WO2020029500A1 - 语音命令定制方法、装置和设备及计算机存储介质 - Google Patents

语音命令定制方法、装置和设备及计算机存储介质 Download PDF

Info

Publication number
WO2020029500A1
WO2020029500A1 PCT/CN2018/121040 CN2018121040W WO2020029500A1 WO 2020029500 A1 WO2020029500 A1 WO 2020029500A1 CN 2018121040 W CN2018121040 W CN 2018121040W WO 2020029500 A1 WO2020029500 A1 WO 2020029500A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice command
voice
user
operation instruction
information
Prior art date
Application number
PCT/CN2018/121040
Other languages
English (en)
French (fr)
Inventor
韦泽光
张玉
陈琳婷
杨煜豪
程万里
Original Assignee
珠海格力电器股份有限公司
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 珠海格力电器股份有限公司 filed Critical 珠海格力电器股份有限公司
Publication of WO2020029500A1 publication Critical patent/WO2020029500A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/22Procedures used during a speech recognition process, e.g. man-machine dialogue
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/22Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/223Execution procedure of a spoken command

Definitions

  • the present invention relates to the field of computer technology, and in particular, to a method, an apparatus, and a device for customizing a voice command, and a computer storage medium.
  • smart home devices provide more convenience to people's lives, and there are more and more control methods for smart home devices.
  • users can control smart home devices through voice, or through applications installed on terminals.
  • applications Application, APP
  • voice the association relationship between the voice library and the function is usually defined in advance when the smart home device leaves the factory, and the user can perform corresponding voice control according to the instructions or prompts.
  • the voice and function instruction sets are predefined, but there are certain differences in the usage habits of different users.
  • the predefined voice control schemes may not meet the user's usage habits.
  • the user's experience when controlling a smart home device by voice is obviously poor, and the limited functions included in the predefined voice control scheme may not meet the user's needs, further reducing the user's experience.
  • Embodiments of the present invention provide a method, an apparatus, and a device for customizing a voice command, and a computer storage medium, which are used to customize a voice control scheme and improve a user experience.
  • a method for customizing a voice command includes:
  • the user's voice information can be collected to generate a voice command, and corresponding operation instructions are generated according to the user's demonstration operation, and the two are associated.
  • the voice command is generated based on the user's voice information.
  • the operation instructions obtained based on the user's demonstration operation can be not limited to the voice control instructions when the smart home device leaves the factory, and the application range is wider.
  • the generating a first voice command according to the collected voice information includes:
  • the first voice command is generated based on the common characteristics.
  • the establishing an association relationship between the first voice command and the first operation instruction includes:
  • the first feedback information indicates that the result of executing the first operation instruction meets the requirements of the user, store the first voice command after associating with the first operation instruction, otherwise prompt the user again Perform a demo operation.
  • the operation instruction is executed once, and then the operation instruction is verified to confirm whether the function implemented by the operation instruction is consistent with the expected effect of the user.
  • the method further includes:
  • the voice information input by the user can be semantically recognized, and the voice command generated based on the voice information can be associated with the semantic recognition result, so that all the instructions associated with the same semantic recognition result can achieve the same Function, so that even if the user enters a dialect or other language, it can support it, improving the generalization ability of speech recognition.
  • the method further includes:
  • the recorded data is recorded. Said second voice command;
  • the voice command that matches the re-entered voice command is updated according to the second voice command.
  • the voice command database may be updated according to the voice command input for the first time to improve the recognition capability of the voice command database.
  • the method further includes:
  • Output a second prompt message for prompting the user whether to set an associated operation instruction when the voice commands received multiple times consecutively fail to match any of the voice commands in the voice command library;
  • the first prompt information is output.
  • a voice command customization device including:
  • a generating unit configured to generate a first voice command according to the collected voice information
  • An output unit configured to output a first prompt message instructing a user to input a demonstration operation for realizing at least one function of the smart home device
  • the generating unit is further configured to generate a first operation instruction for performing an operation step in the demonstration operation process based on the demonstration operation;
  • An association unit configured to establish an association relationship between the first voice command and the first operation instruction, and store the association relationship, so that when a voice command matching the first voice command is received To execute the first operation instruction.
  • the generating unit is specifically configured to:
  • the first voice command is generated based on the common characteristics.
  • association unit is specifically configured to:
  • the first feedback information indicates that the result of executing the first operation instruction meets the requirements of the user, store the first voice command after associating with the first operation instruction, otherwise prompt the user again Perform a demo operation.
  • the device further includes a semantic recognition unit, configured to:
  • the device further includes an update unit, configured to:
  • the recorded data is recorded. Said second voice command;
  • the voice command that matches the re-entered voice command is updated according to the second voice command.
  • the output unit is further configured to:
  • Output a second prompt message for prompting the user whether to set an associated operation instruction when the voice commands received multiple times consecutively fail to match any of the voice commands in the voice command library;
  • the first prompt information is output.
  • a voice command customization device including:
  • At least one processor At least one processor
  • a memory connected in communication with the at least one processor; wherein,
  • the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to the first aspect.
  • a computer storage medium is provided.
  • the computer storage medium stores computer instructions, and when the computer instructions are run on a computer, the computer is caused to execute the method according to the first aspect.
  • FIG. 1 is a schematic flowchart of a voice command customization method according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of correlating a semantic recognition result with a voice command according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of a voice control process according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of a voice command customization device according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a voice command customization device according to an embodiment of the present invention.
  • both the voice and function instruction sets in the voice control scheme are pre-defined, but different users have different usage habits.
  • the predefined voice control scheme may not meet the user's usage habits.
  • the use experience when controlling a smart home device by voice is obviously poor, and the limited functions included in the pre-defined voice control scheme may not meet the user's needs, further reducing the user's experience.
  • embodiments of the present invention provide a method, device, and device for customizing a voice command, and a computer storage medium.
  • a user's voice information can be collected to generate a voice command, and a corresponding operation instruction is generated according to a user's demonstration operation.
  • the two are related, so that the voice command is generated based on the user's voice information, which can better conform to the user's speaking habits, and the operation instructions obtained based on the user's demonstration operation can be not limited to when the smart home device leaves the factory.
  • Voice control instructions for a wider range of applications can be collected to generate a voice command, and a corresponding operation instruction is generated according to a user's demonstration operation.
  • an embodiment of the present invention provides a method for customizing a voice command.
  • the method may be performed by a device (hereinafter referred to as a device) provided with a voice control module, such as a smart home device or a terminal.
  • a device hereinafter referred to as a device
  • a voice control module such as a smart home device or a terminal.
  • the smart home device may be a smart device. Air conditioners, smart gas stoves, smart TVs or smart refrigerators, etc.
