WO2020119437A1 - 语音控制方法、云端服务器及终端设备 - Google Patents

语音控制方法、云端服务器及终端设备 Download PDF

Info

Publication number
WO2020119437A1
WO2020119437A1 PCT/CN2019/120823 CN2019120823W WO2020119437A1 WO 2020119437 A1 WO2020119437 A1 WO 2020119437A1 CN 2019120823 W CN2019120823 W CN 2019120823W WO 2020119437 A1 WO2020119437 A1 WO 2020119437A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice
control instruction
cloud server
voice information
cloud
Prior art date
Application number
PCT/CN2019/120823
Other languages
English (en)
French (fr)
Inventor
黄振兴
许升
尹俊明
舒海
Original Assignee
青岛海尔洗衣机有限公司
海尔智家股份有限公司
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 青岛海尔洗衣机有限公司, 海尔智家股份有限公司 filed Critical 青岛海尔洗衣机有限公司
Priority to EP19895476.0A priority Critical patent/EP3896688A4/en
Priority to JP2021533280A priority patent/JP7159476B2/ja
Priority to US17/413,140 priority patent/US11705129B2/en
Publication of WO2020119437A1 publication Critical patent/WO2020119437A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G06F3/167—Audio in a user interface, e.g. using voice commands for navigating, audio feedback
    • 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
    • G10L15/28—Constructional details of speech recognition systems
    • G10L15/30—Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
    • 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
    • G10L15/28—Constructional details of speech recognition systems
    • G10L15/34—Adaptation of a single recogniser for parallel processing, e.g. by use of multiple processors or cloud computing
    • 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

