CN117811862A - Intelligent household appliance control method, system and storage medium based on voice recognition - Google Patents

Abstract

The invention discloses a voice recognition-based intelligent household appliance control method, a voice recognition-based intelligent household appliance control system and a voice recognition-based intelligent household appliance control storage medium, which belong to the field of intelligent household appliance control, wherein the voice recognition-based intelligent household appliance control method comprises the following steps of: establishing a control network; after the mobile terminal accesses the control network, authenticating the mobile terminal; after receiving control voices, the intelligent household appliances send the received control voices to a cloud module through a central control module, and the cloud module compares whether text information converted by different control voices is the same or not after converting the control voices into text information, and determines correct text information as target information; and if the target information does not comprise the household appliance name, identifying the requirement of the target information based on the current working mode of each intelligent household appliance so as to determine the target equipment in the intelligent household appliance. The intelligent household appliance control method and the intelligent household appliance control system can control the intelligent household appliance under the condition that wake-up words are not included, so that convenience in household appliance control is improved.

Description

Intelligent household appliance control method, system and storage medium based on voice recognition

Technical Field

The invention belongs to the technical field of intelligent household appliance control, and particularly relates to an intelligent household appliance control method, system and storage medium based on voice recognition.

Background

The rise of intelligent home greatly facilitates the life of people. At present, direct human-computer interaction independent voice control is generally performed through a specific voice activation instruction of the household appliance, or intelligent household appliance remote control is generally performed through a mobile phone APP. However, when the individual voice control is performed on the individual intelligent home appliances one to one, the use is inconvenient, after the user must activate the intelligent home appliances by using the designated activation instruction, the user issues the corresponding action command, and when the types of the intelligent home appliances are multiple, the user may not be able to recall the activation instruction, so that the intelligent home appliances cannot be controlled by voice; and through mobile phone APP control, the corresponding APP needs to be opened for control, which increases intermediate control links and reduces the convenience of voice control

Disclosure of Invention

Aiming at the problems, an intelligent household appliance control method, an intelligent household appliance control system and a storage medium based on voice recognition are provided, and the aim is to solve the problem that the intelligent household appliance is inconvenient to control by voice in the prior art.

In a first aspect, the present invention further provides a method for controlling an intelligent home appliance based on voice recognition, where the method includes:

the central control module establishes a control network and accesses the mobile terminal and the intelligent household appliance into the control network;

after the mobile terminal is accessed to the control network, the authentication module authenticates the mobile terminal, and after the authentication is successful, if the mobile terminal is the first accessed device in the control network, the voice module in each intelligent household appliance is activated;

when a plurality of intelligent household appliances receive control voices, the control voices are sent to the central control module, the central control module sends the control voices to the cloud module, the cloud module converts the control voices into text information and compares whether the text information converted by different control voices is the same or not, if yes, one piece of the text information is selected as target information, and if no, the correct text information is determined as the target information;

the cloud module sends the target information to the central control module, and if the target information does not comprise a household appliance name, the cloud module identifies the requirement of the target information based on the current working mode of each intelligent household appliance so as to determine target equipment in the intelligent household appliances, wherein the target equipment is the intelligent household appliance which is directly controlled by the control voice;

the central control module generates a first instruction for controlling the target equipment and a second instruction for controlling the non-target equipment, and each intelligent household appliance executes corresponding actions based on the first instruction and the second instruction respectively.

Further, the cloud module converting the control voice into the text information includes the following steps:

the central control module acquires the identification of each target device in the control network, and acquires the type of the target device and the included working modes based on the identification, wherein each working mode corresponds to a pre-allocated identification code;

an indoor map is generated in the central control module, the indoor map is marked with the position of the intelligent household appliance, after the control voice is received, the sounding position of a user is positioned, the current working mode of the intelligent household appliance, which is positioned in the same room with the sounding position, is obtained, a noise sequence is generated based on the identification code corresponding to the current working mode, and the central control module sends the noise sequence and the control voice to the cloud module;

and after receiving the control voice, the cloud module calls the corresponding separation model based on the noise sequence, separates a human voice track from the control voice by using the called separation model, and then calls the recognition model to convert the human voice track into the text information.

