流程还代表机器可读指令，该机器可读指令包括由处理器执行的程序。该程序可被实体化在被存储于有形计算机可读介质的软件中，该有形计算机可读介质如CD-ROM、软盘、硬盘、数字通用光盘(DVD)、蓝光光盘或其它形式的存储器。替代的，图1、2中的示例方法中的一些步骤或所有步骤可利用专用集成电路(ASIC)、可编程逻辑器件(PLD)、现场可编程逻辑 器件(EPLD)、离散逻辑、硬件、固件等的任意组合被实现。另外，虽然图1、2所示的流程图描述了该数据处理方法，但可对该处理方法中的步骤进行修改、删除或合并。
如上所述，可利用编码指令(如计算机可读指令)来实现图1、2的示例过程，该编程指令存储于有形计算机可读介质上，如硬盘、闪存、只读存储器(ROM)、光盘(CD)、数字通用光盘(DVD)、高速缓存器、随机访问存储器(RAM)和/或任何其他存储介质，在该存储介质上信息可以存储任意时间(例如，长时间，永久地，短暂的情况，临时缓冲，和/或信息的缓存)。如在此所用的，该术语有形计算机可读介质被明确定义为包括任意类型的计算机可读存储的信息。附加地或替代地，可利用编码指令(如计算机可读指令)实现图1、2的示例过程，该编码指令存储于非暂时性计算机可读介质，如硬盘，闪存，只读存储器，光盘，数字通用光盘，高速缓存器，随机访问存储器和/或任何其他存储介质，在该存储介质信息可以存储任意时间(例如，长时间，永久地，短暂的情况，临时缓冲，和/或信息的缓存)。可以理解的，该计算机可读指令还可以存储在网络服务器中、云端平台上，以便于用户使用。
另外，尽管操作以特定顺序被描绘，但这并不应该理解为要求此类操作以示出的特定顺序或以相继顺序完成，或者执行所有图示的操作以获取期望结果。在某些情况下，多任务或并行处理会是有益的。同样地，尽管上述讨论包含了某些特定的实施细节，但这并不应解释为限制任何发明或权利要求的范围，而应解释为对可以针对特定发明的特定实施例的描述。本说明书中在分开的实施例的上下文中描述的某些特征也可以整合实施在单个实施例中。反之，在单个实施例的上下文中描述的各种特征也可以分离地在多个实施例或在任意合适的子组合中实施。
因此，虽然参照特定的示例来描述了本发明，其中这些特定的示例仅仅旨在是示例性的，而不是对本发明进行限制，但对于本领域普通技术人员来说显而易见的是，在不脱离本发明的精神和保护范围的基础上，可以对所公开的实施例进行改变、增加或者删除。

Claims (16)

