WO2022158824A1

WO2022158824A1 - Method and device for controlling electronic apparatus

Info

Publication number: WO2022158824A1
Application number: PCT/KR2022/000919
Authority: WO
Inventors: Jin Guo; Wuzhi LUO; Xiaorong Liang
Original assignee: Samsung Electronics Co., Ltd.
Priority date: 2021-01-21
Filing date: 2022-01-18
Publication date: 2022-07-28
Also published as: CN112856727A

Abstract

A method and device for controlling an electronic apparatus are disclosed. The method includes: acquiring a user voice signal; acquiring environmental information and an operating status of the electronic apparatus; determining a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus; and controlling operation of the electronic apparatus by using the control command.

Description

METHOD AND DEVICE FOR CONTROLLING ELECTRONIC APPARATUS

The present disclosure relates to a control method and a control device of an electronic apparatus, in particular, to a method and device for controlling an electronic apparatus using a user voice signal and environmental information and an operating status of the electronic apparatus.

With the intelligentization of electronic terminals and the development of information network technology, people use voice interaction applications in more and more scenarios. For example, using a smart phone or a smart speaker or an intelligent electronic apparatus itself as an entrance, and using voice interaction applications through voice-text interaction, such as electrical appliances/weather/stocks/music/traffic conditions/alarms/reminds. Here, each voice interaction application is also called semantic skill.

However, a voice interaction logic is variable and complex. A well-designed voice system training process may improve a user's experience of interacting with the system and increase the user's trust in system services and functions. For example, a manufacturer of smart phones and smart speakers that support voice training systems repairs one or more voices according to user settings and gives corresponding one or more responses (for example, voice dialogue/text dialogue/device control instructions/screen recording teachings, etc.). However, under the existing voice and text training systems, users can only set each model one by one, which leads to cumbersome and repetitive operations.

The purpose of embodiments of the present disclosure is to provide a method and device for controlling an electronic apparatus.

According to an aspect of the embodiments of the present disclosure, a method for controlling an electronic apparatus is provided, the method including: acquiring a user voice signal; acquiring environmental information and an operating status of the electronic apparatus; determining a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus; and controlling operation of the electronic apparatus by using the control command.

Optionally, before the determining a control command corresponding to the user voice signal, the method further includes: recognizing the user voice signal through natural language understanding processing to determine an intention indicated by the user voice signal.

Optionally, before the determining a control command corresponding to the user voice signal, the method further includes: determining whether the user voice signal matches a preset habit template in the electronic apparatus; and in response to the user voice signal matching the preset habit template, determining an intention indicated by the user voice signal from the preset habit template.

Optionally, the step of determining a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus, includes: inputting the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus into a control command acquisition model to obtain the control command corresponding to the user voice signal.

Optionally, the control command acquisition model is a model based on a convolutional neural network.

Optionally, the convolutional neural network is a residual neural network.

Optionally, the control command acquisition model is obtained through training sample data or historical data.

Optionally, the control command acquisition model is trained by: inputting a user intention in the sample data or the historical data, the environmental information and the operating status of the electronic apparatus into the control command acquisition model; and adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data.

Optionally, the step of adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data, includes: in response to the output of the control command acquisition model being consistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a positive example; and in response to the output of the control command acquisition model being inconsistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a negative example, to adjust the parameters of the control command acquisition model.

Optionally, the electronic apparatus includes an air conditioner.

Optionally, the environmental information of the electronic apparatus includes an indoor temperature and/or an outdoor temperature, and/or the operating status of the electronic apparatus includes a switch status, a cooling status, and/or a heating status.

According to another aspect of the embodiments of the present disclosure, a device for controlling an electronic apparatus is provided, the device including: a first acquisition module, configured to acquire a user voice signal; a second acquisition module, configured to acquire environmental information and an operating status of the electronic apparatus; a command determination module, configured to determine a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus; and a controlling module, configured to control operation of the electronic apparatus by using the control command.

Optionally, the device further includes: a recognition module, configured to recognize the user voice signal through natural language understanding processing to determine an intention indicated by the user voice signal, before the command determination module determines a control command corresponding to the user voice signal.

Optionally, the device further includes: a recognition module, configured to determine whether the user voice signal matches a preset habit template in the electronic apparatus, before the command determination module determines a control command corresponding to the user voice signal, and in response to the user voice signal matching the preset habit template, determine an intention indicated by the user voice signal from the preset habit template.