Further, the central control module generating the indoor map includes the steps of:

the method comprises the steps that the position of a central control module is preset, the central control module generates a virtual instruction, the virtual instruction is sent to each intelligent household appliance so that the intelligent household appliances execute virtual actions, the central control module and each intelligent household appliance acquire sounds generated by the virtual actions, the azimuth relation between each intelligent household appliance and the central control module is judged based on the size of the sounds, the azimuth relation between each intelligent household appliance and each intelligent household appliance is judged, and a first initial map is generated based on the azimuth relation;

if the designated equipment is not accessed in the control network, the central control module generates notification information and sends the notification information to the mobile terminal, a user is reminded to manually draw a second initial map, and if the designated equipment is accessed, the second initial map stored in the designated equipment is acquired;

and combining the first initial map with the second initial map to generate the indoor map.

Further, determining the target device in the smart home appliance includes the steps of:

the central control module is internally provided with a function table, the function table comprises the function of each intelligent household appliance and a plurality of trigger characters corresponding to the function, the trigger characters comprise positive characters and negative characters, the trigger characters are extracted from the text information, and the function table is used for matching the corresponding function and defining the function as a target function;

the intelligent household appliance which comprises the target function and is in an operation state is acquired and defined as an initial target, if only one initial target exists, the initial target is directly set as target equipment, if a plurality of initial targets exist, the current working modes of the initial targets are continuously acquired, the contribution value of each working mode is prestored by the central control module, the contribution value of the current working mode is acquired, and the target equipment in the initial targets is determined by combining the type of the extracted trigger character and the size of the contribution value.

Further, generating the second instruction for non-target device control includes the steps of:

the intelligent household appliances which are located in the same room at present and positively influence the functions related to the trigger characters are defined as first household appliances, the intelligent household appliances which generate negative influences are defined as second household appliances, the central control module generates positive second instructions for controlling the first household appliances and negative second instructions for controlling the second household appliances, the positive second instructions are used for controlling the functions related to the first household appliances and the trigger characters so as to increase the influence on the environment, and the negative second instructions are used for controlling the second household appliances to reduce the influence on the environment.

Further, authenticating the mobile terminal includes the steps of:

and prompting a user to send out authentication voice during authentication, wherein the central control module confirms whether the voice belongs to a member in the control network according to the tone and the tone of the authentication voice, and distinguishing adult users and child users based on voice tone and tone confirmation, wherein the adult users have control rights of all intelligent household appliances, and all child users have control rights of non-kitchen use of the intelligent household appliances.

Further, the central control module stores names of various intelligent household appliances and wake-up words of family members for the various intelligent household appliances, and if the central control module recognizes that the text information includes the wake-up words, the central control module skips the judging process of the target equipment.

In a second aspect, the present invention provides a voice recognition-based intelligent home appliance control system, including:

the central control module establishes a control network, and accesses a mobile terminal and intelligent household appliances into the control network, the central control module also receives the control voice and sends the control voice to the cloud module, and receives target information from the cloud module, if the target information does not comprise a household appliance name, the requirement of the target information is identified based on the current working mode of each intelligent household appliance so as to determine target equipment in the intelligent household appliances, the target equipment is the intelligent household appliance which is directly controlled by the control voice, and the central control module generates a first instruction for controlling the target equipment and a second instruction for controlling non-target equipment;

the authentication module is used for authenticating the mobile terminal after the mobile terminal is accessed to the control network, and activating the voice module in each intelligent household appliance if the mobile terminal is the first accessed device in the control network after the authentication is successful

The voice module is arranged in each intelligent household appliance and is used for receiving the control voice and sending the control voice to the central control module;

the cloud module is used for comparing whether the text information converted by different control voices is the same or not after converting the control voices into the text information, if yes, selecting one piece of the text information as target information, and if not, determining that the text information is correct as the target information and sending the target information to the central control module.