  • the terminals can be devices such as mobile phones or tablet computers (PADs) that can be installed with APPs for controlling smart home devices. The flow of this method is described below.
  • Step 101 Generate a first voice command according to the collected voice information.
  • a new voice control scheme may be created in a smart home device or APP.
  • the smart home device or APP may provide a visual operation interface for the user. The interface prompts step by step to complete the customization process.
  • a new voice command needs to be generated. Then the user can be prompted to enter voice information. After the user enters the voice information, the smart home device or the terminal where the APP is installed can collect the voice information entered by the user.
  • a microphone is generally provided in the smart home device or the terminal on which the APP is installed, so the smart home device or the terminal on which the APP is installed can collect voice information input by the user through the microphone.
  • a first voice command can be generated according to the voice information input by the user.
  • the voice information input by the user may be collected multiple times, and The voice information collected multiple times is analyzed to extract common features of the voice information, and then a first voice command is generated according to the common features.
  • the collected voice information input by the user three times is "turn on the air conditioner", "help me turn on the air conditioner", and "please turn on the air conditioner”, then the common feature of the three input is "turn on the air conditioner", so the user can input
  • This voice of "turn on the air conditioner” is used as a common feature, and it is used as the first voice command.
  • the vector features in the voice information are separately extracted through the voice recognition model, and then the vector features extracted multiple times are compared to obtain common features, and a first voice command is generated based on the common features. .
  • the user may be prompted to input the next voice information. Specifically, it can be prompted through text information, or can also be prompted through voice.
  • Step 102 The first prompt information is output to instruct the user to input a demonstration operation for realizing at least one function of the smart home device.
  • the first prompt information may be output to prompt the user for a demonstration operation.
  • the first prompt information may be output in the form of text information, for example, the words "Please perform a demonstration operation" may be displayed on the display unit; or the first prompt information may also be output by voice, for example, by using The speaker included in the smart home device or the terminal on which the APP is installed outputs "Please perform a demonstration operation"; of course, it can also be output by combining the above two methods.
  • Step 103 Generate a first operation instruction for performing an operation step in the demonstration operation process based on the demonstration operation.
  • the demonstration operation performed by the user is used to demonstrate the operation steps required to implement at least one function of the smart home device.
  • the device collects the demonstration operation, it can obtain the operation steps included in the demonstration operation.
  • a first operation instruction is generated.
  • each operation step corresponds to a function or a function instruction of the smart home device.
  • the user performed operations such as “turning on the air conditioner”, “adjusting the wind intensity”, and “adjusting the wind direction”.
  • the function instruction may also be other possible function instructions, such as the gear adjustment of a smart home device, or the page jump of an APP.
  • steps 101 and 103 can also be converted, that is, step 103 is performed first, and then step 101 is performed, and then step 102 is used to prompt the user to record a voice command.
  • Step 104 Execute the first operation instruction.
  • Step 105 Determine whether the result of executing the first operation instruction meets the requirements of the user.
  • the first operation instruction may be verified to check whether the first operation instruction can implement the function that the user wants to implement. Therefore, the first operation instruction may be executed once, and then After the execution is completed, a prompt message is output to let the user confirm whether it is what the user expected. After the user performs feedback, the first feedback information may be received, and based on the first feedback information, it is determined whether the result of the first operation instruction meets the requirements of the user.
  • Step 106 If the determination result of step 105 is yes, then establish an association relationship between the first voice command and the first operation instruction, and store the association relationship.
  • the first voice command may be associated with the first operation instruction to generate a new voice control scheme, and the new voice The control scheme is stored. In this way, if the user receives the first voice command again during the use of the smart home device, the first operation instruction associated with the first voice command can be found, and then the first operation instruction is executed to achieve the above.
  • the user demonstrates the functions implemented by the operation.
  • the association between the voice command and the operation instruction may be stored in a storage unit included in the smart home device, so that the smart home device can successfully complete voice control even when there is no network, or the voice command and the The association relationship between the operation instructions is stored on the server side. In this way, the smart home device or APP can obtain the operation instructions associated with the voice command input by the user from the server side.
  • the user can send a voice command entered by the user to the server, and after the server matches the associated operation instruction, it sends the operation instruction to the smart home device to implement voice control; or
  • the APP may send a voice command input by the user to the server, the server sends the matching operation instruction to the APP, and the APP sends the operation instruction to the smart home device to implement voice control.
  • Step 107 If the determination result of step 105 is No, the user is prompted to perform the demo operation again.
  • the user may be prompted to perform the demo operation again, and then a new first operation instruction is generated based on the re-done demo operation.
  • the method further includes the following steps:
  • Step 201 Perform semantic recognition on the collected voice information, and output a semantic recognition result.
  • the users are located in different geographical locations and the types of languages used in daily life are different. For example, Shanghainese may be more accustomed to speak Shanghai dialect, and Sichuanese are more accustomed to speak Sichuan dialect, so users are customizing new voice control
  • dialects may also be input, and different dialects may have the same semantics, so the corresponding operation instructions should be the same. Therefore, in the embodiment of the present invention, after receiving the voice information input by the user, the voice information can also be semantically recognized and the result of the semantic recognition can be output to the user, so that the user can confirm whether the semantic recognition result is correct.
  • the semantic recognition results are generally described in a common language, for example, they can be described in Mandarin. Specifically, during the output, the semantic recognition result may be displayed on the display unit in text form, or the semantic recognition result may be played through a speaker voice.
  • Step 202 Determine whether the semantic recognition result is the semantics expressed by the collected voice information.
  • the user may perform feedback based on the output semantic recognition result.
  • the device may receive the second feedback information input by the user, and determine whether the foregoing semantic recognition result is input by the user based on the second feedback information.
  • the semantics of voice information may be performed by the user.
  • Step 203 If the determination result of step 202 is yes, associate the first voice command with the semantic recognition result.
  • the first voice command can be associated with the semantic recognition result, and the same semantics
  • the operation instructions corresponding to all voice commands associated with the recognition result are the same. In this way, even if the voice commands generated according to different dialects are different, as long as the semantic recognition results associated with these voice commands are the same, the corresponding operation instructions are also the same, so that the voice control can support dialects or spoken life at the same time. .
  • Step 204 If the determination result of step 202 is NO, the received modified semantics of the user is associated with the first voice command.
  • the user may be prompted to input the correct semantics.
  • the user may modify the original semantic recognition result or input the correct semantics by himself, and then the device associates the received modified semantics of the user with the first voice command.