Optionally, the command determination module is configured to input the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus into a control command acquisition model to obtain the control command corresponding to the user voice signal.

Optionally, the convolutional neural network is a residual neural network.

Optionally, the operation of adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data, includes: in response to the output of the control command acquisition model being consistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a positive example; and in response to the output of the control command acquisition model being inconsistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a negative example, to adjust the parameters of the control command acquisition model.

Optionally, the electronic apparatus includes an air conditioner.

According to another aspect of the embodiments of the present disclosure, a computer readable storage medium storing computer programs is provided, the computer programs, when executed by a processor, implement the method for controlling an electronic apparatus.

According to another aspect of the present disclosure, a computing apparatus is provided, the computing apparatus including: a processor; and a memory, storing a computer program, the computer program, when executed by the processor, implements the method for controlling an electronic apparatus.

By using the method and device, an accuracy of generating control commands based on voice signals can be improved, and unnecessary operations for generating control commands can be reduced, and an efficiency of voice control can be improved.

The method and device for controlling an electronic apparatus according to the embodiments of the present disclosure, can accurately generate a control signal of the electronic apparatus based on a user voice signal and environmental information and an operating status of the electronic apparatus, and simply based on a control command acquisition model. There is no need to configure different control command acquisition models for different environmental information and operating status of the electronic apparatus, thereby avoiding complicated operations of users/manufacturers to set various models, and at the same time improving service efficiency of the electronic apparatus.

The above and other objectives and features of embodiments of the present disclosure will become clearer from the following description in conjunction with the accompanying drawings exemplarily showing the embodiments, in which:

Fig. 1 is a flowchart showing a method for controlling an electronic apparatus according to an embodiment of the present disclosure;

Fig. 2 is a block diagram showing a device for controlling an electronic apparatus according to an embodiment of the present disclosure; and

Fig. 3 is a block diagram showing a computing apparatus according to an embodiment of the present disclosure.

The following specific implementations are provided to help readers gain a comprehensive understanding of the methods, devices, and/or systems described herein. However, after understanding the disclosure of the present application, various changes, modifications and equivalents of the methods, devices and/or systems described herein will be clear. For example, the orders of operations described herein are only examples, and are not limited to those set forth herein, but except for operations that must occur in a specific order, they may be changed as will be clear after understanding the disclosure of the present application. In addition, for more clarity and conciseness, descriptions of features known in the art may be omitted.

The features described herein may be implemented in different forms and should not be construed as being limited to the examples described herein. On the contrary, the examples described herein have been provided to show only some of the many feasible methods for implementing the methods, devices, and/or systems described herein, which will be clear after understanding the disclosure of the present application.

As used herein, the term "and/or" includes any one of the associated listed items and any combination of any two or more.

Although terms such as "first", "second" and "third" may be used herein to describe various members, components, regions, layers or parts, these members, components, regions, layers or parts should not be restricted by these terms. On the contrary, these terms are only used to distinguish one member, component, region, layer or part from another member, component, region, layer or part. Therefore, without departing from the teaching of the examples, a first member, a first component, a first region, a first layer, or a first part in the examples described herein may also be referred to as a second member, a second component, a second region, a second layer or a second part.

In the specification, when an element (such as a layer, region, or substrate) is described as being "on", "connected to", or "coupled to" another element, the element may be directly "on" the other element, directly "connected to" or "coupled to" another element, or there may be one or more other elements intervening there between. In contrast, when an element is described as being "directly on", "directly connected to" or "directly coupled to" another element, there may be no intervening elements.

The terms used herein are only used to describe various examples, and are not intended to limit the present disclosure. Unless the context clearly indicates otherwise, the singular form is intended to include the plural form. The terms "including", "comprising" and "having" indicate the existence of the features, numbers, operations, members, elements, and/or combinations thereof, but do not exclude the existence or addition of one or more other features, numbers, operations, members, elements and/or combinations thereof.

Unless defined otherwise, all terms used herein (including technical terms and scientific terms) have the same meaning as commonly understood by those skilled in the art to which the present disclosure belongs after understanding the present disclosure. Unless clearly defined as such herein, terms (such as those defined in the general dictionaries) should be interpreted as having a meaning consistent with their contextual meaning in the related field and in the present disclosure, and should not be interpreted as having ideal or excessive formal meaning.