In a third aspect, the present invention provides a storage medium, where program instructions are stored, where when the program instructions run, the storage medium is controlled to execute the intelligent home appliance control method based on voice recognition.

Compared with the prior art, the invention has the following beneficial effects:

under the condition of establishing a control network, after each intelligent household appliance receives the control voice, the voice is sent to the central control module through the control network, and then the voice is packaged by the central control module and sent to the cloud module, so that the collective recognition of all the control voices at the cloud module is realized, and the comparison of subsequent text information is convenient; through the comparison of the text information, the correct text information can be extracted from the text information to serve as target information, and finally, the central control module analyzes the target information to determine the requirement of control voice, so that the corresponding intelligent household appliance can be controlled under the condition that the control voice does not contain the household appliance name.

The central control module in the invention also monitors the number of the mobile terminals in the control network, if the mobile terminals exist in the control network, the voice recognition function of each intelligent household appliance connected to the control network is activated, if the mobile terminals do not exist in the control network, the voice recognition function of the intelligent household appliance is closed, so that the energy consumption can be saved, and meanwhile, the situation that the intelligent household appliance is started by mistake after receiving the voice of the external environment when no one exists in the household can be avoided.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of a method for controlling an intelligent home appliance based on voice recognition according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an indoor map according to an embodiment of the present invention;

fig. 3 is a block diagram of an intelligent household appliance control system based on voice recognition according to an embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be apparent that the particular embodiments described herein are merely illustrative of the present invention and are some, but not all embodiments of the present invention. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on embodiments of the present invention, are within the scope of the present invention.

Fig. 1 is a flowchart of a voice recognition-based intelligent household appliance control method provided by the invention, and the flowchart specifically includes:

step S1: the central control module establishes a control network and accesses the mobile terminal and the intelligent household appliance into the control network.

Specifically, the central control module may be a bluetooth gateway, the established control network may be a Zigbee or Bee network, and the mobile terminal may be a smart phone, a tablet computer, a notebook computer, or the like. The intelligent household appliances comprise an intelligent refrigerator, an intelligent water heater, an intelligent water dispenser, an intelligent television, an intelligent air conditioner, an intelligent light, an intelligent curtain, an intelligent closestool, a sweeping robot and the like.

Step S2: after the mobile terminal is accessed to the control network, the authentication module authenticates the mobile terminal, and after the authentication is successful, if the mobile terminal is the first accessed device in the control network, the voice module in each intelligent household appliance is activated.

Specifically, after detecting that the mobile terminal is accessed to the network, the central control module sends an authentication prompt to the mobile terminal, and after receiving the authentication prompt, the mobile terminal authenticates by a preset authentication mode, wherein the authentication mode can be password authentication, voiceprint authentication, fingerprint authentication or face recognition authentication. When the mobile terminal which is accessed does not exist in the control network, the voice recognition module in each intelligent household appliance is in a dormant state in a default state, and the control voice is not detected, so that the power consumption can be saved; after that, if the central control module detects that the mobile terminal is accessed in the control network, the voice module is activated again.

Step S3: when a plurality of intelligent household appliances receive control voices, the control voices are sent to a central control module, the central control module sends the control voices to a cloud module, the cloud module compares whether text information converted by different control voices is the same or not after converting the control voices into the text information, one piece of text information is selected as target information under the condition that the text information converted by different control voices is the same, and the correct text information is determined as the target information under the condition that the text information converted by different control voices is not the same.