  • the association relationship may be applied to the voice control.
  • Figure 3 is a schematic diagram of the process of a user controlling a smart home device by voice.
  • Step 301 Receive a second voice command input by a user.
  • Step 302 Determine whether a voice command matching the second voice command exists in the voice command library.
  • the device can receive the second voice command And matching the second voice command with the voice command library to determine whether a voice command matching the second voice command exists in the voice command library.
  • the second voice command is matched with the voice command database, all features included in the second voice command may be compared with features included in each voice command in the voice command database.
  • Step 303 If the determination result of step 302 is yes, execute the operation instruction associated with the voice command matching the second voice command.
  • the smart home device may directly execute an operation instruction associated with a voice command matching the second voice command; when the device is a terminal where the APP is installed, the voice matching the second voice command may be The command-associated operation instruction is sent to the smart home device, so that the smart home device executes the operation instruction.
  • Step 304 If the determination result in step 302 is no, the user is prompted to re-enter the voice command.
  • the device may prompt the device to input the voice command again.
  • the device may temporarily save the second voice command.
  • the device may also output to the user the semantic recognition result associated with the one or more voice commands, so that the user confirms the voice command that needs to be entered, and after the user selects and confirms, executes the corresponding operation instruction.
  • Step 305 Determine whether a voice command matching the re-entered voice command exists in the voice command library.
  • Step 306 If the determination result of step 305 is yes, update the voice command that successfully matches the re-entered voice command according to the second voice command.
  • the voice command that successfully matches the re-entered voice command can be updated according to the second voice command to strengthen the Voice commands to make it more compatible.
  • the second voice command can be compared with the re-entered voice command, and the common features of the two can be extracted and stored.
  • the voice command input by the user for multiple consecutive times does not match successfully, it indicates that there is no voice control scheme associated with the voice command, and then the second prompt information may be output, and the second prompt information is used for Prompt the user whether to set the operation instruction associated with the voice command.
  • the user feedback needs to set the associated operation instruction, output a first prompt message to prompt the user to perform the operation demonstration, and after the user performs the demonstration operation, the newly generated
  • the operation instructions are associated with common features in voice commands that are input multiple times in succession to obtain a new voice control scheme.
  • the user's voice information can be collected to generate a voice command, and the corresponding operation instruction can be generated according to the user's demonstration operation, and the two are associated.
  • the voice command is generated based on the user's voice information, and then It can be more in line with the user's speaking habits, and the operation instructions obtained based on the user's demonstration operation can be not limited to the voice control instructions when the smart home device leaves the factory, and has a wider scope of application.
  • an embodiment of the present invention provides a voice command customization device, including:
  • a generating unit 401 configured to generate a first voice command according to the collected voice information
  • An output unit 402 configured to output a first prompt message instructing a user to input a demonstration operation for realizing at least one function of the smart home device
  • the generating unit 401 is further configured to generate a first operation instruction for performing an operation step in the demonstration operation process based on the demonstration operation;
  • the association unit 403 is configured to establish an association relationship between the first voice command and the first operation instruction, and store the association relationship, so that when a voice command matching the first voice command is received, the first operation instruction is executed.
  • the generating unit 401 is specifically configured to:
  • a first voice command is generated based on a common feature.
  • association unit 403 is specifically configured to:
  • the first voice command is associated with the first operation instruction and stored, otherwise the user is prompted to perform a demo operation again.
  • the device further includes a semantic recognition unit 404, configured to:
  • the device further includes an update unit 405, configured to:
  • the second voice is recorded command
  • the voice command that matches the re-entered voice command is updated according to the second voice command.
  • the output unit 402 is further configured to:
  • Output a second prompt message for prompting the user whether to set an associated operation instruction when the voice commands received multiple times consecutively fail to match any of the voice commands in the voice command library;
  • the first prompt information is output.
  • the device can be used to execute the method provided in the embodiment shown in FIGS. 1-3. Therefore, for the functions that can be implemented by the functional modules of the device, please refer to the description of the embodiment shown in FIG. . Among them, although the semantic recognition unit 404 and the update unit 405 are shown together in FIG. 4, they are not mandatory function units, and therefore are shown by dashed lines.
  • an embodiment of the present invention provides a voice command customization device, including at least one processor 501.
  • the at least one processor 501 is configured to execute the computer program stored in the memory when executed The steps of the voice command customization method provided by the illustrated embodiment.
  • the at least one processor 501 may specifically include a central processing unit (CPU), an application-specific integrated circuit (ASIC), may be one or more integrated circuits for controlling program execution, and may be used
  • a hardware circuit developed by a field programmable gate array (FPGA) can be a baseband processor.
  • the at least one processor 501 may include at least one processing core.
  • the device further includes a memory 502, and the memory 502 may include a read-only memory (ROM), a random access memory (RAM), and a disk memory.
  • the memory 502 is configured to store data required when the at least one processor 501 runs.
  • the number of the memories 502 is one or more.
  • the memory 502 is shown together in FIG. 5, but it needs to be known that the memory 502 is not a required functional module, and therefore is shown by a dotted line in FIG. 5.
  • an embodiment of the present invention provides a computer-readable storage medium.
  • the computer-readable storage medium stores computer instructions.
  • the computer instructions When the computer instructions are run on a computer, the computer executes the operations shown in FIG. 1-3. Shown method.
  • the computer-readable storage medium includes: a universal serial bus flash disk (Universal Serial Bus flash drive (USB), a mobile hard disk, a read-only memory (ROM), and a random access memory (ROM) Random Access Memory (RAM), magnetic disks or compact discs, and other storage media that can store program code.
  • USB Universal Serial Bus flash drive
  • ROM read-only memory
  • RAM Random Access Memory
  • magnetic disks or compact discs and other storage media that can store program code.
  • the disclosed device and method may be implemented in other manners.
  • the device embodiments described above are only schematic.
  • the division of the unit or unit is only a logical function division.
  • the combination can either be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be electrical or other forms.
  • Each functional unit in the embodiment of the present invention may be integrated into one processing unit, or each unit may also be an independent physical module.
  • the integrated unit When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, all or part of the technical solutions of the embodiments of the present invention may be embodied in the form of a software product.
  • the computer software product is stored in a storage medium and includes several instructions for making a computer device, for example, it may be A personal computer, a server, or a network device, or a processor executes all or part of the steps of the method described in each embodiment of the present invention.
  • the foregoing storage medium includes: a universal serial bus flash drive (universal serial bus flash drive), a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disc, and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种语音命令定制方法、装置和设备及计算机存储介质,用于实现语音控制方案的定制,提升用户使用体验。方法包括:根据采集的语音信息生成第一语音命令(101);输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作(102),并基于演示操作生成用于执行演示操作过程中的操作步骤的第一操作指令(103);建立第一语音命令与第一操作指令之间的关联关系,并存储关联关系(106),以使得在接收到与第一语音命令相匹配的语音命令时,执行第一操作指令。