In addition, in the description of the examples, when it is considered that a detailed description of a well-known related structure or function will cause a vague interpretation of the present disclosure, such detailed description will be omitted.

The following describes application scenarios of the method and device for controlling an electronic apparatus according to the present disclosure. The method and device for controlling an electronic apparatus according to the present disclosure may be applied to an air conditioner, but is not limited thereto. For example, the method and device for controlling an electronic apparatus according to the present disclosure may be applied to a display, a speaker, and the like.

In real life, it often involves a plurality of sets of training voice texts and corresponding instructions, especially when an intelligent electronic apparatus has a same voice habit for the same user, the voice habit is often regular. In view of the related habit logic, a plurality of sets of habits may be generated through artificial intelligence training of a set of habits, thereby avoiding the user from manually editing a plurality of sets of habits and improving service efficiency.

According to the prior art, in winter, in response to a user voice signal "the living room is cold", based on a voice command acquisition model, a temperature of an (living room) air conditioner may be raised (heating), or the air conditioner may be turned on (heating). For example, the (living room) air conditioner itself may raise its temperature or may be turned on based on the voice command acquisition model. In addition, an intelligent central control may control the (living room) air conditioner to raise the temperature or control the air conditioner to start based on the voice command acquisition model. However, the command acquisition model is not suitable for summer, but can only control heating based on the user voice signal, and cannot learn. When the season changes, the command acquisition model cannot switch between "heating" and "cooling".

According to the present disclosure, a new command acquisition model may be designed. Based on the new command acquisition model, in response to the user voice signal "the living room is cold", based on environmental information (including an indoor temperature and an outdoor temperature) and an operating status (a switch status/a cooling status/a heating status) of an environment of the air conditioner, the temperature of the (living room) air conditioner may be raised (heating) or the air conditioner may be turned on (heating), or the temperature of the air conditioner may be raised (cooling) or the air conditioner may be turned off (cooling).

Hereinafter, the method and device for controlling an electronic apparatus according to the embodiments of the present disclosure will be described with reference to Fig. 1 and Fig. 2.

Fig. 1 is a flowchart showing a method for controlling an electronic apparatus according to an embodiment of the present disclosure. The method for controlling an electronic apparatus according to an embodiment of the present disclosure may run on the electronic apparatus itself, or may run on an intelligent central control controlling the electronic apparatus, and the present disclosure does not impose any limitation in this regard.

Referring to Fig. 1, in step S110, a user voice signal may be acquired. For example, the user voice signal may be acquired through a microphone set in the electronic apparatus or the intelligent central control. In step S120, environmental information and an operating status of the electronic apparatus may be acquired. As described above, the electronic apparatus may be an air conditioner, but is not limited thereto. For example, the electronic apparatus may be a display, a speaker, or the like. The environmental information of the electronic apparatus includes an indoor temperature and/or an outdoor temperature, and the operating status of the electronic apparatus includes a switch status, a cooling status, and/or a heating status. For example, the environmental information of the electronic apparatus may be acquired through various sensors set in the electronic apparatus or the intelligent central control, and the operating status of the electronic apparatus may be acquired through a controller/processor of the electronic apparatus. The order of step S110 and step S120 is not limited. For example, step S110 may be performed before step S120, may be performed after step S120, or may be performed simultaneously with step S120.

Next, in step S130, a control command corresponding to the user voice signal may be determined based on the acquired environmental information and the operating status of the electronic apparatus.

In order to determine the control command corresponding to the user voice signal, before step S120, the method for controlling an electronic apparatus according to an embodiment of the present disclosure may include a step as follows: recognizing the user voice signal through natural language understanding processing (NLU) to determine an intention indicated by the user voice signal. For example, when the user voice signal is "it is very hot", it may be determined through NLU that an intention indicated by the user voice signal is "cooling down", and the intention is related to control of the electronic apparatus. However, when the user voice signal is "how is the weather today", it may be determined through NLU that an intention indicated by the user voice signal is "asking for the weather", and the intention has nothing to do with the control of the electronic apparatus.