Definitions

  • the invention relates to the technical field of voice control, and in particular to a voice control method, a cloud server and a terminal device.
  • the washing machine With the improvement of people's living standards, the washing machine has become one of the necessary household appliances in people's homes. In order to meet the differentiated laundry needs of users, the existing washing machines have more and more functions. The laundry mode and The adjustment options for laundry parameters are also getting more and more complicated, which undoubtedly increases the user's learning and use costs.
  • the existing voice recognition module is mainly divided into an offline voice recognition module and an online voice recognition module.
  • the offline voice recognition module responds quickly, However, it can only recognize specific command keywords, and the number of keywords is often limited by the storage space. Users who want to start a specific laundry program must accurately remember the corresponding keywords, and the experience is very bad.
  • the online voice recognition module can perform semantic recognition.
  • the laundry program required by the user can be accurately executed, but the online voice recognition module requires Upload the command to the cloud for analysis, the response is slow, causing the user to wait too long, so the experience is not very good.
  • the present invention provides a voice control method, the voice control method includes :
  • Terminal equipment receives voice information
  • the terminal device queries the local voice library for control instructions corresponding to the voice information
  • the terminal device When the control instruction corresponding to the voice information is not found in the local voice library, the terminal device uploads the voice information to the cloud server;
  • the cloud server queries the cloud voice library for control instructions corresponding to the voice information
  • the cloud server When a control instruction corresponding to the voice information is found in the cloud voice library, the cloud server sends the control instruction to the terminal device;
  • the terminal device receives the control instruction and performs a corresponding operation based on the control instruction.
  • the voice control method further includes:
  • the cloud server determines the number of times the mapping relationship between the voice information and the control instruction is recorded in the cloud voice library;
  • the cloud server compares the number of recording times with the preset number of times
  • the cloud server sends the mapping relationship to the terminal device;
  • the terminal device receives the mapping relationship and stores the mapping relationship to the local voice library.
  • the voice control method further includes:
  • the cloud server parses the control instruction corresponding to the voice information
  • the cloud server When the control instruction corresponding to the voice information is parsed, the cloud server sends the control instruction to the terminal device, and records the mapping relationship between the voice information and the control instruction to the cloud voice In the library
  • the terminal device receives the control instruction and performs a corresponding operation based on the control instruction.
  • the present invention also provides a voice control method, including:
  • the cloud server receives the voice information uploaded by the terminal device
  • the cloud server queries the cloud voice library for control instructions corresponding to the voice information
  • the cloud server When a control instruction corresponding to the voice information is queried in the cloud voice library, the cloud server sends the control instruction to the terminal device, so that the terminal device performs a corresponding operation based on the control instruction;
  • the terminal device uploads the voice information when the control instruction corresponding to the voice information cannot be found from the local voice library.
  • the voice control method further includes:
  • the cloud server determines the number of times the mapping relationship between the voice information and the control instruction is recorded in the cloud voice library;
  • the cloud server compares the number of recording times with the preset number of times
  • the cloud server sends the mapping relationship to the terminal device, so that the terminal device stores the mapping relationship to the local voice library.
  • the voice control method further includes:
  • the cloud server parses the control instruction corresponding to the voice information
  • the cloud server sends the control instruction to the terminal device, so that the terminal device performs a corresponding operation based on the control instruction;
  • the cloud server records the mapping relationship between the voice information and the control instruction in the cloud voice library.
  • the present invention also provides a voice control method, including:
  • Terminal equipment receives voice information
  • the terminal device queries the local voice library for control instructions corresponding to the voice information
  • the terminal device When the control instruction corresponding to the voice information is not found in the local voice library, the terminal device uploads the voice information to the cloud server, so that the cloud server queries the voice information from the cloud voice library
  • the corresponding control instruction or the control instruction corresponding to the voice information is parsed when not found in the cloud voice library;
  • the terminal device receives the searched or parsed control instruction corresponding to the voice information from the cloud server, and performs a corresponding operation based on the control instruction.
  • control method after the step of “uploading the voice information to the cloud server by the terminal device”, the control method further includes:
  • the terminal device receives the mapping relationship between the voice information and the control instruction from the cloud server, and stores the mapping relationship to the local voice library;
  • the cloud server determines that the mapping relationship between the voice information and the control instruction is in the cloud voice
  • the number of records in the library the mapping relationship is sent by the cloud server when the number of records is greater than a preset number of times.
  • the present invention also provides a cloud server including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be loaded by the processor for execution
  • a cloud server including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be loaded by the processor for execution
  • the present invention also provides a terminal device including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be loaded by the processor for execution
  • a terminal device including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be loaded by the processor for execution
  • the voice control method according to any one of the above preferred technical solutions of the third aspect.
  • the voice control method includes: the terminal device receives voice information; the terminal device queries the local voice library for control instructions corresponding to the voice information; when in the local voice library When the control instruction corresponding to the voice information is not found, the terminal device uploads the voice information to the cloud server; the cloud server queries the control instruction corresponding to the voice information from the cloud voice library; when the control instruction corresponding to the voice information is found in the cloud voice library At this time, the cloud server sends a control instruction to the terminal device; the terminal device receives the control instruction and performs the corresponding operation based on the control instruction.
  • the voice control method of the present invention can increase the response speed of a terminal device with a voice recognition function, and greatly improve the user experience of the user. Specifically, after receiving the voice information, the terminal device first performs a local query. If a control instruction corresponding to the voice information is found in the local voice library, the corresponding operation is performed according to the control instruction. Because the response speed of the local voice library is fast, the operation of the terminal device is performed quickly, and the user experience is good. If the control instruction corresponding to the voice information is not found in the local voice library, the cloud server queries the cloud voice library by uploading the voice information to the cloud server.
  • the cloud server sends a control instruction to the terminal device, so that the terminal device performs a corresponding operation based on the control instruction. Because the cloud server has super storage and computing capabilities, its cloud voice library stores a large amount of voice information and corresponding control instructions, so when no control instructions are queried in the local voice library, the cloud server performs a larger range The query can avoid the problems of small recognition range and low recognition accuracy only through local recognition, and improve the user experience.
  • the number of recording times of the mapping relationship between the voice information and the corresponding control instruction is further determined by the cloud server, and when the number of recording times is greater than the preset number of times , The corresponding mapping relationship is sent to the terminal device, so that the terminal device stores the mapping relationship in the local voice library.
  • the voice control method of the present invention also enables the terminal device to automatically update the local voice library based on the user's usage habits, so that the terminal device The degree of intelligence of the voice recognition function has been greatly improved, which in turn greatly improves the response speed of the terminal device and the user experience.
  • the voice information is parsed through the cloud server, and when the control instruction corresponding to the voice information is parsed, the control instruction is sent to the terminal Device, so that the terminal device performs the corresponding operation based on the control instruction. Since the cloud control method of the present invention can perform speech semantic recognition and has a wide recognition range and high accuracy, it can not find the corresponding in the cloud voice library Control instructions to further improve recognition efficiency and recognition accuracy, and enhance user experience.
  • FIG. 1 is a flowchart of an implementation manner of the voice control method of the present invention
  • FIG. 3 is a flowchart of still another embodiment of the voice control method of the present invention.
  • FIG. 4 is a logic diagram of an embodiment of the voice control method of the present invention.
  • FIG. 5 is a logic diagram of another embodiment of the voice control method of the present invention.
  • FIG. 1 is a flowchart of an implementation manner of the voice control method of the present invention.
  • the execution subject of the method is a washing machine and a cloud server (hereinafter referred to as cloud), wherein the washing machine is provided with a control module and a voice recognition module, the voice recognition module can recognize the control instructions corresponding to the voice information, and the voice information can be converted into control instructions
  • the control instruction is sent to the control module, so that the control module controls the washing machine to perform the corresponding operation based on the control instruction.
  • the cloud server can also recognize the control instructions corresponding to the voice information, and the cloud server can also directly communicate with the voice recognition module or indirectly communicate with the voice recognition module through communication with the washing machine to receive the voice information sent by the voice recognition module or the washing machine, And the instruction information or the mapping relationship between the voice information and the control instruction is sent to the voice recognition module or the washing machine, so that the washing machine can perform the corresponding operation.
  • the form of the cloud server is not limited in this embodiment, and any remote server that is not local can be used as the cloud server in the present invention.
  • the voice control method of the present invention mainly includes the following steps:
  • the washing machine receives voice information.
  • the voice information may be obtained through a collection element such as a microphone provided on the washing machine, or may be obtained through conversion after receiving an audio file, which is not limited in this embodiment.
  • the microphone is always on to obtain voice information, or when the user presses the voice input button set on the washing machine and is on to obtain voice information; again, for example, the user , Tablet, etc.) input voice information, so that the washing machine converts into voice information after receiving the audio file sent by the mobile device.
  • the washing machine queries the local voice library for control instructions corresponding to the voice information.
  • a local voice library is stored in the washing machine, and the mapping relationship between voice information and control instructions is stored in the local voice library.
  • the washing machine controls the voice recognition module to query the mapping relationship stored in the local voice library to identify whether there is The control instruction corresponding to the voice information.
  • the washing machine uploads the voice information to the cloud server.
  • the control instruction is sent to the control module, so that the control module controls the washing machine to perform corresponding operations, such as performing a washing program, a dehydration program, etc., based on the control instruction.
  • the voice recognition module or the washing machine uploads the voice information to the cloud server, so as to process the voice information using the powerful processing capability of the cloud server.
  • the voice recognition module or the washing machine uploads the voice information to the cloud server in a wired or wireless manner, which is not limited in this embodiment.
  • the cloud server queries the cloud voice library for control instructions corresponding to the voice information.
  • a cloud voice library is stored in a cloud server, and a cloud voice library stores a mapping relationship between a large amount of voice information and control instructions.
  • the mapping relationship may be stored in advance, or may be used by all users connected to the server Generated in the process; after receiving the voice information, the cloud server recognizes whether there is a control instruction corresponding to the voice information from the mapping relationship stored in the cloud voice library.
  • the cloud server sends the control instruction to the washing machine. For example, when a control instruction corresponding to the voice information is found in the cloud voice library, the cloud server sends the control instruction to the washing machine, so that the washing machine can perform the corresponding operation based on the control instruction.
  • the washing machine receives the control instruction and executes the corresponding operation based on the control instruction.
  • the control module receives the control instruction and controls the washing machine to perform corresponding operations based on the control instruction, such as executing a washing program, a dehydration program, and the like.
  • the voice control method of the present invention can increase the response speed of the washing machine and greatly improve the user experience of the user. Specifically, after receiving the voice information, the washing machine first controls the voice recognition module to perform a local query. If a control instruction corresponding to the voice information is found in the local voice library, the control module controls the washing machine to perform the corresponding operation according to the control instruction. Since the response speed of the local voice library is fast, the operation of the washing machine in this case is performed quickly and the user experience is good. If the control instruction corresponding to the voice information is not found in the local voice library, the cloud server queries the cloud voice library by uploading the voice information to the cloud server.
  • the cloud server sends the control instruction to the control module of the washing machine, so that the control module controls the washing machine to perform corresponding operations based on the control instruction. Because the cloud server has super storage and computing capabilities, and its cloud voice library stores a large number of mapping relationships, when no control command is queried in the local voice library, a larger range of queries can be performed through the cloud server to avoid The problem of small recognition range and low recognition accuracy when performing voice recognition only locally improves user experience.