After the intelligent household appliance receives the control voice, the received control voice is sent to the central control module, and the central control module sends the control voice to the cloud module for recognition and comparison through the Internet; the reason for this comparison is that, as shown in fig. 2, a television TV, a sound TC, a central control module CL, and a floor sweeping machine SR are placed in a living room, a washing machine HW is placed in a balcony, and a refrigerator RF is placed in a kitchen; when a user sends out control voice at the position P1, the intelligent home appliances are far away from the position P1, so that the sizes of the received voice of the home appliances at different positions are different and are smaller, the voice quality received by which equipment is better cannot be ensured, and further, the converted text information cannot be determined to be correct, so that comparison is needed; the voting method is adopted for comparison, if two pieces of text information are identical in 3 pieces of information, the content of the two pieces of text information is set as target information.

Step S4: and the cloud module sends the target information to the central control module, and if the target information does not comprise the household appliance name, the cloud module identifies the requirement of the target information based on the current working mode of each intelligent household appliance so as to determine the target equipment in the intelligent household appliance, wherein the target equipment is the intelligent household appliance to be directly controlled by the control voice.

Step S5: the central control module generates a first instruction for controlling the target equipment and a second instruction for controlling the non-target equipment, and each intelligent household appliance executes corresponding actions based on the first instruction and the second instruction respectively.

The control voices in the embodiment include two types of home appliance names and home appliance names, for example, the control voice without a home appliance name is "too noisy", the control voice with a home appliance name is "air conditioner, temperature is adjusted to be equal to or less than the same as the first party," help me to increase the temperature of the air conditioner "; when the converted text information does not include the appliance name, the device to which the target information is directed is determined based on the current operation mode of each appliance, such as "noisy" may be directed to a television or sound, it is necessary to determine which device is the target device, and then a first instruction for the target device and a second instruction for a non-target device are generated. The specific manner in which the target device is identified and the first and second instructions are generated will be described in detail later.

Under the condition of establishing a control network, after each intelligent household appliance receives the control voice, the voice is sent to the central control module through the control network, and then the voice is packaged by the central control module and sent to the cloud module, so that the collective recognition of all the control voices at the cloud module is realized, and the comparison of subsequent text information is convenient; through the comparison of the text information, the correct text information can be extracted from the text information to serve as target information, and finally, the central control module analyzes the target information to determine the requirement of control voice, so that the corresponding intelligent household appliance can be controlled under the condition that the control voice does not contain the household appliance name.

The central control module in the invention also monitors the number of the mobile terminals in the control network, if the mobile terminals exist in the control network, the voice recognition function of each intelligent household appliance connected to the control network is activated, if the mobile terminals do not exist in the control network, the voice recognition function of the intelligent household appliance is closed, so that the energy consumption can be saved, and meanwhile, the situation that the intelligent household appliance is started by mistake after receiving the voice of the external environment when no one exists in the household can be avoided.

Particularly, the intelligent household appliance control method and the intelligent household appliance control system can control the intelligent household appliance under the condition that wake-up words are not included, so that convenience in household appliance control is improved.

In this embodiment, the cloud module converting the control voice into text information includes the following steps:

the central control module acquires the identification of each target device in the control network, and acquires the type of the target device and the included working modes based on the identification, wherein each working mode corresponds to a pre-allocated identification code.

Specifically, a working mode comparison table is arranged in the central control module, the working mode comparison table comprises identifiers of all target devices, working modes and identification codes, specifically, the central control module determines the type of intelligent household appliances in a control network through the identifiers, determines the working modes according to the current identification codes of the intelligent household appliances, for example, the identifier 10000 is an air conditioner, the corresponding identification code 01 is a refrigerating mode, 02 is a heating mode, 01000 is a washing machine, the identification code A is a washing mode and B is a dewatering mode.

An indoor map is generated in the central control module, the indoor map is marked with the position of the intelligent household appliance, after control voice is received, the sounding position of a user is positioned, the current working mode of the intelligent household appliance, which is positioned in the same room with the sounding position, is obtained, a noise sequence is generated based on the identification code corresponding to the current working mode, and the central control module sends the noise sequence and the control voice to the cloud module together.