In another aspect, in order to determine the control command corresponding to the user voice signal, before step S120, the method for controlling an electronic apparatus according to an embodiment of the present disclosure may include steps as follows: determining whether the user voice signal matches a preset habit template in the electronic apparatus; and in response to the user voice signal matching the preset habit template, determining an intention indicated by the user voice signal from the preset habit template. For example, a user may pre-set a habit template "Aoligei" to be associated with an intention "cooling down". When the user voice signal is "Aoligei", it may be determined that the user voice signal matches the preset habit template in the electronic apparatus, and the intention "cooling down" indicated by the user voice signal may be determined from the preset habit template. However, when the user voice signal is "Fighting", it may be determined that the user voice signal does not match the preset habit template in the electronic apparatus, so that the intention indicated by the user voice signal cannot be acquired.

After determining the intention indicated by the user voice signal, in step S130, the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus may be input into a control command acquisition model to obtain the control command corresponding to the user voice signal. In step S140, the obtained control command may be used to control operation of the electronic apparatus.

For example, if the user voice signal is "the living room is cold" and the intention indicated by the voice signal is determined to be "warming up", the indoor temperature is 22 degrees (℃), the outdoor temperature is 32 degrees (℃), and the operating status of the electronic apparatus is "colling". Through the control command acquisition model, the control command "turn off the air conditioner" or "set the air conditioner temperature to 26 degrees (℃)" corresponding to the user voice signal may be obtained. Optionally, based on the user voice signal "too hot", and the determined intention indicated by the voice signal "cooling down", the indoor temperature is 30 degrees (℃), the outdoor temperature is 1 degree (℃), and the operating status of the electronic apparatus is "heating", through the control command acquisition model, the control command "turn off the air conditioner" or "set the air conditioner temperature to 22 degrees (℃)" corresponding to the user voice signal may be obtained.

According to an embodiment of the present disclosure, the control command acquisition model may be a model based on a convolutional neural network (CNN), and the convolutional neural network may be a residual neural network (Resnet), but is not limited thereto. Those skilled in the art may configure the CNN according to actual needs, and detailed description thereof will be omitted.

The following describes a training process of the control command acquisition model.

The control command acquisition model is obtained through training sample data or historical data. To this end, first, historical data including a user intention, the environmental information and the operating status of the electronic apparatus input may be collected, or sample data including a user intention, the environmental information and the operating status of the electronic apparatus input may be set. Then, the user intention, the environmental information and the operating status of the electronic apparatus in the sample data or the historical data are input into the control command acquisition model. Finally, based on an output of the control command acquisition model and expected decision data, parameters of the control command acquisition model are adjusted. Optionally, the expected decision data as an expected output of the control command acquisition model may be preset. When the output of the control command acquisition model is consistent with the expected decision data, the output of the control command acquisition model may be fed back to the control command acquisition model in a form of a positive example to strengthen the parameters of the control command acquisition model currently set. On the other hand, when the output of the control command acquisition model is inconsistent with the expected decision data, the output of the control command acquisition model may be fed back to the control command acquisition model in a form of a negative example to adjust the parameters of the control command acquisition model. By repeating the above training process, optimal parameters of the control command acquisition model may be finally determined. Optionally, the above training process may be performed by a manufacturer of the electronic apparatus in a manufacturing/testing phase of the electronic apparatus, or may be performed by a user of the electronic apparatus during using the electronic apparatus.

For example, a sample intention or a preset intention (such as "warming up") and the operating status of the air conditioner (such as "cooling") may be input into the control command acquisition model, and the indoor temperature (such as 22℃) and the outdoor temperature (such as 22℃) may be input into the control command acquisition model. The control command acquisition model may classify the operating status and a set temperature of the air conditioner through Resnet based on the input sample intention or preset intention, the operating status of the air conditioner, the indoor temperature and the outdoor temperature, and output the operating status and the set temperature of the air conditioner. When the operating status and the set temperature of the air conditioner (for example, "cooling" and "26℃") output by the control command acquisition model are consistent with the expected decision data, the operating status and the set temperature of the air conditioner output by the control command acquisition model are fed back to the control command acquisition model in the form of a positive example to strengthen the parameters of the control command acquisition model currently set. However, when the operating status and the set temperature of the air conditioner (for example, "cooling" and "24℃") output by the control command acquisition model are inconsistent with the expected decision data, the operating status and the set temperature of the air conditioner output by the control command acquisition model are fed back to the control command acquisition model in the form of a negative example to adjust the parameters of the control command acquisition model.