  • the voice control method further includes the following steps:
  • the cloud server determines how many times the mapping relationship between the voice information and the control instruction is recorded in the cloud voice library. For example, when querying the control instruction corresponding to the voice information in the cloud voice library, it proves that the cloud voice library already stores the mapping relationship between the voice information and the control instruction. The cloud server can further determine the cloud after sending the control instruction to the washing machine The number of records of this mapping relationship in the voice library.
  • the cloud server compares the number of recording times with the preset number of times. For example, after determining the number of recording times from the cloud voice library, the cloud server updates the current mapping relationship to the number of recording times (that is, the number of recording times+1) and compares it with the preset number of times.
  • the preset number of times is preferably 10-30 times, and more preferably 20 times.
  • the setting of the preset number of times can be adjusted by those skilled in the art based on different application scenarios.
  • the current record may be updated to the number of recording times, or may be directly compared without updating.
  • the cloud server sends the mapping relationship to the washing machine, the washing machine receives the mapping relationship, and stores the mapping relationship to the local voice library for the next time.
  • the washing machine can directly query the corresponding control instruction from the local voice library.
  • the number of recordings is less than the preset number, it is proved that the mapping relationship is used less frequently. In this case, there is no need to send the mapping relationship to the washing machine, so as to avoid too many mapping relationships and occupy storage space of the local voice library.
  • the cloud server parses the control command corresponding to the voice information; for example, the cloud server does not find the same mapping relationship in the cloud voice library. It is not recorded in the cloud voice library, that is, the mapping relationship has not been used by the user or other users. At this time, the cloud server performs voice and/or semantic analysis and recognition of the voice information.
  • the cloud server fails to parse out the control command corresponding to the voice message, it can send a parse failure command to the washing machine, so that the control module can control the washing machine to issue a prompt message after receiving the parse failure command, such as a prompt sound to prompt the user to re-enter, or control The indicator flashes, etc.
  • the cloud server When the cloud server successfully parses out the control instruction corresponding to the voice information, it can send the control instruction to the washing machine, so that the washing machine can perform the corresponding operation based on the control instruction. In addition, while the cloud server successfully parses out the control instructions corresponding to the voice information, the cloud server can also record the mapping relationship to the cloud voice library, so that the cloud server can easily recall from the cloud voice library when the same mapping relationship occurs next time. Take the record times of the mapping relationship.
  • the washing machine receives control instructions and performs corresponding operations based on the control instructions. For example, after the cloud server successfully parses out the control instruction and sends it to the washing machine, the control module receives the control instruction and controls the washing machine to perform corresponding operations based on the control instruction, such as executing a washing program, a dehydration program, and the like.
  • the control method of the present invention also enables the washing machine to automatically update the local voice library based on the user's usage habits, thereby greatly improving the intelligence of the washing machine's voice recognition function, greatly increasing the response speed of the washing machine and the user's Use experience.
  • the local voice library due to the limited storage capacity of the local voice library, it is not possible to store a large number of mapping relationships like a cloud server, and after applying the voice control method of the present invention, the local voice library always stores some of the most frequently used mapping relationships of users , So that the washing machine can be highly adapted to the user's usage habits, and truly achieve intelligent control of the washing machine.
  • the control instruction corresponding to the voice information is not found in the cloud voice library, the voice information is parsed through the cloud server, and when the control instruction corresponding to the voice information is parsed, the control instruction is sent to the washing machine Therefore, the washing machine performs corresponding operations based on the control instruction. Since the cloud control method of the present invention can perform speech semantic recognition, and the recognition range is wide and the accuracy is high, the corresponding control can not be queried in the cloud voice library Further increase the recognition efficiency and recognition accuracy when instructing, and improve the user experience.
  • the cloud server may determine the number of times the mapping relationship is recorded in the cloud voice library at the same time or before sending the control instruction to the washing machine. This change in the sequence of steps does not deviate from the present invention. The principle should therefore fall within the protection scope of the present invention.
  • FIG. 2 is a flowchart of another implementation manner of the voice control method of the present invention.
  • the execution subject of this method is a cloud server (hereinafter referred to as cloud).
  • cloud The setting modes of the washing machine and the cloud server involved in the following embodiments are the same as those in the first embodiment, and are not repeated here.
  • the voice control method of the present invention mainly includes the following steps:
  • the cloud server receives the voice information uploaded by the washing machine
  • the cloud server queries the cloud voice library for control instructions corresponding to the voice information
  • the cloud server sends the control instruction to the washing machine, so that the washing machine performs the corresponding operation based on the control instruction; wherein, the washing machine cannot find the voice information from the local voice library Voice messages are uploaded during the corresponding control instructions.
  • the washing machine uploads voice information to the cloud server when the control instruction corresponding to the voice information cannot be queried from the local voice library, so that the cloud server queries the control instruction corresponding to the voice information in the cloud voice library and controls the control when the query is found
  • the instruction is sent to the washing machine, so that the washing machine can perform the corresponding operation based on the control instruction.
  • the voice control method of the present invention can improve the recognition accuracy of the washing machine, and avoid the accuracy of the washing machine only through the limited mapping relationship in the local voice library. Low problems greatly improve the user experience.
  • the voice control method of the present invention further includes after step S202:
  • the cloud server determines the number of records of the mapping relationship between the voice information and the control instruction in the cloud voice library; the cloud server compares the number of records with the preset number of times; When the number of recording times is greater than the preset number of times, the cloud server sends the mapping relationship to the washing machine, so that the washing machine stores the mapping relationship to the local voice library.
  • the cloud server parses the control command corresponding to the voice information; when the control command corresponding to the voice information is parsed, the cloud server sends the control command to the washing machine, so that the washing machine is based on the control The instruction performs the corresponding operation; and the cloud server records the mapping relationship between the voice information and the control instruction in the cloud voice library.
  • the cloud server When querying the control instruction corresponding to the voice information in the cloud voice library, the cloud server determines the number of recording times of the mapping relationship between the voice information and the corresponding control instruction, and when the recording number is greater than the preset number, the corresponding mapping The relationship is sent to the washing machine so that the washing machine stores the mapping relationship in the local voice library.
  • the voice control method of the present invention enables the washing machine to automatically update the local voice library in the washing machine based on the user's usage habits, making the washing machine's voice recognition function intelligent The degree has been greatly improved, thereby further improving the response speed of the washing machine and the user's experience.
  • the cloud server parses the voice information to obtain the corresponding control command, and sends the control command to the washing machine, so that the washing machine can be based on the control command
  • the voice control method of the present invention can also improve the recognition accuracy of the washing machine, avoid the problem of low accuracy caused when the mapping relationship is not stored in the cloud voice library, and greatly improve the user experience.
  • FIG. 3 is a flowchart of still another implementation manner of the voice control method of the present invention.
  • the execution subject of this method is a washing machine.
  • the setting modes of the washing machine and the cloud server (or cloud for short) involved in the following embodiments are the same as those in the first embodiment, and will not be repeated here.
  • the voice control method of the present invention mainly includes the following steps:
  • the washing machine receives voice information
  • the washing machine queries the local voice library for control instructions corresponding to the voice information;
  • the washing machine uploads the voice information to the cloud server, so that the cloud server queries the control instruction corresponding to the voice information from the cloud voice library or does not find it in the cloud voice library. Analyze the control instructions corresponding to the voice information when inquired;
  • the washing machine receives a control instruction corresponding to the queried or parsed voice information from the cloud server, and performs a corresponding operation based on the control instruction.
  • the voice control method of the present invention enables the operation of the washing machine. Fast and good user experience.
  • the control instruction corresponding to the voice information cannot be found from the local voice library, upload the voice information to the cloud server, so that the cloud server queries the control instruction corresponding to the voice information from the cloud voice library or does not find it in the cloud voice library
  • the control command corresponding to the voice information is parsed, and then the washing machine receives the control command corresponding to the queried or parsed voice information, and performs corresponding operations based on the control command.
  • the voice control method of the present invention can also recognize voice information locally Through the analysis of the cloud server, the recognition accuracy of the washing machine is improved to avoid the problem of low accuracy caused by the limited mapping relationship in the local voice library, and the user experience is greatly improved.
  • the voice control method of the present invention further includes after step S303:
  • the washing machine receives the mapping relationship between the voice information and the control instructions from the cloud server and stores the mapping relationship to the local voice library; where, when the cloud server queries the cloud voice library for the control instruction corresponding to the voice information, the cloud server determines the voice The number of times the mapping relationship between information and control instructions is recorded in the cloud voice library; the mapping relationship is sent by the cloud server when the number of recordings is greater than the preset number.
  • the voice control method of the present invention enables the washing machine to automatically update the local voice library in the washing machine based on the user's usage habits.
  • the intelligence of the voice recognition function has been greatly improved, thereby further improving the response speed of the washing machine and the user's experience.
  • the present invention also provides a cloud server including a memory, a processor, and a computer program, wherein the memory is connected to the processor, the computer program is stored in the memory, and is configured to be loaded by the processor to execute the embodiment The voice control method described in the second.
  • the present invention also provides a washing machine including a memory, a processor, and a computer program, wherein the memory is connected to the processor, the computer program is stored in the memory, and is configured to be loaded by the processor to perform the third embodiment The voice control method.
  • FIG. 4 is a logic diagram of an embodiment of the voice control method of the present invention
  • FIG. 5 is a logic diagram of another embodiment of the voice control method of the present invention.
  • the washing machine receives the user's voice information, and the voice recognition module queries the corresponding control instruction in the ontology voice library based on the voice information; when the voice information is found in the local voice library When the corresponding control instruction is received, the voice recognition module sends the control instruction to the control module, and the control module controls the washing machine to perform the corresponding operation based on the control instruction; when the control instruction corresponding to the voice information is not queried, the washing machine uses the voice information via wireless Upload to the cloud, the cloud server queries the cloud voice library for the control instruction corresponding to the voice information, and when the cloud voice library finds the control instruction, the control instruction is returned to the control module of the washing machine, so that the washing machine control module Based on the control instruction, the washing machine is controlled to perform the corresponding operation; when the control instruction corresponding to the voice information is not found in the cloud voice library, the cloud server further parses the voice information.
  • control instruction corresponding to the voice information is successfully parsed, Then return the control instruction to the control module of the washing machine, so that the control module of the washing machine controls the washing machine to perform corresponding operations based on the control instruction; if the cloud server does not successfully parse out the control instruction corresponding to the voice message, the failed instruction is returned to the control of the washing machine The module, so that the control module controls the washing machine to emit a prompt sound, prompting the user to re-enter voice information.
  • the cloud after receiving the voice information uploaded by the washing machine, the cloud queries the control instruction corresponding to the voice information from the cloud voice library; if no control corresponding to the voice information is found When instructing, the voice information is parsed.
  • the command is ended; if the corresponding control command is parsed, the mapping relationship between the voice information and the control command is recorded to the cloud voice In the library; if the control instruction corresponding to the voice information is queried, the query number of the mapping relationship between the voice information and the control instruction in the cloud voice library is further queried; the record number is updated (that is, the record number +1) and the record is judged Whether the number of times is greater than N times, if the number of recording times is not greater than N times, the command is ended; if the number of recording times is greater than N times, the mapping relationship is sent to the washing machine.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Acoustics & Sound (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Control Of Washing Machine And Dryer (AREA)
  • Telephonic Communication Services (AREA)