Specifically, as shown in fig. 2, which is an indoor map, after each intelligent home appliance receives a control voice, the position P2 where the user is located is first located, and the central control module determines the name of the intelligent home appliance sending the control voice based on the identifier attached to the control voice; because the television TV, the loudspeaker box TC and the sweeping robot SR are located in the same room as the user and are in the operation mode, the central control module obtains the working modes of the television TV, the loudspeaker box TC and the sweeping robot SR based on the identification codes and generates a corresponding noise sequence, for example, the television identification code is W12, the sound volume is 12, the loudspeaker box identification code is Y20, the sound volume is 20, the sweeping robot SR is U, and the sound volume is 20, and the sweeping robot SR is in the powerful sweeping mode, then the noise sequence W12Y20U is generated and sent to the cloud module.

The cloud module is provided with a plurality of separation models in advance, after receiving control voice, the cloud module calls the corresponding separation model based on the noise sequence, separates a voice track from the control voice by using the called separation model, and then calls the recognition model to convert the voice track into text information.

The cloud module is internally provided with a pre-trained separation model, the separation model is constructed based on a neural network technology, the neural network model can be BP, RBF or GAN, and the like, and a plurality of separation models exist, and different models are generated through different training data; if the separation model 1 is generated by training a noise sample set comprising human voice, television voice between 10-20, sound box in rock mode and sound volume under 10-20 and the sweeping robot under strong sweeping, the separation model 2 is generated by training a noise sample set comprising human voice and the sweeping robot under ordinary sweeping; therefore, the cloud module calls the corresponding separation model to process after receiving the noise sequence; in addition, the recognition model in the embodiment is also constructed based on the neural network technology, such as the RNN model; the method has the advantages that different models are built for processing according to different noise conditions, and the model processing effect trained through the special sample set is better than that of a general model, so that different models are called for processing based on different noise environments, the voice separated from the model is clearer, and the accuracy of converting subsequent voice into text is improved.

In this embodiment, the central control module generating the indoor map includes the steps of:

the method comprises the steps of presetting the position of a central control module, generating a virtual instruction by the central control module, sending the virtual instruction to each intelligent household appliance so that the intelligent household appliances execute virtual actions, acquiring sounds generated by the virtual actions by the central control module and each intelligent household appliance, judging the azimuth relation between each intelligent household appliance and the central control module and the azimuth relation between the intelligent household appliances based on the sound, and generating a first initial map based on the azimuth relation.

Specifically, the position of the central control module is marked in a blank map in a manual mode, then the central control module generates a virtual instruction and sends the virtual instruction to the intelligent household appliance, for example, the washing machine enters a dehydration mode, a fan is started, a refrigerator sends warning sound and the like, then the intelligent household appliance positions the positions of other household appliances through sounds and directions sent by other household appliances, various methods such as a sound transmission speed method and a voiceprint positioning method exist through the sound positioning positions, and particularly, a microphone array is inevitably carried in equipment carrying the voice module, which is basic hardware of voice recognition, and the relative direction between two equipment can be determined through the sound transmission speed method on the premise that the equipment comprises the microphone array; if the washing machine HW emits sound, a circular microphone array is arranged in the central control module, wherein the direction of the connection line between the microphone which receives the sound first and the center of the array is the direction of the washing machine HW relative to the central control module, and the time difference between the time when the washing machine HW emits the sound and the time when the central control module receives the sound is multiplied by the sound propagation speed, the approximate distance between the microphone array and the central control module can be obtained, so that the relative position of the washing machine HW and the central control module can be obtained, and the judging mode of the directions of other household appliances and the central control module is the same and is not repeated here. After the relative position relation between each household appliance and the central control module is known, the position relation between the household appliances is determined; after determining the relative positions between the individual appliances, a first initial map may be generated.

If the designated equipment is not accessed in the control network, the central control module generates notification information and sends the notification information to the mobile terminal, a user is reminded to manually draw a second initial map, and if the designated equipment is accessed, the second initial map stored in the designated equipment is acquired from the designated equipment.

And combining the first initial map with the second initial map to generate an indoor map.