Description

语音命令定制方法、装置和设备及计算机存储介质
本申请要求于2018年8月6日提交中国专利局、申请号为201810887444.X、发明名称为“语音命令定制方法、装置和设备及计算机存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明涉及计算机技术领域,特别涉及一种语音命令定制方法、装置和设备及计算机存储介质。
背景技术
目前,智能家居设备给人们的生活提供了更多的便利,对于智能家居设备的控制方法也愈来愈多,例如,用户可以通过语音对智能家居设备进行控制,或者通过终端上安装的应用程序(Application,APP)对智能家居设备进行控制。其中,对于通过语音进行控制的方式,通常是在智能家居设备出厂时就预先定义好语音库与功能的关联关系,用户可以根据说明书或者提示进行相应的语音控制。
因此,现有的语音控制方案中语音和功能指令集都是预先定义好的,但是不同的用户的使用习惯存在一定的差异,预先定义好的语音控制方案可能并不符合用户的使用习惯,这样,用户在通过语音控制智能家居设备时的使用体验显然不佳,并且,预先定义的语音控制方案中包括的功能有限,可能并不能满足用户的使用需求,进一步降低了用户的使用体验。
发明内容
本发明实施例提供一种语音命令定制方法、装置和设备及计算机存储介质,用于实现语音控制方案的定制,提升用户使用体验。
第一方面,提供一种语音命令定制方法,该方法包括:
根据采集的语音信息生成第一语音命令;
输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作,并基于所述演示操作生成用于执行所述演示操作过程中的操作步骤的第一操作指令;
建立所述第一语音命令与所述第一操作指令之间的关联关系,并存储所述关联关系,以使得在接收到与所述第一语音命令相匹配的语音命令时,执行所述第一操作指令。
本发明实施例中,可以采集用户的语音信息生成语音命令,并根据用户的演示操作生成相应的操作指令,并将这两者进行关联,这样,语音命令是根据用户的语音信息来生成的,进而能够更加符合用户的说话习惯,且基于用户演示操作得到的操作指令可以不局限于智能家居设备出厂时的语音控制指令,适用范围更广。
可选的,所述根据采集的语音信息生成第一语音命令,包括:
根据采集到的多条语音信息进行分析,以提取所述多条语音信息的共同特征;
基于所述共同特征生成所述第一语音命令。
可选的,所述建立所述第一语音命令与所述第一操作指令之间的关联关系,包括:
执行所述第一操作指令;
基于接收到的第一反馈信息,确定执行所述第一操作指令的结果是否符合用户的要求;
在所述第一反馈信息表明执行所述第一操作指令的结果符合所述用户的要求时,将所述第一语音命令与所述第一操作指令关联后进行存储,否则提示所述用户再次进行演示操作。
本发明实施例中,在生成操作指令之后,执行一次操作指令,进而对该操作指令进行校验,以确认操作指令实现的功能是否符合用户的预期效果。
可选的,所述方法还包括:
对采集到的语音信息进行语义识别,并输出语义识别结果;
基于接收到的第二反馈信息,确定所述语义识别结果是否为所述采集到的语音信息所表达的语义;
若确定结果为是,则将所述第一语音命令与所述语义识别结果进行关联,否则,将接收到的用户修改后的语义与所述第一语音命令进行关联,其中,与同一语义识别结果关联的所有语音命令所对应的操作指令相同。
本发明实施例中,可以对用户输入的语音信息进行语义识别,并将根据该语音信息生成的语音命令与语义识别结果进行关联,这样,与同一语义识别结果关联的所有指令就可以实现相同的功能,这样,即使用户输入方言或者其他语种,都可以进行支持,提高语音识别的泛化能力。
可选的,所述方法还包括:
在接收到的第二语音命令未与已存储的语音命令库中的任何语音命令匹配成功,但与所述语音命令库中至少一条语音命令的相似度不小于第一预设阈值时,记录所述第二语音命令;
提示用户重新进行语音命令的输入;
若重新输入的语音命令匹配成功,则根据所述第二语音命令更新与所述重新输入的语音命令匹配成功的语音命令。
本发明实施例中,若是用户初次输入的语音命令未被识别,而再次输入的语音命令识别成功,那么就可以根据初次输入的语音命令对语音命令库进行更新,以提升语音命令库的识别能力。
可选的,所述方法还包括:
在连续多次接收到的语音命令均未与语音命令库中的任何语音命令匹配成功时,输出用于提示用户是否设置相关联的操作指令的第二提示信息;
在接收到的第三反馈信息表明需要设置与所述第四语音命令关联的操作指令时,输出所述第一提示信息。
第二方面,提供一种语音命令定制装置,包括:
生成单元,用于根据采集的语音信息生成第一语音命令;
输出单元,用于输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作;
所述生成单元,还用于基于所述演示操作生成用于执行所述演示操作过程中的操作步骤的第一操作指令;
关联单元,用于建立所述第一语音命令与所述第一操作指令之间的关联关系,并存储所述关联关系,以使得在接收到与所述第一语音命令相匹配的语音命令时,执行所述第一操作指令。
可选的,所述生成单元具体用于:
根据采集到的多条语音信息进行分析,以提取所述多条语音信息的共同特征;
基于所述共同特征生成所述第一语音命令。
可选的,所述关联单元具体用于:
执行所述第一操作指令;
基于接收到的第一反馈信息,确定执行所述第一操作指令的结果是否符合用户的要求;
在所述第一反馈信息表明执行所述第一操作指令的结果符合所述用户的要求时,将所述第一语音命令与所述第一操作指令关联后进行存储,否则提示所述用户再次进行演示操作。
可选的,所述装置还包括语义识别单元,用于:
对采集到的语音信息进行语义识别,并输出语义识别结果;
基于接收到的第二反馈信息,确定所述语义识别结果是否为所述采集到的语音信息所表达的语义;
若确定结果为是,则将所述第一语音命令与所述语义识别结果进行关联,否则,将接收到的用户修改后的语义与所述第一语音命令进行关联,其中,与 同一语义识别结果关联的所有语音命令所对应的操作指令相同。
可选的,所述装置还包括更新单元,用于:
在接收到的第二语音命令未与已存储的语音命令库中的任何语音命令匹配成功,但与所述语音命令库中至少一条语音命令的相似度不小于第一预设阈值时,记录所述第二语音命令;
提示用户重新进行语音命令的输入;
若重新输入的语音命令匹配成功,则根据所述第二语音命令更新与所述重新输入的语音命令匹配成功的语音命令。
可选的,所述输出单元还用于:
在连续多次接收到的语音命令均未与语音命令库中的任何语音命令匹配成功时,输出用于提示用户是否设置相关联的操作指令的第二提示信息;
在接收到的第三反馈信息表明需要设置与所述第四语音命令关联的操作指令时,输出所述第一提示信息。
第三方面,提供一种语音命令定制设备,包括:
至少一个处理器;以及
与所述至少一个处理器通信连接的存储器;其中,
所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行如第一方面所述的方法。
第四方面,提供一种计算机存储介质,
所述计算机存储介质存储有计算机指令,当所述计算机指令在计算机上运行时,使得计算机执行如第一方面所述的方法。
附图说明
图1为本发明实施例提供的语音命令定制方法的流程示意图;
图2为本发明实施例提供的将语义识别结果与语音命令进行关联的流程示意图;
图3为本发明实施例提供的语音控制过程的流程示意图;
图4为本发明实施例提供的语音命令定制装置的一种结构示意图;
图5为本发明实施例提供的语音命令定制设备的一种结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述。
下面介绍本发明实施例的技术背景。
目前,语音控制方案中语音和功能指令集都是预先定义好的,但是不同的用户的使用习惯存在一定的差异,预先定义好的语音控制方案可能并不符合用户的使用习惯,这样,用户在通过语音控制智能家居设备时的使用体验显然不佳,并且,预先定义的语音控制方案中包括的功能有限,可能并不能满足用户的使用需求,进一步降低了用户的使用体验。
鉴于此,本发明实施例提供一种语音命令定制方法、装置和设备及计算机存储介质,在该方法中,可以采集用户的语音信息生成语音命令,并根据用户的演示操作生成相应的操作指令,并将这两者进行关联,这样,语音命令是根据用户的语音信息来生成的,进而能够更加符合用户的说话习惯,且基于用户演示操作得到的操作指令可以不局限于智能家居设备出厂时的语音控制指令,适用范围更广。