Abstract

涉及语音控制技术领域，具体涉及一种语音控制方法、云端服务器及终端设备。旨在解决现有的具有语音识别功能的洗衣机存在的用户体验差的问题。语音控制方法包括：终端设备接收语音信息；终端设备从本地语音库中查询语音信息对应的控制指令；当在本地语音库中未查询到语音信息对应的控制指令时，终端设备将语音信息上传至云端服务器；云端服务器从云端语音库中查询语音信息对应的控制指令；当在云端语音库中查询到语音信息对应的控制指令时，云端服务器向终端设备发送控制指令；终端设备接收控制指令，并基于控制指令执行相应的操作。能够提高终端设备的响应速度，大幅提升用户的用户体验。

Description

语音控制方法、云端服务器及终端设备 技术领域
本发明涉语音控制技术领域，具体涉及一种语音控制方法、云端服务器及终端设备。
背景技术
随着人们生活水平的日益提高，洗衣机已成为人们家中必备的家用电器之一，为了满足用户的差异化洗衣需求，现有洗衣机的功能越来越多，随之而来的是洗衣模式和洗衣参数的调节选项也越来越烦杂，这无疑增加了用户的学习和使用成本。
为解决功能的增多带来的学习和使用成本增加的问题，现有部分洗衣机搭载了语音识别模块，通过语音识别模块识别出用户所说的指令来启动相应的洗衣程序，从而避免繁杂的洗衣模式和洗衣参数的调节过程。虽然上述解决方案一定程度上简化了洗衣机的操作，但是也不可避免地存在如下问题：现有的语音识别模块主要分为离线语音识别模块和在线语音识别模块，离线语音识别模块虽然响应速度快，但是其只能识别特定的命令关键词，而且关键词的数量往往受存储空间的限制，用户想要开启特定的洗衣程序，必须准确地记住相对应的关键词，体验非常不好。相比于离线语音识别模块，在线语音识别模块则可以进行语义识别，通过将用户说出的命令语句上传到云端进行分析和解析，准确的执行用户所需的洗衣程序，但是在线语音识别模块需要将命令上传至云端解析，响应较慢，造成用户等待时间过长，因此体验也不是很好。
相应地，本领域需要一种新的语音控制方法来解决上述问题。
发明内容
为了解决现有技术中的上述问题，即为了解决现有的具有语音识别功能的洗衣机存在的用户体验差的问题，第一方面，本发明提供了一种语音控制方法，所述语音控制方法包括：
终端设备接收语音信息；
所述终端设备从本地语音库中查询所述语音信息对应的控制指令；
当在所述本地语音库中未查询到所述语音信息对应的控制指令时，所述终端设备将所述语音信息上传至云端服务器；
所述云端服务器从云端语音库中查询所述语音信息对应的控制指令；
当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令；
所述终端设备接收所述控制指令，并基于所述控制指令执行相应的操作。
在上述语音控制方法的优选技术方案中，在“所述云端服务器从云端语音库中查询所述语音信息对应的控制指令”的步骤之后，所述语音控制方法还包括：
当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；
所述云端服务器比较所述记录次数与预设次数的大小；
当所述记录次数大于所述预设次数时，所述云端服务器将所述映射关系发送至终端设备；
所述终端设备接收所述映射关系，并将所述映射关系存储至所述本地语音库。
在上述语音控制方法的优选技术方案中，所述语音控制方法还包括：
当在所述云端语音库中未查询到所述语音信息对应的控制指令时，所述云端服务器解析所述语音信息对应的控制指令；
当解析出所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，并将所述语音信息与所述控制指令之间的映射关系记录到所述云端语音库中；
所述终端设备接收所述控制指令，并基于所述控制指令执行相应的操作。
第二方面，本发明还提供了一种语音控制方法，所述语音控制方法包括：
云端服务器接收终端设备上传的语音信息；
所述云端服务器从云端语音库中查询所述语音信息对应的控制指令；
当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，以便所述终端设备基于所述控制指令执行相应的操作；
其中，所述终端设备在从本地语音库中无法查询到所述语音信息对应的控制指令时上传所述语音信息。
在上述语音控制方法的优选技术方案中，在“所述云端服务器从云端语音库中查询所述语音信息对应的控制指令”的步骤之后，所述语音控制方法还包括：
当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；
所述云端服务器比较所述记录次数与预设次数的大小；
当所述记录次数大于所述预设次数时，所述云端服务器将所述映射关系发送至所述终端设备，以便所述终端设备将所述映射关系存储至所述本地语音库。
在上述语音控制方法的优选技术方案中，所述语音控制方法还包括：
当在所述云端语音库中未查询到所述语音信息对应的控制指令时，所述云端服务器解析所述语音信息对应的控制指令；
当解析出所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，以便所述终端设备基于所述控制指令执行相应的操作；以及
所述云端服务器将所述语音信息与所述控制指令之间的映射关系记录到所述云端语音库中。
第三方面，本发明还提供了一种语音控制方法，所述语音控制方法包括：
终端设备接收语音信息；
所述终端设备从本地语音库中查询所述语音信息对应的控制指令；
当在所述本地语音库中未查询到所述语音信息对应的控制指令时，所述终端设备将所述语音信息上传至云端服务器，以便所述云端服务器从云端语音库中查询所述语音信息对应的控制指令或在所述云端语音库中未查询到时解析所述语音信息对应的控制指令；
所述终端设备从所述云端服务器接收查询到的或解析出的所述语音信息对应的控制指令，并基于所述控制指令执行相应的操作。
在上述语音控制方法的优选技术方案中，在“所述终端设备将所述语音信息上传至云端服务器”的步骤之后，所述控制方法还包括：
所述终端设备从所述云端服务器接收所述语音信息与所述控制指令之间的映射关系，并将所述映射关系存储至所述本地语音库；
其中，当所述云端服务器在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；所述映射关系是当所述记录次数大于预设次数时所述云端服务器发送的。
本发明还提供了一种云端服务器，所述云端服务器包括：存储器；处理器；以及计算机程序；其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器加载以便执行上述第二方面的优选技术方案中任一项所述的语音控制方法。
本发明还提供了一种终端设备，所述终端设备包括：存储器；处理器；以及计算机程序；其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器加载以便执行上述第三方面的优选技术方案中任一项所述的语音控制方法。
本领域技术人员能够理解的是，在本发明的优选技术方案中，语音控制方法包括：终端设备接收语音信息；终端设备从本地语音库中查询语音信息对应的控制指令；当在本地语音库中未查询到语音信息对应的控制指令时，终端设备将语音信息上传至云端服务器；云端服务器从云端语音库中查询语音信息对应的控制指令；当在云端语音库中查询 到语音信息对应的控制指令时，云端服务器向终端设备发送控制指令；终端设备接收控制指令，并基于控制指令执行相应的操作。
通过上述控制方式，本发明的语音控制方法能够提高带有语音识别功能的终端设备的响应速度，大幅提升用户的用户体验。具体而言，终端设备在接收语音信息后，首先进行本地查询，如果本地语音库中查询到该语音信息对应的控制指令，则按照该控制指令执行相应的操作。由于本地语音库的响应速度快，因此终端设备的操作执行快，用户体验好。如果本地语音库中未查询到该语音信息对应的控制指令，则通过将语音信息上传至云端服务器由云端服务器在云端语音库中进行查询，如果在云端语音库中查询到该语音信息对应的控制指令，云端服务器将控制指令发送给终端设备，从而终端设备基于该控制指令执行相应的操作。由于云端服务器具有超强的存储能力和运算能力，其云端语音库中存储有大量的语音信息和对应的控制指令，因此在本地语音库中未查询到控制指令时，通过云端服务器进行更大范围的查询，能够避免只通过本地识别时识别范围小、识别准确率低的问题，提升用户体验。
进一步地，当在云端语音库中查询到语音信息对应的控制指令时，通过云端服务器进一步确定语音信息与对应的控制指令之间的映射关系的记录次数，并且在该记录次数大于预设次数时，将对应的映射关系发送至终端设备，从而终端设备将该映射关系存储在本地语音库中，本发明的语音控制方法还使得终端设备能够基于用户的使用习惯自动更新本地语音库，使终端设备的语音识别功能的智能程度得到极大地提升，进而大幅度提高终端设备的响应速度和用户的使用体验。