The second initial map is a house type map of a room, and the designated equipment is a sweeping robot, so that the sweeping robot can traverse the room and establish a room map when in use, and under the condition that the sweeping robot exists in the house, the second initial map can be directly obtained from the house, and the first initial map and the second initial map are combined to generate an indoor map comprising the intelligent household appliance position; in the absence of the sweeping robot, the user is prompted to manually draw a house pattern on the first initial map to generate a specific second initial map.

In this embodiment, determining the target device in the smart home appliance includes the following steps:

the central control module is internally provided with a function table, the function table comprises functions of each intelligent household appliance and a plurality of trigger characters corresponding to the functions, the trigger characters comprise positive characters and negative characters, the trigger characters are extracted from text information, the corresponding functions are matched based on the function table, and the function table is defined as a target function.

In the function table, the corresponding functions of the air conditioner comprise heating, cooling, dehumidifying and the like, the trigger characters of the heating function comprise cooling, heating and the like, the sound box and the television comprise audio playing functions, the corresponding trigger characters comprise big, small, noisy, inaudible and the like, the type of the positive trigger character is small and inaudible, and the negative trigger character is big and noisy; after receiving the text information, the central control module extracts trigger characters in the text information, for example, the text information is "noisy" and extracts "noisy" therefrom, and then the corresponding function is acquired as an audio playing function and defined as a target function.

The method comprises the steps of obtaining intelligent household appliances which comprise target functions and are in an operation state, defining the intelligent household appliances as initial targets, directly setting the initial targets as target devices if only one initial target exists, continuously obtaining current working modes of all the initial targets if a plurality of initial targets exist, storing contribution values of each working mode in advance by a central control module, obtaining the contribution values of the current working modes, and determining the target devices in the initial targets by combining the types of the extracted trigger characters and the sizes of the contribution values.

Then, the intelligent household appliance in the running state is acquired, referring to fig. 2, a user sends out voice "real o" at a position P2, the television and the sound box are defined as initial targets in the running state of the television and the sound box, the working modes of the television and the sound box are acquired, the contribution values of various working modes of the initial targets are preset, for example, the contribution value of the sound box in a rock mode and the contribution value of the sound volume 20 is 30, the contribution value of the television in a normal mode and the contribution value of the sound volume is 15 when the sound volume is 12, and the initial equipment with the largest contribution value is set as target equipment because the trigger character "o" is a negative character, and then a first instruction for controlling the sound volume of the sound box to be reduced is generated, namely, the sound is considered to be overlarge at the moment, and the user is influenced to watch the television. For another example, the user utters the voice "too little sound" and "little" is the forward trigger character, so the initial device with the smallest contribution value is set as the target device, and the first instruction for controlling the television volume to increase is generated, that is, the television sound is considered too little at this time. For another example, if the user utters a voice "too little sound" but only the television is in an operating state in the environment, the first instruction for controlling the volume of the television to be increased is directly generated.

In this embodiment, generating the second instruction for non-target device control includes the steps of:

the intelligent household appliance which is located in the same room with the sounding position and has positive influence on the function related to the trigger character is defined as a first household appliance, the intelligent household appliance which has negative influence is defined as a second household appliance, the central control module generates a positive second instruction for controlling the first household appliance and a negative second instruction for controlling the second household appliance, the positive second instruction is used for controlling the function related to the first household appliance and the trigger character so as to increase the influence on the environment, and the negative second instruction is used for controlling the second household appliance to reduce the influence on the environment.

Specifically, if the user sends out control voice that is "too noisy", non-target equipment located in the room is obtained, as shown in fig. 2, if the target equipment is sound TC, that is, sound of the sound TC is too loud to influence the user to watch television, the non-target equipment comprises a television TV, a sweeping robot SR and an intelligent sliding DOOR, a positive second instruction for the television TV and the intelligent sliding DOOR is generated, and a negative second instruction for the sweeping robot SR is generated; specifically, by increasing the volume of the television TV, the user can hear the television sound more easily, and the sliding DOOR DOOR is controlled to be closed, so that the influence on the environment is isolated, namely, the influence on the environment is reduced by reducing the cleaning force of the sweeping robot SR, namely, the influence on the environment is reduced; in addition, if the user sends out the instruction of "noisy" many times, further generate the instruction of moving the robot SR out of the room, thus further reduce the influence on living room environment.