  1. 一种用于数据可视化的信息的处理方法，其特征在于，包括：
    对所接收的输入信息进行可识别性分析；以及，
    确定所述输入信息是否被正确识别，当所述输入信息被正确识别时，基于所述输入信息的识别结果来确定与所述识别结果相对应的交互指令，进而执行所述交互指令。
  2. 根据权利要求1所述的方法，其特征在于，所述确定所述输入信息是否被正确识别包括：
    将能够被识别的所述输入信息转换为具有指定呈现形式的媒介信息，并基于所述媒介信息的确认信息来确定所述输入信息是否被正确识别，所述确认信息用于指示所述媒介信息是否正确地呈现了所述输入信息。
  3. 根据权利要求1所述的方法，其特征在于，所述基于所述输入信息的识别结果来确定与所述识别结果相对应的交互指令包括：将所述识别结果在数据库中进行查找匹配；当所述数据库中存有与所述识别结果相对应的数据字段时，基于所述识别结果直接确定与所述识别结果相对应的交互指令。
  4. 根据权利要求3所述的方法，其特征在于，所述基于所述输入信息的识别结果来确定与所述识别结果相对应的交互指令包括：将所述识别结果在数据库中进行查找匹配，当所述数据库中不存在与所述识别结果相对应的数据字段时，基于所述识别结果来确定关键字集；以及，基于所述关键字集来确定与所述识别结果相对应的交互指令。
  5. 根据权利要求1所述的方法，其特征在于，还包括：
    当对所述输入信息进行接收时，判断所述输入信息是否被成功接收，其中，当所述输入信息未被成功接收，则生成用于指示接收失败的第一反馈信息。
  6. 根据权利要求1所述的方法，其特征在于，所述对所接收的输入信息进行可识别性分析包括：
    基于用于识别所述输入信息的识别模型来对所述输入信息进行分析，进而确 定所述输入信息的可识别性；其中，当所述输入信息无法被识别时，生成用于指示所述输入信息无法被识别的第二反馈信息。
  7. 根据权利要求2所述的方法，其特征在于，当所述输入信息未被正确识别时，产生用于指示所述输入信息识别错误的第三反馈信息。
  8. 根据权利要求4所述的方法，其特征在于，基于所述输入信息的识别结果来确定关键字集包括：
    将所述输入信息识别为语义文本，从所述语义文本中抽取所述关键字集，其中，所述关键字集包括至少一个字段。
  9. 根据权利要求4所述的方法，其特征在于，所述基于所述关键字集来确定与所述识别结果相对应的交互指令包括：
    基于所述关键字集来与数据库中的数据字段进行比对；
    当所述关键字集中的字段与所述数据库中的数据字段相匹配时，基于匹配结果来确定所述交互指令。
  10. 根据权利要求9所述的方法，其特征在于，所述基于所述关键字集来确定与所述识别结果相对应的交互指令还包括：
    当所述关键字集中的字段与所述数据库中的数据字段不匹配时，生成第四反馈信息，其中，所述第四反馈信息用于指示所述关键字集中的字段与所述数据库中的数据字段无法匹配。
  11. 根据权利要求1至10任一项所述的方法，其特征在于，所述输入信息包括以下项中的至少一项：语音、触摸或肢体动作。
  12. 根据权利要求11所述的方法，其特征在于，所述方法还包括当对所述输入信息进行接收时，判断所述输入信息是否被成功接收，其中所述输入信息包括语音；其中，判断所述输入信息是否被成功接收包括：基于第一阈值来判断所述语音是否被成功接收。
  13. 根据权利要求12所述的方法，其特征在于，所述第一阈值包括以下项中的一种或者多种组合：语音长度阈值、语音强度阈值、语音频域阈值。
  14. 根据权利要求2所述的方法，其特征在于，所述媒介信息包括以下项中的至少一项：视频、音频、图片或文字。
  15. 一种用于数据可视化的信息处理装置，其特征在于，包括：
    处理器；
    存储器，其用于存储指令，当所述指令在执行时，所述处理器执行如权利要求1至14任一项所述的用于可视化的信息的处理方法的步骤。
  16. 一种计算机可读存储介质，具有存储在其上的计算机可读程序指令，其特征在于，所述计算机可读程序指令执行时，实现如权利要求1至14中任一项所述的用于数据可视化的信息的处理方法的步骤。
PCT/CN2018/116415 2017-11-21 2018-11-20 用于数据可视化的信息的处理方法以及装置 WO2019101067A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2019542091A JP6887508B2 (ja) 2017-11-21 2018-11-20 データ可視化のための情報処理方法及び装置
KR1020197023144A KR20190107063A (ko) 2017-11-21 2018-11-20 데이터 시각화를 위한 정보 처리 방법 및 장치
US16/354,678 US20190213998A1 (en) 2017-11-21 2019-03-15 Method and device for processing data visualization information

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711166559.1A CN108108391A (zh) 2017-11-21 2017-11-21 用于数据可视化的信息的处理方法以及装置
CN201711166559.1 2017-11-21

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/354,678 Continuation US20190213998A1 (en) 2017-11-21 2019-03-15 Method and device for processing data visualization information

Publications (1)

Publication Number Publication Date
WO2019101067A1 true WO2019101067A1 (zh) 2019-05-31

Family

ID=62207647

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/116415 WO2019101067A1 (zh) 2017-11-21 2018-11-20 用于数据可视化的信息的处理方法以及装置

Country Status (5)