进一步地，当在云端语音库中未查询到语音信息对应的控制指令时，通过云端服务器对该语音信息进行解析，并在解析出该语音信息对应的控制指令时，将该控制指令发送给终端设备，从而终端设备基于该控制指令执行相应的操作，本发明的语音控制方法由于云端服务器能够进行语音语义识别，并且其识别范围广、准确性高，因此能够在云端语音库中未查询到对应的控制指令时进一步提高识别效率和识别准确率，提升用户体验。
附图说明
下面参照附图并结合洗衣机来描述本发明的语音控制方法、云端服务器及终端设备。附图中：
图1为本发明的语音控制方法的一种实施方式流程图；
图2为本发明的语音控制方法的又一种实施方式的流程图；
图3为本发明的语音控制方法的再一种实施方式的流程图；
图4为本发明的语音控制方法的一种实施方式的逻辑图；
图5为本发明的语音控制方法的又一种实施方式的逻辑图。
具体实施方式
下面参照附图来描述本发明的优选实施方式。本领域技术人员应当理解的是，这些实施方式仅仅用于解释本发明的技术原理，并非旨在限制本发明的保护范围。例如，虽然本实施方式是结合洗衣机进行描述的，但是本发明的应用场景并非只限于此，在不偏离本发明原理的条件下，本领域技术人员可以将本发明的控制方法应用于其他终端设备，只要该终端设备具有语音识别功能即可。如，显然本发明的控制方法还可以应用于带有语音识别功能的电冰箱、电视机、音箱等。
实施例一
首先参照图1，对本发明的语音控制方法的第一种实施方式进行描述。其中，图1为本发明的语音控制方法的一种实施方式流程图。该方法的执行主体是洗衣机和云端服务器(以下或简称云端)，其中，洗衣机设置有控制模块和语音识别模块，语音识别模块能够识别语音信息对应的控制指令，并且在语音信息能够转化为控制指令时将控制指令发送给控制模块，从而控制模块基于该控制指令控制洗衣机执行相应的操作。其中，云端服务器同样能够识别语音信息对应的控制指令，并且云端服务器还能够直接与语音识别模块通信或通过与洗衣机通信从而间接与语音识别模块通信，以接收语音识别模块或洗衣机发送的语音信息，以及向语音识别模块或洗衣机发送指令信息或语音信息与控制指令之间的映射关系，以便洗衣机能够执行相应的操作。同样地，本实施方式中对云端服务器的形式不做限制，任何不在本地的远程服务器均可作为云端服务器应用于本发明中。
如图1所示，为了解决现有的具有语音识别功能的洗衣机存在的用户体验差的问题，本发明的语音控制方法主要包括以下步骤：
S101、洗衣机接收语音信息。例如，语音信息可以通过洗衣机上设置的麦克风等采集元件获取，也可以通过接收音频文件后转换获取，本实施方式对此不做限制。如在洗衣机开机后，麦克风一直处于导通状态从而获取语音信息，或者在用户按下洗衣机上设置的语音输入按钮后处于导通状态而获取语音信息；再如，用户通过对移动设备(如手机、平板电脑等)输入语音信息，从而洗衣机通过接收移动设备发送的音频文件后转换成语音信息。
S102、洗衣机从本地语音库中查询语音信息对应的控制指令。例如，洗衣机中存储有本地语音库，本地语音库中存储有语音信息与控制指令的映射关系，洗衣机在接收语音信息后，通过控制语音识别模块查询本地语音库中存储的映射关系来识别是否有该语音信息对应的控制指令。
S103、当在本地语音库中未查询到语音信息对应的控制指令时，洗衣机将语音信息上传至云端服务器。例如，当在本地语音库中查询到语音信息对应的控制指令时将该控制指令发送给控制模块，从而控制模块基于该控制指令控制洗衣机执行相应的操作，如执行洗衣程序、脱水程序等。当在本地语音库中未查询到该语音信息对应的控制指令时，语音识别模块或洗衣机将该语音信息上传至云端服务器，以便利用云端服务器强大的处理能力对该语音信息进行处理。其中语音识别模块或洗衣机通过有线或无线的方式将语音信息上传至云端服务器，本实施方式对此不做限制。
S104、云端服务器从云端语音库中查询语音信息对应的控制指令。例如，在云端服务器中存储有云端语音库，云端语音库中存储有大量的语音信息与控制指令的映射关系，该映射关系可以是事先存储的，也可以是连接到该服务器的所有用户在使用过程中生成的；云端服务器在接收到语音信息后，从云端语音库中存储的映射关系来识别是否有该语音信息对应的控制指令。
S105、当在所述云端语音库中查询到所述语音信息对应的控制指令时，云端服务器向洗衣机发送该控制指令。例如，当在云端语音 库中查询到该语音信息对应的控制指令时，云端服务器向洗衣机发送该控制指令，以便洗衣机能够基于该控制指令执行相应的操作。
S106、洗衣机接收控制指令，并基于控制指令执行相应的操作。例如，在云端服务器将控制指令发送给洗衣机后，控制模块接收该控制指令，并基于该控制指令控制洗衣机执行相应的操作，如执行洗衣程序、脱水程序等。
通过上述控制方式，本发明的语音控制方法能够提高洗衣机的响应速度，大幅提升用户的用户体验。具体而言，洗衣机在接收语音信息后，首先控制语音识别模块进行本地查询，如果本地语音库中查询到该语音信息对应的控制指令，则控制模块按照该控制指令控制洗衣机执行相应的操作。由于本地语音库的响应速度快，因此这种情况下洗衣机的操作执行快，用户体验好。如果本地语音库中未查询到该语音信息对应的控制指令，则通过将语音信息上传至云端服务器由云端服务器在云端语音库中进行查询，如果在云端语音库中查询到该语音信息对应的控制指令，云端服务器将控制指令发送给洗衣机的控制模块，从而控制模块基于该控制指令控制洗衣机执行相应的操作。由于云端服务器具有超强的存储能力和运算能力，其云端语音库中存储有大量的映射关系，因此在本地语音库中未查询到控制指令时，通过云端服务器进行更大范围的查询，能够避免只通过本地进行语音识别时识别范围小、识别准确率低的问题，提升用户体验。
作为一种优选的实施方式，在步骤S104之后，所述语音控制方法还包括如下步骤：
当在云端语音库中查询到语音信息对应的控制指令时，云端服务器确定语音信息与控制指令之间的映射关系在云端语音库中的记录次数。例如，在云端语音库中查询到语音信息对应的控制指令时，证明云端语音库已经存储有该语音信息与控制指令之间的映射关系，云端服务器可以在向洗衣机发送控制指令之后，进一步确定云端语音库中该映射关系的记录次数。
云端服务器比较记录次数与预设次数的大小。例如，云端服务器从云端语音库中确定出记录次数之后，将本次的映射关系更新至该记录次数(即记录次数+1)并与预设次数相比较。其中，预设次数优选 地为10-30次，更为优选的为20次。当然，预设次数的设置本领域技术人员可以基于不用的应用场景进行调整。此外，记录次数在与预设次数比较之前，可以将本次的记录更新至记录次数中，也可以不更新直接比较。
当记录次数大于预设次数时，证明该映射关系使用频率较高，此时云端服务器将该映射关系发送至洗衣机，洗衣机接收该映射关系，并将该映射关系存储至本地语音库，以便下次用户输入相同的语音信息时，洗衣机能够直接从本地语音库中查询到相应的控制指令。当记录次数小于预设次数时，证明该映射关系使用的频率较低，此时无需将该映射关系发送至洗衣机，以避免映射关系过多而占用本地语音库的存储空间。
当在云端语音库中未查询到语音信息对应的控制指令时，云端服务器解析语音信息对应的控制指令；例如，云端服务器在云端语音库中未查询到相同的映射关系，证明该映射关系之前并未记录在云端语音库中，也就是该用户或其他用户均未使用过该映射关系，此时云端服务器对该语音信息进行语音和/或语义的解析识别。当云端服务器未能解析出语音信息对应的控制指令时，可以向洗衣机发送解析失败指令，从而控制模块收到该解析失败指令后控制洗衣机发出提示信息，如发出提示音提示用户重新输入，或控制指示灯闪烁等。当云端服务器成功解析出语音信息对应的控制指令时，可以向洗衣机发送该控制指令，以便洗衣机能够基于该控制指令执行相应的操作。此外，在云端服务器成功解析出语音信息对应的控制指令的同时，云端服务器还可以将该映射关系记录到云端语音库中，以便下次有相同的映射关系时方便云端服务器从云端语音库中调取该映射关系的记录次数。
洗衣机接收控制指令，并基于控制指令执行相应的操作。例如，在云端服务器成功解析出控制指令并发送给洗衣机后，控制模块接收该控制指令，并基于该控制指令控制洗衣机执行相应的操作，如执行洗衣程序、脱水程序等。
通过上述设置方式，本发明的控制方法还使得洗衣机能够基于用户的使用习惯自动更新本地语音库，从而使洗衣机的语音识别功能的智能程度得到极大地提升，大幅度提高洗衣机的响应速度和用户的使 用体验。实际上，由于本地语音库的存储容量有限，不能够像云端服务器一样存储大量的映射关系，而在应用本发明的语音控制方法后，本地语音库中始终存储着用户使用频率最高的一些映射关系，从而使洗衣机能够高度适应用户的使用习惯，真正做到洗衣机的智能化控制。进一步地，当在云端语音库中未查询到语音信息对应的控制指令时，通过云端服务器对该语音信息进行解析，并在解析出该语音信息对应的控制指令时，将该控制指令发送给洗衣机，从而洗衣机基于该控制指令执行相应的操作，本发明的语音控制方法由于云端服务器能够进行语音语义识别，并且其识别范围广、准确性高，因此能够在云端语音库中未查询到对应的控制指令时进一步提高识别效率和识别准确率，提升用户体验。