In this embodiment, authenticating the mobile terminal includes the steps of:

the central control module confirms whether the voice belongs to members in the control network according to the tone and the tone of the authentication voice, and distinguishes adult users and child users based on the voice tone and the tone confirmation, wherein the adult users have the control authority of all intelligent household appliances, and all child users have the control authority of non-kitchen intelligent household appliances.

In this embodiment, the central control module stores names of various intelligent home appliances and wake-up words of family members for the various intelligent home appliances, and if the central control module identifies that the text information includes the wake-up words, the central control module skips the judging process for the target device.

As shown in fig. 3, the present invention provides an intelligent home appliance control system based on voice recognition, comprising:

the central control module establishes a control network, accesses the mobile terminal and the intelligent home appliances into the control network, sends control voice received by the central control module to the cloud module, receives target information from the cloud module, identifies the requirement of the target information based on the working mode of each intelligent home appliance at present if the target information does not comprise the home appliance name so as to determine target equipment in the intelligent home appliances, wherein the target equipment is the intelligent home appliance which is directly controlled by the control voice, and generates a first instruction for controlling the target equipment and a second instruction for controlling non-target equipment;

the authentication module is used for authenticating the mobile terminal after the mobile terminal is accessed to the control network, and activating the voice module in each intelligent household appliance if the mobile terminal is the first accessed device in the control network after the authentication is successful

The voice module is arranged in each intelligent household appliance, and is used for receiving the control voice and sending the control voice to the central control module;

after the cloud module converts the control voice into text information, comparing whether the text information converted by different control voices is the same or not, if yes, selecting one piece of text information as target information, and if not, determining the correct text information as the target information and sending the target information to the central control module.

The invention provides a storage medium, which stores program instructions, wherein the program instructions control equipment where the storage medium is located to execute the intelligent household appliance control method based on voice recognition.

The foregoing examples merely illustrate preferred embodiments of the invention, which are described in greater detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Claims (9)

1. The intelligent household appliance control method based on voice recognition is characterized by comprising the following steps of:

the central control module establishes a control network and accesses the mobile terminal and the intelligent household appliance into the control network;

after the mobile terminal is accessed to the control network, the authentication module authenticates the mobile terminal, and after the authentication is successful, if the mobile terminal is the first accessed device in the control network, the voice module in each intelligent household appliance is activated;

when a plurality of intelligent household appliances receive control voices, the control voices are sent to the central control module, the central control module sends the control voices to the cloud module, the cloud module converts the control voices into text information and compares whether the text information converted by different control voices is the same or not, if yes, one piece of the text information is selected as target information, and if no, the correct text information is determined as the target information;

the cloud module sends the target information to the central control module, and if the target information does not comprise a household appliance name, the cloud module identifies the requirement of the target information based on the current working mode of each intelligent household appliance so as to determine target equipment in the intelligent household appliances, wherein the target equipment is the intelligent household appliance which is directly controlled by the control voice;

the central control module generates a first instruction for controlling the target equipment and a second instruction for controlling the non-target equipment, and each intelligent household appliance executes corresponding actions based on the first instruction and the second instruction respectively.

2. The voice recognition-based intelligent home appliance control method according to claim 1, wherein the cloud module converting the control voice into the text information comprises the steps of:

the central control module acquires the identification of each target device in the control network, and acquires the type of the target device and the included working modes based on the identification, wherein each working mode corresponds to a pre-allocated identification code;

an indoor map is generated in the central control module, the indoor map is marked with the position of the intelligent household appliance, after the control voice is received, the sounding position of a user is positioned, the current working mode of the intelligent household appliance, which is positioned in the same room with the sounding position, is obtained, a noise sequence is generated based on the identification code corresponding to the current working mode, and the central control module sends the noise sequence and the control voice to the cloud module;

and after receiving the control voice, the cloud module calls the corresponding separation model based on the noise sequence, separates a human voice track from the control voice by using the called separation model, and then calls the recognition model to convert the human voice track into the text information.