下面结合附图介绍本发明实施例提供的技术方案。
请参见图1,本发明实施例提供一种语音命令定制方法,该方法可以通过智能家居设备或者终端等设置有语音控制模块的设备(后续简称设备)来执行,例如,智能家居设备可以是智能空调、智能燃气灶、智能电视或者智能冰箱等等,终端可以是手机或者平板电脑(PAD)等可以安装用于控制智能家居设备的APP的设备。该方法的流程描述如下。
步骤101:根据采集的语音信息生成第一语音命令。
本发明实施例中,在用户想要定制新的语音控制方案时,可以在智能家居 设备或者APP中新建新的语音控制方案,智能家居设备或者APP可以提供给用户可视化的操作界面,用户根据操作界面上的提示一步步进行操作,以完成定制过程。
在定制新的语音控制方案时,需要生成新的语音命令,那么可以提示用户输入语音信息,在用户输入语音信息之后,智能家居设备或者安装APP的终端就可以采集到用户输入的语音信息。具体的,智能家居设备或者安装APP的终端中一般设置有麦克风,因此智能家居设备或者安装APP的终端可以通过麦克风采集用户输入的语音信息。
其中,在采集到用户输入的语音信息之后,就可以根据用户输入的语音信息生成第一语音命令。
本发明实施例中,由于用户每一次说话时的语气或者内容或多或少都可能有所不同,因此为了提升生成的语音命令的识别率,还可以多次采集用户输入的语音信息,并对多次采集的语音信息进行分析,以提取出这些语音信息的共同特征,进而根据共同特征来生成第一语音命令。例如,采集到的用户3次输入的语音信息分别为“打开空调”“帮我打开空调”“请打开空调”,那么3次输入的共同特征即为“打开空调”,因而就可以将用户输入的“打开空调”的这一段语音作为共同特征,并将其作为第一语音命令。或者,在采集到用户的语音信息之后,通过语音识别模型分别提取语音信息中的矢量特征,进而将多次提取的矢量特征进行对比,获取共同特征,并基于该共同特征来生成第一语音命令。
在实际应用中,在采集到一次用户输入的语音信息之后,则可以提示用户进行下一次语音信息的输入。具体的,可以通过文字信息进行提示,或者还可以通过语音进行提示。
步骤102:输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作。
本发明实施例中,在生成第一语音命令之后,则可以输出第一提示信息, 以提示用户进行演示操作。其中,第一提示信息可以通过以文字信息的形式进行输出,例如,可以在显示单元上显示“请进行演示操作”字样;或者,第一提示信息还可以通过语音方式进行输出,例如,可以通过智能家居设备或者安装APP的终端中包括的扬声器输出“请进行演示操作”;当然,也可以通过上述两种方式结合进行输出。
步骤103:基于演示操作生成用于执行演示操作过程中的操作步骤的第一操作指令。
本发明实施例中,用户进行的演示操作用于演示实现智能家居设备的至少一个功能所需的操作步骤,这样,设备在采集到该演示操作之后,则可以获取该演示操作中包括的操作步骤以及操作步骤之间的顺序关系,进而基于这些操作步骤以及操作步骤之间的顺序关系来生成第一操作指令。
其中,对于设备来讲,每一个操作步骤即对应着智能家居设备的一个功能或者一个功能指令。例如,用户在演示操作中,依次执行了“打开空调”“调节出风强度”“调节风向”等操作,那么对于设备来讲,则可以获取这几个操作对应的功能指令以及功能指令间的顺序关系,那么基于此生成的第一操作指令则会依次执行上述操作。当然,功能指令还可以是其他可能的功能指令,例如智能家居设备的档位调节,或者APP的页面跳转等。
在具体实施过程中,步骤101和步骤103的顺序也可以进行转换,也就是先执行步骤103,然后再执行步骤101,那么步骤102即用于提示用户录制语音命令。
步骤104:执行第一操作指令。
步骤105:确定执行第一操作指令的结果是否符合用户的要求。
本发明实施例中,在生成第一操作指令之后,可以对第一操作指令进行校验,以检验第一操作指令是否能够实现用户想要实现的功能,因此可以执行一次第一操作指令,然后在执行完成之后,输出提示信息,让用户确认是否到该用户的预期。在用户进行反馈之后,则可以接收到第一反馈信息,并基于第一 反馈信息确定第一操作指令的结果是否符合用户的要求。
步骤106:若步骤105的确定结果为是,则建立第一语音命令与第一操作指令之间的关联关系,并存储关联关系。
本发明实施例中,若是第一反馈信息表明执行第一操作指令的结果符合要求,则可以将第一语音命令与第一操作指令进行关联,以生成新的语音控制方案,并对新的语音控制方案进行存储,这样,在智能家居设备的使用过程中,若是再次接收到用户输入第一语音命令,则可以查找到与之关联的第一操作指令,进而执行第一操作指令,来实现上述用户演示操作所实现的功能。
具体的,可以将语音命令与操作指令之间的关联关系保存在智能家居设备包括的存储单元,这样,即使没有网络时,智能家居设备也能够顺利完成语音控制,或者,还可以将语音命令与操作指令之间的关联关系保存在服务器端,这样,智能家居设备或者APP可以从服务器端来获取与用户输入的语音命令相关联的操作指令。例如,用户通过APP控制智能家居设备时,则可以将用户输入的语音命令发送给服务器端,服务器端匹配到相关联的操作指令后,将操作指令发送给智能家居设备,从而实现语音控制;或者,也可以是APP将用户输入的语音命令发送给服务器端,服务器端将匹配到的操作指令反馈给APP,APP再将操作指令发送给智能家居设备,实现语音控制。
步骤107:若步骤105的确定结果为否,则提示用户再次进行演示操作。
本发明实施例中,若是第一反馈信息表明执行第一操作指令的结果不符合要求,那么就可以提示用户重新进行演示操作,再基于重新进行的演示操作生成新的第一操作指令。
请参见图2,在接收到用户输入的语音信息之后,该方法还包括如下步骤:
步骤201:对采集到的语音信息进行语义识别,并输出语义识别结果。
本发明实施例中,用户所在的地理位置不同,在日常生活中所习惯的语言类型不同,例如上海人可能更习惯说上海话,四川人更习惯说四川话,因而用户在定制新的语音控制方案时,也有可能会输入方言,而不同的方言有可能对 应的语义是相同的,那么所对应的操作指令就理应相同。因此,本发明实施例中,在接收到用户输入的语音信息之后,还可以对语音信息进行语义识别,并将语义识别结果输出给用户,这样就可以让用户确认语义识别结果是否正确。一般来讲,语义识别结果一般采用通用语言进行描述,例如可以采用普通话进行描述。具体的,在进行输出时,可以将语义识别结果以文字形式显示在显示单元上,或者通过扬声器语音播放语义识别结果。
步骤202:确定语义识别结果是否为采集到的语音信息所表达的语义。
本发明实施例中,用户可以基于输出的语义识别结果进行反馈,相对应的,设备就可以接收到用户输入的第二反馈信息,并基于第二反馈信息确定上述语义识别结果是否为用户输入的语音信息所表达的语义。
步骤203:若步骤202的确定结果为是,则将第一语音命令与语义识别结果进行关联。
本发明实施例中,若是语义识别结果为用户输入的语音信息所表达的语义,也就是说语义识别结果是正确的,那么就可以将第一语音命令与语义识别结果进行关联,并且,同一语义识别结果关联的所有语音命令所对应的操作指令相同。这样,即使根据不同的方言生成的语音命令不同,但是只要这些语音命令关联的语义识别结果是相同的,那么所对应的操作指令也是相同的,进而使得语音控制能够同时支持方言,或者生活口语等。
步骤204:若步骤202的确定结果为否,则接收到的用户修改后的语义与所述第一语音命令进行关联。
本发明实施例中,若是语义识别结果并不是用户输入的语音信息所表达的语义,那么就可以提示用户输入正确的语义。用户可以在原有语义识别结果的基础上进行修改,或者自行输入正确的语义,然后设备将接收到的用户修改后的语义与所述第一语音命令进行关联。
本发明实施例中,在存储了语音命令与操作指令之间的关联关系之后,则可以将该关联关系应用与语音控制。请参见图3,为用户通过语音控制智能家 居设备的流程示意图。
步骤301:接收用户输入的第二语音命令。
步骤302:确定语音命令库中是否存在与第二语音命令匹配的语音命令。
在用户想要通过语音对智能家居设备进行控制时,例如用户想要实现的功能需要输入第二语音命令,那么用户就可以输入第二语音命令,相应的,设备就可以接收到第二语音命令,并将第二语音命令与语音命令库进行匹配,以确定语音命令库中是否存在与第二语音命令匹配的语音命令。具体的,在将第二语音命令与语音命令库进行匹配时,可以将第二语音命令包括的所有特征与语音命令库中每一条语音命令包括的特征进行比对。