需要说明的是，上述优选的实施方式仅仅用于阐述本实用新型的原理，并非旨在于限制本实用新型的保护范围。在不偏离本实用新型原理的前提下，本领域技术人员可以对上述设置方式进行调整，以便本实用新型能够适用于更加具体的应用场景。例如，在一种可替换的实施方式中，云端服务器可以在向洗衣机发送控制指令的同时或之前，确定云端语音库中该映射关系的记录次数，这种步骤顺序的更改并未偏离本发明的原理，因此理应落入本发明的保护范围之中。
实施例二
参照图2，对本发明的语音控制方法的第二种实施方式进行描述。其中，图2为本发明的语音控制方法的又一种实施方式流程图。该方法的执行主体是云端服务器(以下或简称云端)。其中，以下实施方式涉及到的洗衣机和云端服务器的设置方式与实施例一相同，在此不再赘述。
如图2所示，本发明的语音控制方法主要包括以下步骤：
S201、云端服务器接收洗衣机上传的语音信息；
S202、云端服务器从云端语音库中查询语音信息对应的控制指令；
S203、当在云端语音库中查询到语音信息对应的控制指令时，云端服务器向洗衣机发送控制指令，以便洗衣机基于控制指令执行相应的操作；其中，洗衣机在从本地语音库中无法查询到语音信息对应的控制指令时上传语音信息。
通过洗衣机在从本地语音库中无法查询到语音信息对应的控制指令时上传语音信息至云端服务器，从而云端服务器在云端语音库中查询该语音信息对应的控制指令，并在查询到时将该控制指令发送给洗衣机，以便洗衣机能够基于该控制指令执行相应的操作，本发明的语音控制方法能够提高洗衣机的识别准确率，避免洗衣机只通过本地语音库中有限的映射关系进行识别带来的准确率低的问题，大幅提升用户体验。
为进一步提升用户体验，作为一种优选的实施方式，本发明的语音控制方法在步骤S202之后还包括：
当在云端语音库中查询到语音信息对应的控制指令时，云端服务器确定语音信息与控制指令之间的映射关系在云端语音库中的记录次数；云端服务器比较记录次数与预设次数的大小；当记录次数大于预设次数时，云端服务器将映射关系发送至洗衣机，以便洗衣机将映射关系存储至本地语音库。当在云端语音库中未查询到语音信息对应的控制指令时，云端服务器解析语音信息对应的控制指令；当解析出语音信息对应的控制指令时，云端服务器向洗衣机发送控制指令，以便洗衣机基于控制指令执行相应的操作；以及云端服务器将语音信息与控制指令之间的映射关系记录到云端语音库中。
在云端语音库中查询到语音信息对应的控制指令时，通过云端服务器确定语音信息与对应的控制指令之间的映射关系的记录次数，并且在该记录次数大于预设次数时，将对应的映射关系发送至洗衣机，以便洗衣机将该映射关系存储在本地语音库中，本发明的语音控制方法使得洗衣机还能够基于用户的使用习惯自动更新洗衣机中的本地语音库，使洗衣机的语音识别功能的智能程度得到极大地提升，从而进一步提高洗衣机的响应速度和用户的使用体验。进一步地，在云端语音库中未查询到语音信息对应的控制指令时，通过云端服务器对该语音信息进行解析得到对应的控制指令，并将该控制指令发送给洗衣机，以便洗衣机能够基于该控制指令执行相应的操作，本发明的语音控制方法还能够提高洗衣机的识别准确率，避免在云端语音库中也未存储映射关系时带来的准确率低的问题，大幅提升用户体验。
实施例三
参照图3，对本发明的语音控制方法的第三种实施方式进行描述。其中，图3为本发明的语音控制方法的再一种实施方式流程图。该方法的执行主体是洗衣机。其中，以下实施方式涉及到的洗衣机和云端服务器(或简称云端)的设置方式与实施例一相同，在此不再赘述。
如图3所示，本发明的语音控制方法主要包括以下步骤：
S301、洗衣机接收语音信息；
S302、洗衣机从本地语音库中查询语音信息对应的控制指令；
S303、当在本地语音库中未查询到语音信息对应的控制指令时，洗衣机将语音信息上传至云端服务器，以便云端服务器从云端语音库中查询语音信息对应的控制指令或在云端语音库中未查询到时解析语音信息对应的控制指令；
S304、洗衣机从云端服务器接收查询到的或解析出的语音信息对应的控制指令，并基于控制指令执行相应的操作。
通过洗衣机接收到语音信息后先在本地语音库中查询控制指令，如果本地语音库中查询到相应的控制指令则按照该控制指令执行相应的操作，本发明的语音控制方法能够使洗衣机的操作执行快，用户体验好。而在从本地语音库中无法查询到语音信息对应的控制指令时，通过上传语音信息至云端服务器，以便云端服务器从云端语音库中查询语音信息对应的控制指令或在云端语音库中未查询到时解析语音信息对应的控制指令，然后洗衣机接收查询到的或解析出的语音信息对应的控制指令，并基于控制指令进行相应的操作，本发明的语音控制方法还能够在无法在本地识别语音信息时通过云端服务器的解析提高洗衣机的识别准确率，避免洗衣机只通过本地语音库中有限的映射关系进行识别带来的准确率低的问题，大幅提升用户体验。
为进一步提升用户体验，作为一种优选的实施方式，本发明的语音控制方法在步骤S303之后还包括：
洗衣机从云端服务器接收语音信息与控制指令之间的映射关系，并将映射关系存储至本地语音库；其中，当云端服务器在云端语音库中查询到语音信息对应的控制指令时，云端服务器确定语音信息与控制指令之间的映射关系在云端语音库中的记录次数；映射关系是当记录次数大于预设次数时云端服务器发送的。
通过云端服务器解析出语音信息对应的控制指令时，确定语音信息与对应的控制指令之间的映射关系在云端语音库中的记录次数，并且在该记录次数大于预设次数时，将对应的映射关系发送至洗衣机，从而洗衣机接收该映射关系并将该映射关系存储在本地语音库中，本发明的语音控制方法使得洗衣机还能够基于用户的使用习惯自动更新洗衣机中的本地语音库，使洗衣机的语音识别功能的智能程度得到极大地提升，从而进一步提高洗衣机的响应速度和用户的使用体验。
此外，本发明还提供了一种云端服务器，该云端服务器包括存储器、处理器以及计算机程序，其中存储器与处理器连接，计算机程序存储在存储器中，并被配置为由处理器加载以便执行实施例二中所述的语音控制方法。
此外，本发明还提供了一种洗衣机，该洗衣机包括存储器、处理器以及计算机程序，其中存储器与处理器连接，计算机程序存储在存储器中，并被配置为由处理器加载以便执行实施例三中所述的语音控制方法。
下面参照图4和图5，对本发明的控制方法的控制过程进行描述。其中，图4为本发明的语音控制方法的一种实施方式的逻辑图；图5为本发明的语音控制方法的又一种实施方式的逻辑图。
首先参照图4，在一种可能的实施方式中，洗衣机接收用户的语音信息，语音识别模块基于该语音信息在本体语音库中查询对应的控制指令；当在本地语音库中查询到该语音信息对应的控制指令时，语音识别模块将控制指令发送给控制模块，控制模块基于该控制指令控制洗衣机执行相应的操作；当未查询到该语音信息对应的控制指令时，洗衣机将该语音信息通过无线的方式上传至云端，云端服务器在云端语音库中查询该语音信息对应的控制指令，当在云端语音库中查询到控制指令时，则将该控制指令返回洗衣机的控制模块，从而洗衣机的控制模块基于该控制指令控制洗衣机执行相应的操作；当在云端语音库中未查询到语音信息对应的控制指令时，云端服务器进一步对该语音信息进行解析，如果成功解析出该语音信息对应的控制指令，则将该控制指令返回洗衣机的控制模块，从而洗衣机的控制模块基于该控制指令控制洗衣机执行相应的操作；如果云端服务器未成功解析出该语音信息对应的控制 指令，则将失败指令返回洗衣机的控制模块，从而控制模块控制洗衣机发出提示音，提示用户重新输入语音信息。
接下来参照图5，在另一种可能的实施方式中，云端在接收到洗衣机上传的语音信息后，从云端语音库中查询该语音信息对应的控制指令；如果未查询到语音信息对应的控制指令时，则对语音信息进行解析，如果未解析出该语音信息对应的指令，则结束命令；如果解析出对应的控制指令，则将该语音信息与控制指令之间的映射关系记录到云端语音库中；如果查询到语音信息对应的控制指令，则进一步查询该语音信息与该控制指令之间的映射关系在云端语音库中的记录次数；更新记录次数(即记录次数+1)并判断记录次数是否大于N次，如果记录次数不大于N次，则结束命令；如果记录次数大于N次，则将该映射关系发送给洗衣机。
至此，已经结合附图所示的优选实施方式描述了本发明的技术方案，但是，本领域技术人员容易理解的是，本发明的保护范围显然不局限于这些具体实施方式。在不偏离本发明的原理的前提下，本领域技术人员可以对相关技术特征作出等同的更改或替换，这些更改或替换之后的技术方案都将落入本发明的保护范围之内。