Using the above method for controlling an electronic apparatus according to an embodiment of the present disclosure, there is no need to configure different control command acquisition models for different environmental information and operating status of the electronic apparatus, avoiding complicated operations of users/manufacturers to set various models, and at the same time improving service efficiency of the electronic apparatus.

Fig. 2 is a block diagram showing a device for controlling an electronic apparatus according to an embodiment of the present disclosure. The device for controlling an electronic apparatus according to an embodiment of the present disclosure may be set on the electronic apparatus, or may be set on an intelligent central control controlling the electronic apparatus, and the present disclosure does not impose any limitation in this regard.

Referring to Fig. 2, a device 200 for controlling an electronic apparatus according to an embodiment of the present disclosure may include a first acquisition module 210, a second acquisition module 220, a command determination module 230 and a controlling module 240. The first acquisition module 210 may be configured to acquire a user voice signal. The second acquisition module 220 may be configured to acquire environmental information and an operating status of the electronic apparatus. The command determination module 230 may be configured to determine a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus. The controlling module 240 may be configured to control operation of the electronic apparatus by using the control command. As described above, the electronic apparatus may be an air conditioner, but is not limited thereto. For example, the electronic apparatus may be a display, a speaker, or the like. The environmental information of the electronic apparatus includes an indoor temperature and/or an outdoor temperature, and the operating status of the electronic apparatus includes a switch status, a cooling status, and/or a heating status. For example, the environmental information of the electronic apparatus may be acquired through various sensors set in the electronic apparatus or the intelligent central control, and the operating status of the electronic apparatus may be acquired through a controller/processor of the electronic apparatus.

The device 200 for controlling an electronic apparatus according to an embodiment of the present disclosure may further include a recognition module (not shown). The recognition module may be configured to recognize the user voice signal through natural language understanding processing to determine an intention indicated by the user voice signal, before the command determination module determines a control command corresponding to the user voice signal, or determine whether the user voice signal matches a preset habit template in the electronic apparatus, before the command determination module determines a control command corresponding to the user voice signal, and in response to the user voice signal matching the preset habit template, determine an intention indicated by the user voice signal from the preset habit template. Then, the command determination module 230 may be configured to input the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus into a control command acquisition model to obtain the control command corresponding to the user voice signal.

According to an embodiment of the present disclosure, the control command acquisition model may be a model based on a convolutional neural network (CNN), and the convolutional neural network may be a residual neural network (Resnet), but is not limited thereto. The control command acquisition model may be obtained through training sample data or historical data. A user intention in the sample data or the historical data, the environmental information and the operating status of the electronic apparatus may be input into the control command acquisition model, then parameters of the control command acquisition model may be adjusted based on an output of the control command acquisition model and expected decision data. For example, when the output of the control command acquisition model is consistent with the expected decision data, the output of the control command acquisition model may be fed back to the control command acquisition model in a form of a positive example, and when the output of the control command acquisition model is inconsistent with the expected decision data, the output of the control command acquisition model may be fed back to the control command acquisition model in a form of a negative example to adjust the parameters of the control command acquisition model.

Referring to Fig. 3, a computing apparatus 300 according to an embodiment of the present disclosure may include a processor 310 and a memory 320. The processor 310 may include (but is not limited to) a central processing unit (CPU), a digital signal processor (DSP), a microcomputer, a field programmable gate array (FPGA), a system on chip (SoC), a microprocessor, an application specific integrated circuit (ASIC), etc. The memory 320 stores computer programs to be executed by the processor 310. The memory 320 includes a high-speed random access memory and/or a non-volatile computer readable storage medium. When the processor 310 executes the computer programs stored in the memory 320, the method for controlling an electronic apparatus as described above may be implemented. Optionally, the computing apparatus 300 may communicate with the electronic apparatus or an intelligent central control controlling the electronic apparatus in a wired/wireless communication.