步骤303:若步骤302的确定结果为是,则执行与第二语音命令匹配的语音命令关联的操作指令。
具体的,当设备为智能家居设备时,智能家居设备可以直接执行与第二语音命令匹配的语音命令关联的操作指令;当设备为安装APP的终端时,可以将与第二语音命令匹配的语音命令关联的操作指令发送给智能家居设备,以使得智能家居设备执行该操作指令。
步骤304:若步骤302的确定结果为否,则提示用户重新进行语音命令的输入。
本发明实施例中,若是用户输入的第二语音命令与语音命令库匹配失败,也就是第二语音命令未被识别成功,则设备可以提示设备重新进行语音命令的输入。
具体的,随着时间的推移,用户可能对定制语音控制方案时输入的语音命令有所遗忘,但是能记得大概内容,因而第二语音命令可能未能够与语音命令库中的任何语音命令匹配成功,但是第二语音命令与语音命令库中的其中一条或者多条语音命令的相似度不小于第一预设阈值,那么设备可以将第二语音命令暂时保存。当然,设备还可以将上述一条或者多条语音命令关联的语义识别结果输出给用户,让用户确认需要输入的语音命令,在用户进行选择并确认后, 执行相对应的操作指令。
步骤305:确定语音命令库中是否存在与重新输入的语音命令匹配的语音命令。
步骤306:若步骤305的确定结果为是,则根据第二语音命令更新与重新输入的语音命令匹配成功的语音命令。
本发明实施例中,若是用户重新输入的语音命令匹配成功了,并且初次输入的第二语音命令与语音命令库中的其中一条或者多条语音命令的相似度不小于第一预设阈值,那么表明用户初次输入的第二语音命令与重新输入的语音命令所表达的语义应是相同的,那么就可以根据第二语音命令对与重新输入的语音命令匹配成功的语音命令进行更新,以强化该语音命令,使其能够兼容性更强。具体的,可以将第二语音命令与重新输入的语音命令匹配成功的语音命令进行比对,提取这两者的共同特征并进行存储。
本发明实施例中,若是用户连续多次输入的语音命令均未匹配成功,则表明并未与该语音命令相关联的语音控制方案,那么就可以输出第二提示信息,第二提示信息用于提示用户是否设置与该语音命令关联的操作指令,在用户反馈需要设置相关联的操作指令时,输出第一提示信息,以提示用户进行操作演示,并在用户进行演示操作之后,将新生成的操作指令与根据连续多次输入的语音命令中的共同特征进行关联,以得到新的语音控制方案。
综上所述,可以采集用户的语音信息生成语音命令,并根据用户的演示操作生成相应的操作指令,并将这两者进行关联,这样,语音命令是根据用户的语音信息来生成的,进而能够更加符合用户的说话习惯,且基于用户演示操作得到的操作指令可以不局限于智能家居设备出厂时的语音控制指令,适用范围更广。
请参见图4,基于同一发明构思,本发明实施例提供一种语音命令定制装置,包括:
生成单元401,用于根据采集的语音信息生成第一语音命令;
输出单元402,用于输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作;
生成单元401,还用于基于演示操作生成用于执行演示操作过程中的操作步骤的第一操作指令;
关联单元403,用于建立第一语音命令与第一操作指令之间的关联关系,并存储关联关系,以使得在接收到与第一语音命令相匹配的语音命令时,执行第一操作指令。
可选的,生成单元401具体用于:
根据采集到的多条语音信息进行分析,以提取多条语音信息的共同特征;
基于共同特征生成第一语音命令。
可选的,关联单元403具体用于:
执行第一操作指令;
基于接收到的第一反馈信息,确定执行第一操作指令的结果是否符合用户的要求;
在第一反馈信息表明执行第一操作指令的结果符合用户的要求时,将第一语音命令与第一操作指令关联后进行存储,否则提示用户再次进行演示操作。
可选的,该装置还包括语义识别单元404,用于:
对采集到的语音信息进行语义识别,并输出语义识别结果;
基于接收到的第二反馈信息,确定语义识别结果是否为采集到的语音信息所表达的语义;
若确定结果为是,则将第一语音命令与语义识别结果进行关联,否则,将接收到的用户修改后的语义与第一语音命令进行关联,其中,与同一语义识别结果关联的所有语音命令所对应的操作指令相同。
可选的,该装置还包括更新单元405,用于:
在接收到的第二语音命令未与已存储的语音命令库中的任何语音命令匹配成功,但与语音命令库中至少一条语音命令的相似度不小于第一预设阈值时, 记录第二语音命令;
提示用户重新进行语音命令的输入;
若重新输入的语音命令匹配成功,则根据第二语音命令更新与重新输入的语音命令匹配成功的语音命令。
可选的,输出单元402还用于:
在连续多次接收到的语音命令均未与语音命令库中的任何语音命令匹配成功时,输出用于提示用户是否设置相关联的操作指令的第二提示信息;
在接收到的第三反馈信息表明需要设置与第四语音命令关联的操作指令时,输出第一提示信息。
该设备可以用于执行图1-3所示的实施例所提供的方法,因此,对于该设备的各功能模块所能够实现的功能等可参考图1所示的实施例的描述,不多赘述。其中,语义识别单元404和更新单元405虽然在图4中一并示出,但是并非是必选的功能单元,因此以虚线示出。
请参见图5,基于同一发明构思,本发明实施例提供一种语音命令定制设备,包括至少一个处理器501,至少一个处理器501用于执行存储器中存储的计算机程序时实现图1-3所示的实施例提供的语音命令定制方法的步骤。
可选的,至少一个处理器501具体可以包括中央处理器(CPU)、特定应用集成电路(application specific integrated circuit,ASIC),可以是一个或多个用于控制程序执行的集成电路,可以是使用现场可编程门阵列(field programmable gate array,FPGA)开发的硬件电路,可以是基带处理器。
可选的,至少一个处理器501可以包括至少一个处理核心。
可选的,该设备还包括存储器502,存储器502可以包括只读存储器(read only memory,ROM)、随机存取存储器(random access memory,RAM)和磁盘存储器。存储器502用于存储至少一个处理器501运行时所需的数据。存储器502的数量为一个或多个。其中,存储器502在图5中一并示出,但需要知道的是存储器502不是必选的功能模块,因此在图5中以虚线示出。
基于同一发明构思,本发明实施例提供一种计算机可读存储介质,所述计算机可读存储介质存储有计算机指令,当所述计算机指令在计算机上运行时,使得计算机执行如图1-3所示的方法。
在具体的实施过程中,计算机可读存储介质包括:通用串行总线闪存盘(Universal Serial Bus flash drive,USB)、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的存储介质。
在本发明实施例中,应该理解到,所揭露的设备和方法,可以通过其它的方式实现。例如,以上所描述的设备实施例仅仅是示意性的,例如,所述单元或单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个***,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,设备或单元的间接耦合或通信连接,可以是电性或其它的形式。
在本发明实施例中的各功能单元可以集成在一个处理单元中,或者各个单元也可以均是独立的物理模块。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明实施例的技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备,例如可以是个人计算机,服务器,或者网络设备等,或处理器(processor)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:通用串行总线闪存盘(universal serial bus flash drive)、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,以上实施例仅用以对本申请的技术方案进行了详细介绍,但以上实施例的说明只是用于帮助理解本发明实施例的方法,不应理解为对本发明 实施例的限制。本技术领域的技术人员可轻易想到的变化或替换,都应涵盖在本发明实施例的保护范围之内。