Claims (10)

  1. 一种语音控制方法，其特征在于，所述语音控制方法包括：
    终端设备接收语音信息；
    所述终端设备从本地语音库中查询所述语音信息对应的控制指令；
    当在所述本地语音库中未查询到所述语音信息对应的控制指令时，所述终端设备将所述语音信息上传至云端服务器；
    所述云端服务器从云端语音库中查询所述语音信息对应的控制指令；
    当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令；
    所述终端设备接收所述控制指令，并基于所述控制指令执行相应的操作。
  2. 根据权利要求1所述的语音控制方法，其特征在于，在“所述云端服务器从云端语音库中查询所述语音信息对应的控制指令”的步骤之后，所述语音控制方法还包括：
    当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；
    所述云端服务器比较所述记录次数与预设次数的大小；
    当所述记录次数大于所述预设次数时，所述云端服务器将所述映射关系发送至终端设备；
    所述终端设备接收所述映射关系，并将所述映射关系存储至所述本地语音库。
  3. 根据权利要求2所述的语音控制方法，其特征在于，所述语音控制方法还包括：
    当在所述云端语音库中未查询到所述语音信息对应的控制指令时，所述云端服务器解析所述语音信息对应的控制指令；
    当解析出所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，并将所述语音信息与所述控制指令之间的映 射关系记录到所述云端语音库中；
    所述终端设备接收所述控制指令，并基于所述控制指令执行相应的操作。
  4. 一种语音控制方法，其特征在于，所述语音控制方法包括：
    云端服务器接收终端设备上传的语音信息；
    所述云端服务器从云端语音库中查询所述语音信息对应的控制指令；
    当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，以便所述终端设备基于所述控制指令执行相应的操作；
    其中，所述终端设备在从本地语音库中无法查询到所述语音信息对应的控制指令时上传所述语音信息。
  5. 根据权利要求4所述的语音控制方法，其特征在于，在“所述云端服务器从云端语音库中查询所述语音信息对应的控制指令”的步骤之后，所述语音控制方法还包括：
    当在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；
    所述云端服务器比较所述记录次数与预设次数的大小；
    当所述记录次数大于所述预设次数时，所述云端服务器将所述映射关系发送至所述终端设备，以便所述终端设备将所述映射关系存储至所述本地语音库。
  6. 根据权利要求5所述的语音控制方法，其特征在于，所述语音控制方法还包括：
    当在所述云端语音库中未查询到所述语音信息对应的控制指令时，所述云端服务器解析所述语音信息对应的控制指令；
    当解析出所述语音信息对应的控制指令时，所述云端服务器向所述终端设备发送所述控制指令，以便所述终端设备基于所述控制指令执行相应的操作；以及
    所述云端服务器将所述语音信息与所述控制指令之间的映射关系记录到所述云端语音库中。
  7. 一种语音控制方法，其特征在于，所述语音控制方法包括：
    终端设备接收语音信息；
    所述终端设备从本地语音库中查询所述语音信息对应的控制指令；
    当在所述本地语音库中未查询到所述语音信息对应的控制指令时，所述终端设备将所述语音信息上传至云端服务器，以便所述云端服务器从云端语音库中查询所述语音信息对应的控制指令或在所述云端语音库中未查询到时解析所述语音信息对应的控制指令；
    所述终端设备从所述云端服务器接收查询到的或解析出的所述语音信息对应的控制指令，并基于所述控制指令执行相应的操作。
  8. 根据权利要求7所述的语音控制方法，其特征在于，在“所述终端设备将所述语音信息上传至云端服务器”的步骤之后，所述控制方法还包括：
    所述终端设备从所述云端服务器接收所述语音信息与所述控制指令之间的映射关系，并将所述映射关系存储至所述本地语音库；
    其中，当所述云端服务器在所述云端语音库中查询到所述语音信息对应的控制指令时，所述云端服务器确定所述语音信息与所述控制指令之间的映射关系在所述云端语音库中的记录次数；所述映射关系是当所述记录次数大于预设次数时所述云端服务器发送的。
  9. 一种云端服务器，其特征在于，所述云端服务器包括：
    存储器；
    处理器；以及
    计算机程序；
    其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器加载以便执行权利要求4至6中任一项所述的语音控制方法。
  10. 一种终端设备，其特征在于，所述终端设备包括：
    存储器；
    处理器；以及
    计算机程序；
    其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器加载以便执行权利要求7至8中任一项所述的语音控制方法。
PCT/CN2019/120823 2018-12-11 2019-11-26 语音控制方法、云端服务器及终端设备 WO2020119437A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP19895476.0A EP3896688A4 (en) 2018-12-11 2019-11-26 VOICE CONTROL METHOD, CLOUD SERVER AND TERMINAL DEVICE
JP2021533280A JP7159476B2 (ja) 2018-12-11 2019-11-26 音声制御方法、クラウドサーバおよび端末装置
US17/413,140 US11705129B2 (en) 2018-12-11 2019-11-26 Voice control method, cloud server and terminal device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811513020.3A CN111312234A (zh) 2018-12-11 2018-12-11 语音控制方法、云端处理器及终端设备
CN201811513020.3 2018-12-11

Publications (1)

Publication Number Publication Date
WO2020119437A1 true WO2020119437A1 (zh) 2020-06-18