Country Link
US (1) US20190213998A1 (zh)
JP (1) JP6887508B2 (zh)
KR (1) KR20190107063A (zh)
CN (1) CN108108391A (zh)
WO (1) WO2019101067A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108108391A (zh) * 2017-11-21 2018-06-01 众安信息技术服务有限公司 用于数据可视化的信息的处理方法以及装置
CN109241464A (zh) * 2018-07-19 2019-01-18 上海小蚁科技有限公司 用于数据大屏的数据展示方法及装置、存储介质、终端
CN111510671A (zh) * 2020-03-13 2020-08-07 海信集团有限公司 一种监控视频调取显示的方法及智能终端
CN111610949A (zh) * 2020-05-28 2020-09-01 广州市玄武无线科技股份有限公司 数据大屏的展示方法及装置、电子设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140207452A1 (en) * 2013-01-24 2014-07-24 Microsoft Corporation Visual feedback for speech recognition system
CN105005578A (zh) * 2015-05-21 2015-10-28 中国电子科技集团公司第十研究所 多媒体目标信息可视化分析系统
CN106980689A (zh) * 2017-03-31 2017-07-25 邢加和 一种通过语音交互实现数据可视化的方法
CN107300970A (zh) * 2017-06-05 2017-10-27 百度在线网络技术（北京）有限公司 虚拟现实交互方法和装置
CN108108391A (zh) * 2017-11-21 2018-06-01 众安信息技术服务有限公司 用于数据可视化的信息的处理方法以及装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000019307A1 (fr) * 1998-09-25 2000-04-06 Hitachi, Ltd. Procede et dispositif d'interaction de traitement
JP3705735B2 (ja) * 2000-08-29 2005-10-12 シャープ株式会社 オンデマンド・インタフェース装置とそのウィンドウ表示装置
US7437291B1 (en) * 2007-12-13 2008-10-14 International Business Machines Corporation Using partial information to improve dialog in automatic speech recognition systems
CN103065640B (zh) * 2012-12-27 2017-03-01 上海华勤通讯技术有限公司 语音信息可视化的实现方法
JPWO2017130486A1 (ja) * 2016-01-28 2018-11-22 ソニー株式会社 情報処理装置、情報処理方法およびプログラム
US10373612B2 (en) * 2016-03-21 2019-08-06 Amazon Technologies, Inc. Anchored speech detection and speech recognition
WO2017168936A1 (ja) * 2016-03-31 2017-10-05 ソニー株式会社 情報処理装置、情報処理方法、及びプログラム
CN107199971B (zh) * 2017-05-03 2020-03-13 深圳车盒子科技有限公司 车载语音交互方法、终端及计算机可读存储介质
CN107193948B (zh) * 2017-05-22 2018-04-20 邢加和 人机对话数据分析方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140207452A1 (en) * 2013-01-24 2014-07-24 Microsoft Corporation Visual feedback for speech recognition system
CN105005578A (zh) * 2015-05-21 2015-10-28 中国电子科技集团公司第十研究所 多媒体目标信息可视化分析系统
CN106980689A (zh) * 2017-03-31 2017-07-25 邢加和 一种通过语音交互实现数据可视化的方法
CN107300970A (zh) * 2017-06-05 2017-10-27 百度在线网络技术（北京）有限公司 虚拟现实交互方法和装置
CN108108391A (zh) * 2017-11-21 2018-06-01 众安信息技术服务有限公司 用于数据可视化的信息的处理方法以及装置

Also Published As

Publication number Publication date
JP6887508B2 (ja) 2021-06-16
KR20190107063A (ko) 2019-09-18
CN108108391A (zh) 2018-06-01
US20190213998A1 (en) 2019-07-11
JP2020507165A (ja) 2020-03-05

Similar Documents

Publication Publication Date Title
WO2019101067A1 (zh) 用于数据可视化的信息的处理方法以及装置
US11797772B2 (en) Word lattice augmentation for automatic speech recognition
CN112115706B (zh) 文本处理方法、装置、电子设备及介质
US9805718B2 (en) Clarifying natural language input using targeted questions
CN109635270B (zh) 双向概率性的自然语言重写和选择
CN105931644B (zh) 一种语音识别方法及移动终端
CN107909998B (zh) 语音指令处理方法、装置、计算机设备和存储介质
JP2016529603A (ja) オンライン音声翻訳方法及び装置
WO2017166631A1 (zh) 语音信号处理方法、装置和电子设备
US11120802B2 (en) Diarization driven by the ASR based segmentation
US20220115000A1 (en) On-device personalization of speech synthesis for training of speech recognition model(s)
CN113412514A (zh) 用于设备上语音识别模型的训练的文本片段的设备上语音合成
EP3724875B1 (en) Text independent speaker recognition
WO2018094852A1 (zh) 一种虚拟现实设备及虚拟现实设备的输入控制方法
KR20150077580A (ko) 음성 인식 기반 서비스 제공 방법 및 그 장치
CN113225612A (zh) 字幕生成方法、装置、计算机可读存储介质及电子设备
KR102345625B1 (ko) 자막 생성 방법 및 이를 수행하는 장치
CN108268443B (zh) 确定话题点转移以及获取回复文本的方法、装置
KR20230005966A (ko) 거의 일치하는 핫워드 또는 구문 검출
US10468031B2 (en) Diarization driven by meta-information identified in discussion content
CN106980640B (zh) 针对照片的交互方法、设备和计算机可读存储介质
JP2013050742A (ja) 音声認識装置および音声認識方法
CN113658593B (zh) 基于语音识别的唤醒实现方法及装置
JP2023007014A (ja) 応答システム、応答方法、および応答プログラム
KR20130137367A (ko) 이미지 기반 도서 관련 서비스 제공 시스템 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18881284

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019542091

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20197023144

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18881284

Country of ref document: EP

Kind code of ref document: A1