Claims (14)

  1. 一种语音命令定制方法,其特征在于,包括:
    根据采集的语音信息生成第一语音命令;
    输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作,并基于所述演示操作生成用于执行所述演示操作过程中的操作步骤的第一操作指令;
    建立所述第一语音命令与所述第一操作指令之间的关联关系,并存储所述关联关系,以使得在接收到与所述第一语音命令相匹配的语音命令时,执行所述第一操作指令。
  2. 如权利要求1所述的方法,其特征在于,所述根据采集的语音信息生成第一语音命令,包括:
    根据采集到的多条语音信息进行分析,以提取所述多条语音信息的共同特征;
    基于所述共同特征生成所述第一语音命令。
  3. 如权利要求1所述的方法,其特征在于,所述建立所述第一语音命令与所述第一操作指令之间的关联关系,包括:
    执行所述第一操作指令;
    基于接收到的第一反馈信息,确定执行所述第一操作指令的结果是否符合用户的要求;
    在所述第一反馈信息表明执行所述第一操作指令的结果符合所述用户的要求时,将所述第一语音命令与所述第一操作指令关联后进行存储,否则提示所述用户再次进行演示操作。
  4. 如权利要求1所述的方法,其特征在于,所述方法还包括:
    对采集到的语音信息进行语义识别,并输出语义识别结果;
    基于接收到的第二反馈信息,确定所述语义识别结果是否为所述采集到的语音信息所表达的语义;
    若确定结果为是,则将所述第一语音命令与所述语义识别结果进行关联,否则,将接收到的用户修改后的语义与所述第一语音命令进行关联,其中,与同一语义识别结果关联的所有语音命令所对应的操作指令相同。
  5. 如权利要求1-4任一所述的方法,其特征在于,所述方法还包括:
    在接收到的第二语音命令未与已存储的语音命令库中的任何语音命令匹配成功,但与所述语音命令库中至少一条语音命令的相似度不小于第一预设阈值时,记录所述第二语音命令;
    提示用户重新进行语音命令的输入;
    若重新输入的语音命令匹配成功,则根据所述第二语音命令更新与所述重新输入的语音命令匹配成功的语音命令。
  6. 如权利要求1-4任一所述的方法,其特征在于,所述方法还包括:
    在连续多次接收到的语音命令均未与语音命令库中的任何语音命令匹配成功时,输出用于提示用户是否设置相关联的操作指令的第二提示信息;
    在接收到的第三反馈信息表明需要设置与所述第四语音命令关联的操作指令时,输出所述第一提示信息。
  7. 一种语音命令定制装置,其特征在于,包括:
    生成单元,用于根据采集的语音信息生成第一语音命令;
    输出单元,用于输出第一提示信息指示用户输入实现智能家居设备的至少一个功能的演示操作;
    所述生成单元,还用于基于所述演示操作生成用于执行所述演示操作过程中的操作步骤的第一操作指令;
    关联单元,用于建立所述第一语音命令与所述第一操作指令之间的关联关系,并存储所述关联关系,以使得在接收到与所述第一语音命令相匹配的语音命令时,执行所述第一操作指令。
  8. 如权利要求7所述的装置,其特征在于,所述生成单元具体用于:
    根据采集到的多条语音信息进行分析,以提取所述多条语音信息的共同特 征;
    基于所述共同特征生成所述第一语音命令。
  9. 如权利要求7所述的装置,其特征在于,所述关联单元具体用于:
    执行所述第一操作指令;
    基于接收到的第一反馈信息,确定执行所述第一操作指令的结果是否符合用户的要求;
    在所述第一反馈信息表明执行所述第一操作指令的结果符合所述用户的要求时,将所述第一语音命令与所述第一操作指令关联后进行存储,否则提示所述用户再次进行演示操作。
  10. 如权利要求7所述的装置,其特征在于,所述装置还包括语义识别单元,用于:
    对采集到的语音信息进行语义识别,并输出语义识别结果;
    基于接收到的第二反馈信息,确定所述语义识别结果是否为所述采集到的语音信息所表达的语义;
    若确定结果为是,则将所述第一语音命令与所述语义识别结果进行关联,否则,将接收到的用户修改后的语义与所述第一语音命令进行关联,其中,与同一语义识别结果关联的所有语音命令所对应的操作指令相同。
  11. 如权利要求7-10任一所述的装置,其特征在于,所述装置还包括更新单元,用于:
    在接收到的第二语音命令未与已存储的语音命令库中的任何语音命令匹配成功,但与所述语音命令库中至少一条语音命令的相似度不小于第一预设阈值时,记录所述第二语音命令;
    提示用户重新进行语音命令的输入;
    若重新输入的语音命令匹配成功,则根据所述第二语音命令更新与所述重新输入的语音命令匹配成功的语音命令。
  12. 如权利要求7-10任一所述的装置,其特征在于,所述输出单元还用 于:
    在连续多次接收到的语音命令均未与语音命令库中的任何语音命令匹配成功时,输出用于提示用户是否设置相关联的操作指令的第二提示信息;
    在接收到的第三反馈信息表明需要设置与所述第四语音命令关联的操作指令时,输出所述第一提示信息。
  13. 一种语音命令定制设备,其特征在于,包括:
    至少一个处理器;以及
    与所述至少一个处理器通信连接的存储器;其中,
    所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行如权利要求1~6任一权利要求所述的方法。
  14. 一种计算机存储介质,其特征在于:
    所述计算机存储介质存储有计算机指令,当所述计算机指令在计算机上运行时,使得计算机执行如权利要求1-6中任一项所述的方法。
PCT/CN2018/121040 2018-08-06 2018-12-14 语音命令定制方法、装置和设备及计算机存储介质 WO2020029500A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810887444.X 2018-08-06
CN201810887444.XA CN108831469B (zh) 2018-08-06 2018-08-06 语音命令定制方法、装置和设备及计算机存储介质

Publications (1)

Publication Number Publication Date
WO2020029500A1 true WO2020029500A1 (zh) 2020-02-13

Family

ID=64153673

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/121040 WO2020029500A1 (zh) 2018-08-06 2018-12-14 语音命令定制方法、装置和设备及计算机存储介质