Family

ID=71075418

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/120823 WO2020119437A1 (zh) 2018-12-11 2019-11-26 语音控制方法、云端服务器及终端设备

Country Status (5)

Country Link
US (1) US11705129B2 (zh)
EP (1) EP3896688A4 (zh)
JP (1) JP7159476B2 (zh)
CN (1) CN111312234A (zh)
WO (1) WO2020119437A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220028384A1 (en) * 2018-12-11 2022-01-27 Qingdao Haier Washing Machine Co., Ltd. Voice control method, cloud server and terminal device

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111312234A (zh) * 2018-12-11 2020-06-19 青岛海尔洗衣机有限公司 语音控制方法、云端处理器及终端设备
CN112581953A (zh) * 2020-11-09 2021-03-30 金茂智慧科技(广州)有限公司 一种在线与离线语音结合的控制面板
CN112581952A (zh) * 2020-11-09 2021-03-30 金茂智慧科技(广州)有限公司 一种在线与离线语音结合的智能设备控制方法
CN114244879A (zh) * 2021-12-15 2022-03-25 北京声智科技有限公司 一种工业控制系统、工业控制方法和电子设备
CN114550719A (zh) * 2022-02-21 2022-05-27 青岛海尔科技有限公司 语音控制指令的识别方法和装置、存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102496364A (zh) * 2011-11-30 2012-06-13 苏州奇可思信息科技有限公司 基于云端网络的交互式语音识别方法
CN102708865A (zh) * 2012-04-25 2012-10-03 北京车音网科技有限公司 语音识别方法、装置及系统
CN104538034A (zh) * 2014-12-31 2015-04-22 深圳雷柏科技股份有限公司 一种语音识别方法及系统
WO2016063158A1 (en) * 2014-10-20 2016-04-28 Vocalzoom Systems Ltd. System and method for operating devices using voice commands
CN108510984A (zh) * 2017-02-24 2018-09-07 九阳股份有限公司 一种智能家电设备的语音协同控制方法
CN108520747A (zh) * 2018-03-29 2018-09-11 浙江吉利汽车研究院有限公司 一种具有语音识别功能的车载控制装置

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA1229922A (en) * 1985-01-17 1987-12-01 James K. Baker Speech recognition training method
CN1859592A (zh) * 2005-12-28 2006-11-08 华为技术有限公司 一种模块化语种解析装置及其实现方法
AU2014200663B2 (en) * 2010-01-26 2015-05-28 Google Inc. Integration of embedded and network speech recognizers
JP5887162B2 (ja) 2012-02-23 2016-03-16 株式会社Ｎｔｔドコモ 通信端末、制御方法及びプログラム
KR101330671B1 (ko) 2012-09-28 2013-11-15 삼성전자주식회사 전자장치, 서버 및 그 제어방법
KR20140055502A (ko) 2012-10-31 2014-05-09 삼성전자주식회사 방송 수신 장치, 서버 및 그 제어 방법
US9734151B2 (en) 2012-10-31 2017-08-15 Tivo Solutions Inc. Method and system for voice based media search
CN104575494A (zh) * 2013-10-16 2015-04-29 中兴通讯股份有限公司 一种语音处理的方法和终端
US10031721B2 (en) * 2014-05-15 2018-07-24 Tyco Safety Products Canada Ltd. System and method for processing control commands in a voice interactive system
US10224034B2 (en) * 2016-02-03 2019-03-05 Hua Xu Voice recognition system and construction method thereof
CN105913847B (zh) * 2016-06-01 2020-10-16 北京京东尚科信息技术有限公司 语音控制系统、用户端设备、服务器和中央控制单元
CN108010522A (zh) * 2016-11-02 2018-05-08 松下电器（美国）知识产权公司 信息处理方法以及记录介质
CA3155434A1 (en) * 2017-02-07 2018-08-16 Lutron Technology Company Llc Audio-based load control system
CN108538291A (zh) * 2018-04-11 2018-09-14 百度在线网络技术（北京）有限公司 语音控制方法、终端设备、云端服务器及系统
CN108958053A (zh) * 2018-07-20 2018-12-07 杭州电子科技大学 基于树莓派的微信远程监控方法
CN111312234A (zh) * 2018-12-11 2020-06-19 青岛海尔洗衣机有限公司 语音控制方法、云端处理器及终端设备
CN111312253A (zh) * 2018-12-11 2020-06-19 青岛海尔洗衣机有限公司 语音控制方法、云端服务器及终端设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102496364A (zh) * 2011-11-30 2012-06-13 苏州奇可思信息科技有限公司 基于云端网络的交互式语音识别方法
CN102708865A (zh) * 2012-04-25 2012-10-03 北京车音网科技有限公司 语音识别方法、装置及系统
WO2016063158A1 (en) * 2014-10-20 2016-04-28 Vocalzoom Systems Ltd. System and method for operating devices using voice commands
CN104538034A (zh) * 2014-12-31 2015-04-22 深圳雷柏科技股份有限公司 一种语音识别方法及系统
CN108510984A (zh) * 2017-02-24 2018-09-07 九阳股份有限公司 一种智能家电设备的语音协同控制方法
CN108520747A (zh) * 2018-03-29 2018-09-11 浙江吉利汽车研究院有限公司 一种具有语音识别功能的车载控制装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3896688A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220028384A1 (en) * 2018-12-11 2022-01-27 Qingdao Haier Washing Machine Co., Ltd. Voice control method, cloud server and terminal device
US11967320B2 (en) * 2018-12-11 2024-04-23 Qingdao Haier Washing Machine Co., Ltd. Processing voice information with a terminal device and a cloud server to control an operation

Also Published As

Publication number Publication date
JP2022513785A (ja) 2022-02-09
CN111312234A (zh) 2020-06-19
EP3896688A4 (en) 2022-05-18
EP3896688A1 (en) 2021-10-20
JP7159476B2 (ja) 2022-10-24
US20220028383A1 (en) 2022-01-27
US11705129B2 (en) 2023-07-18

Similar Documents

Publication Publication Date Title
WO2020119437A1 (zh) 语音控制方法、云端服务器及终端设备
WO2020119438A1 (zh) 语音控制方法、云端服务器及终端设备
US20220197593A1 (en) Conditionally assigning various automated assistant function(s) to interaction with a peripheral assistant control device
US11822857B2 (en) Architecture for a hub configured to control a second device while a connection to a remote system is unavailable
US10074365B2 (en) Voice control method, mobile terminal device, and voice control system
WO2019091171A1 (zh) 语音控制智能家电的方法、装置、系统和电子设备
WO2016206494A1 (zh) 语音控制方法、装置和移动终端
US20230053765A1 (en) Speech Control Method and Apparatus, Server, Terminal Device, and Storage Medium
WO2020014899A1 (zh) 语音控制方法、中控设备和存储介质
US20140129032A1 (en) Genetic learning for environmental control automation
JP2014071449A (ja) 電子装置、サーバー及びその制御方法
CN108989397B (zh) 数据推荐方法、装置及存储介质
US20230335137A1 (en) Hub device, multi-device system including the hub device and plurality of devices, and method of operating the same
US20140278440A1 (en) Framework for voice controlling applications
CN114067798A (zh) 一种服务器、智能设备及智能语音控制方法
WO2020135080A1 (zh) 电视设备显示的控制方法、系统、终端及存储介质
JP7288885B2 (ja) 音声対話方法、装置、機器および記憶媒体
CN110691016B (zh) 一种基于音频设备实现的交互方法及音频设备
CN112084210A (zh) 数据处理方法、装置、电子设备及存储介质
CN110555075B (zh) 数据处理方法、装置、电子设备以及计算机可读存储介质
KR20200127823A (ko) 허브 디바이스, 허브 디바이스 및 복수의 디바이스를 포함하는 멀티 디바이스 시스템 및 그 동작 방법
US20230022642A1 (en) Audio services agent manager
KR102575697B1 (ko) 음향을 제어하는 공용 단말, 방법 및 컴퓨터 프로그램
CN110769177B (zh) 计算机可读存储介质、基于通讯软件的多媒体资源录制方法及系统
CN115841815A (zh) 一种显示设备和语音报修方法

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: 19895476

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021533280

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019895476

Country of ref document: EP

Effective date: 20210712