The method for controlling an electronic apparatus according to an embodiment of the present disclosure may be written as computer programs and stored on the computer readable storage medium. When the computer programs are executed by the processor, the method for controlling an electronic apparatus as described above may be implemented. Examples of the computer readable storage medium include: read only memory (ROM), random access programmable read only memory (PROM), electrically erasable programmable read only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or CD storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as multimedia cards, secure digital (SD) cards, or extreme digital (XD) cards), magnetic tape, floppy disk, magneto-optical data storage apparatus, optical data storage apparatus, hard disk, solid state disk, and any other apparatuses. The any other apparatuses are configured to store the computer programs and any associated data, data files and data structures in a non-transitory mode and provide the computer programs and any associated data, data files and data structures to the processor or computer, enabling the processor or computer to execute the computer programs. In one example, the computer programs and any associated data, data files and data structures are distributed on a networked computer system, so that the computer programs and any associated data, data files and data structures are stored, accessed, and executed in a distributed mode through one or more processors or computers.

Although some embodiments of the present disclosure have been shown and described, those skilled in the art should understand that these embodiments may be modified without departing from the principle and spirit of the present disclosure whose scope is defined by the claims and their equivalents.

Claims

A method for controlling an electronic apparatus, the method comprising:

acquiring a user voice signal;

acquiring environmental information and an operating status of the electronic apparatus;

determining a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus; and

controlling operation of the electronic apparatus by using the control command.
The method according to claim 1, wherein, before the determining a control command corresponding to the user voice signal, the method further comprises: recognizing the user voice signal through natural language understanding processing to determine an intention indicated by the user voice signal.
The method according to claim 1, wherein, before the determining a control command corresponding to the user voice signal, the method further comprises:

determining whether the user voice signal matches a preset habit template in the electronic apparatus; and

in response to the user voice signal matching the preset habit template, determining an intention indicated by the user voice signal from the preset habit template.
The method according to claim 2, wherein, the step of determining a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus, comprises:

inputting the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus into a control command acquisition model to obtain the control command corresponding to the user voice signal.
The method according to claim 4, wherein the control command acquisition model is obtained through training sample data or historical data.
The method according to claim 5, wherein the control command acquisition model is trained by:

inputting a user intention in the sample data or the historical data, the environmental information and the operating status of the electronic apparatus into the control command acquisition model; and

adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data.
The method according to claim 6, wherein, the step of adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data, comprise:

in response to the output of the control command acquisition model being consistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a positive example; and

in response to the output of the control command acquisition model being inconsistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a negative example, to adjust the parameters of the control command acquisition model.
The method according to claim 1, wherein the electronic apparatus comprises an air conditioner, and

wherein, the environmental information of the electronic apparatus comprises an indoor temperature and/or an outdoor temperature, and/or

the operating status of the electronic apparatus comprises a switch status, a cooling status, and/or a heating status.
A device for controlling an electronic apparatus, the device comprising:

a first acquisition module, configured to acquire a user voice signal;

a second acquisition module, configured to acquire environmental information and an operating status of the electronic apparatus;

a command determination module, configured to determine a control command corresponding to the user voice signal based on the acquired environmental information and the operating status of the electronic apparatus; and

a controlling module, configured to control operation of the electronic apparatus by using the control command.
The device according to claim 9, wherein the device further comprises:

a recognition module, configured to recognize the user voice signal through natural language understanding processing to determine an intention indicated by the user voice signal, before the command determination module determines a control command corresponding to the user voice signal.
The device according to claim 9, wherein the device further comprises:

a recognition module, configured to determine whether the user voice signal matches a preset habit template in the electronic apparatus, before the command determination module determines a control command corresponding to the user voice signal, and in response to the user voice signal matching the preset habit template, determine an intention indicated by the user voice signal from the preset habit template.
The device according to claim 10, wherein the command determination module is configured to input the determined intention indicated by the user voice signal, the environmental information and the operating status of the electronic apparatus into a control command acquisition model to obtain the control command corresponding to the user voice signal.
The device according to claim 12, wherein the control command acquisition model is obtained through training sample data or historical data.
The device according to claim 13, wherein the control command acquisition model is trained by:

inputting a user intention in the sample data or the historical data, the environmental information and the operating status of the electronic apparatus into the control command acquisition model; and

adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data.
The device according to claim 14, wherein, the operation of adjusting parameters of the control command acquisition model based on an output of the control command acquisition model and expected decision data, comprise:

in response to the output of the control command acquisition model being consistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a positive example; and

in response to the output of the control command acquisition model being inconsistent with the expected decision data, feeding back the output of the control command acquisition model to the control command acquisition model in a form of a negative example, to adjust the parameters of the control command acquisition model.