Country Status (2)

Country Link
CN (1) CN108831469B (zh)
WO (1) WO2020029500A1 (zh)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108831469B (zh) * 2018-08-06 2021-02-12 珠海格力电器股份有限公司 语音命令定制方法、装置和设备及计算机存储介质
CN109584875A (zh) * 2018-12-24 2019-04-05 珠海格力电器股份有限公司 一种语音设备控制方法、装置、存储介质及语音设备
CN109901707A (zh) * 2018-12-27 2019-06-18 安徽语讯科技有限公司 一种配置到***内的学习型***操作模块
CN109871119A (zh) * 2018-12-27 2019-06-11 安徽语讯科技有限公司 一种学习型智能语音操作方法和***
US11170774B2 (en) * 2019-05-21 2021-11-09 Qualcomm Incorproated Virtual assistant device
CN110570867A (zh) * 2019-09-12 2019-12-13 安信通科技(澳门)有限公司 一种本地新增语料的语音处理方法及***
CN110580904A (zh) * 2019-09-29 2019-12-17 百度在线网络技术(北京)有限公司 通过语音控制小程序的方法、装置、电子设备及存储介质
CN110784384B (zh) * 2019-10-16 2021-11-02 杭州九阳小家电有限公司 一种家电语音技能的生成方法及智能家电
CN111785265A (zh) * 2019-11-26 2020-10-16 北京沃东天骏信息技术有限公司 智能音箱设置方法和装置、控制方法和装置、智能音箱
CN111063353B (zh) * 2019-12-31 2022-11-11 思必驰科技股份有限公司 允许自定义语音交互内容的客户端处理方法及用户终端
CN111261158A (zh) * 2020-01-15 2020-06-09 上海思依暄机器人科技股份有限公司 一种功能菜单定制方法、语音快捷控制方法和机器人
CN113160807A (zh) * 2020-01-22 2021-07-23 广州汽车集团股份有限公司 一种语料库更新方法及其***、语音控制设备
CN111179933A (zh) * 2020-01-23 2020-05-19 珠海荣邦电子科技有限公司 一种语音控制方法、装置及智能终端
CN114067792B (zh) * 2020-08-07 2024-06-14 北京猎户星空科技有限公司 一种智能设备的控制方法及装置
CN114246450B (zh) * 2020-09-21 2024-02-06 佛山市顺德区美的电热电器制造有限公司 信息处理方法、装置、烹饪设备及计算机可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030110040A1 (en) * 2001-12-07 2003-06-12 Creative Logic Solutions Inc. System and method for dynamically changing software programs by voice commands
CN103646646A (zh) * 2013-11-27 2014-03-19 联想(北京)有限公司 一种语音控制方法及电子设备
CN103713905A (zh) * 2013-12-29 2014-04-09 广州视源电子科技股份有限公司 一种操作步骤自定义方法、装置及***
CN106484270A (zh) * 2016-09-12 2017-03-08 深圳市金立通信设备有限公司 一种语音操作事件添加方法及终端
CN108831469A (zh) * 2018-08-06 2018-11-16 珠海格力电器股份有限公司 语音命令定制方法、装置和设备及计算机存储介质

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE60128816T2 (de) * 2000-09-08 2008-02-07 Koninklijke Philips Electronics N.V. Spracherkennungsverfahren mit ersetzungsbefehl
CN101937693B (zh) * 2010-08-17 2012-04-04 深圳市子栋科技有限公司 基于语音命令的视音频播放方法及***
CN102842306B (zh) * 2012-08-31 2016-05-04 深圳Tcl新技术有限公司 语音控制方法及装置、语音响应方法及装置
CN105845136A (zh) * 2015-01-13 2016-08-10 中兴通讯股份有限公司 语音控制方法、装置及终端
CN105989841B (zh) * 2015-02-17 2019-12-27 上海汽车集团股份有限公司 车载语音控制方法及装置
CN105931637A (zh) * 2016-04-01 2016-09-07 金陵科技学院 一种可自定义指令识别的语音拍照***
CN108174030B (zh) * 2017-12-26 2020-11-17 努比亚技术有限公司 定制化语音控制的实现方法、移动终端及可读存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030110040A1 (en) * 2001-12-07 2003-06-12 Creative Logic Solutions Inc. System and method for dynamically changing software programs by voice commands
CN103646646A (zh) * 2013-11-27 2014-03-19 联想(北京)有限公司 一种语音控制方法及电子设备
CN103713905A (zh) * 2013-12-29 2014-04-09 广州视源电子科技股份有限公司 一种操作步骤自定义方法、装置及***
CN106484270A (zh) * 2016-09-12 2017-03-08 深圳市金立通信设备有限公司 一种语音操作事件添加方法及终端
CN108831469A (zh) * 2018-08-06 2018-11-16 珠海格力电器股份有限公司 语音命令定制方法、装置和设备及计算机存储介质

Also Published As

Publication number Publication date
CN108831469A (zh) 2018-11-16
CN108831469B (zh) 2021-02-12

Similar Documents

Publication Publication Date Title
WO2020029500A1 (zh) 语音命令定制方法、装置和设备及计算机存储介质
US11600265B2 (en) Systems and methods for determining whether to trigger a voice capable device based on speaking cadence
US10489112B1 (en) Method for user training of information dialogue system
US9953648B2 (en) Electronic device and method for controlling the same
US20160293168A1 (en) Method of setting personal wake-up word by text for voice control
JP4942970B2 (ja) 音声認識における動詞誤りの回復
US10811005B2 (en) Adapting voice input processing based on voice input characteristics
KR102108500B1 (ko) 번역 기반 통신 서비스 지원 방법 및 시스템과, 이를 지원하는 단말기
US20170046124A1 (en) Responding to Human Spoken Audio Based on User Input
US20160328205A1 (en) Method and Apparatus for Voice Operation of Mobile Applications Having Unnamed View Elements
US20150371628A1 (en) User-adapted speech recognition
KR20200012933A (ko) 어시스턴트 애플리케이션을 위한 음성 사용자 인터페이스 단축
WO2017012511A1 (zh) 语音控制方法、装置及投影仪设备
US10860289B2 (en) Flexible voice-based information retrieval system for virtual assistant
US20060253272A1 (en) Voice prompts for use in speech-to-speech translation system
CN107331400A (zh) 一种声纹识别性能提升方法、装置、终端及存储介质
WO2020024620A1 (zh) 语音信息的处理方法以及装置、设备和存储介质
TW200813812A (en) Methods and apparatus for improved voice recognition and voice recognition systems
WO2020233363A1 (zh) 语音识别的方法、装置、电子设备和存储介质
WO2019228138A1 (zh) 音乐播放方法、装置、存储介质及电子设备
WO2019239656A1 (ja) 情報処理装置および情報処理方法
KR20190001435A (ko) 음성 입력에 대응하는 동작을 수행하는 전자 장치
WO2017092322A1 (zh) 智能电视的浏览器操作方法及智能电视
KR102584324B1 (ko) 음성 인식 서비스 제공 방법 및 이를 위한 장치
TWI770395B (zh) 語音控制電視版銀行轉帳之裝置與方法

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: 18929666

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18929666

Country of ref document: EP

Kind code of ref document: A1