3. The voice recognition-based intelligent home appliance control method of claim 2, wherein the central control module generating the indoor map comprises the steps of:

the method comprises the steps that the position of a central control module is preset, the central control module generates a virtual instruction, the virtual instruction is sent to each intelligent household appliance so that the intelligent household appliances execute virtual actions, the central control module and each intelligent household appliance acquire sounds generated by the virtual actions, the azimuth relation between each intelligent household appliance and the central control module is judged based on the size of the sounds, the azimuth relation between each intelligent household appliance and each intelligent household appliance is judged, and a first initial map is generated based on the azimuth relation;

if the designated equipment is not accessed in the control network, the central control module generates notification information and sends the notification information to the mobile terminal, a user is reminded to manually draw a second initial map, and if the designated equipment is accessed, the second initial map stored in the designated equipment is acquired;

and combining the first initial map with the second initial map to generate the indoor map.

4. The voice recognition-based smart home control method of claim 2, wherein determining the target device in the smart home comprises:

the central control module is internally provided with a function table, the function table comprises the function of each intelligent household appliance and a plurality of trigger characters corresponding to the function, the trigger characters comprise positive characters and negative characters, the trigger characters are extracted from the text information, and the function table is used for matching the corresponding function and defining the function as a target function;

the intelligent household appliance which comprises the target function and is in an operation state is acquired and defined as an initial target, if only one initial target exists, the initial target is directly set as target equipment, if a plurality of initial targets exist, the current working modes of the initial targets are continuously acquired, the contribution value of each working mode is prestored by the central control module, the contribution value of the current working mode is acquired, and the target equipment in the initial targets is determined by combining the type of the extracted trigger character and the size of the contribution value.

5. The voice recognition-based intelligent appliance control method of claim 4, wherein generating the second instruction for non-target device control comprises the steps of:

the intelligent household appliances which are located in the same room at present and positively influence the functions related to the trigger characters are defined as first household appliances, the intelligent household appliances which generate negative influences are defined as second household appliances, the central control module generates positive second instructions for controlling the first household appliances and negative second instructions for controlling the second household appliances, the positive second instructions are used for controlling the functions related to the first household appliances and the trigger characters so as to increase the influence on the environment, and the negative second instructions are used for controlling the second household appliances to reduce the influence on the environment.

6. The voice recognition-based intelligent home appliance control method according to claim 2, wherein authenticating the mobile terminal comprises the steps of:

and prompting a user to send out authentication voice during authentication, wherein the central control module confirms whether the voice belongs to a member in the control network according to the tone and the tone of the authentication voice, and distinguishing adult users and child users based on voice tone and tone confirmation, wherein the adult users have control rights of all intelligent household appliances, and all child users have control rights of non-kitchen use of the intelligent household appliances.

7. The voice recognition-based intelligent home appliance control method according to claim 6, wherein the central control module stores names of various intelligent home appliances and wake-up words of family members for the various intelligent home appliances, and if the central control module recognizes that the text information includes the wake-up words, the central control module skips the judgment process for the target device.

8. A voice recognition-based intelligent home appliance control system for implementing the voice recognition-based intelligent home appliance control method as set forth in any one of claims 1 to 7, comprising:

the central control module establishes a control network, and accesses a mobile terminal and intelligent household appliances into the control network, the central control module also receives the control voice and sends the control voice to the cloud module, and receives target information from the cloud module, if the target information does not comprise a household appliance name, the requirement of the target information is identified based on the current working mode of each intelligent household appliance so as to determine target equipment in the intelligent household appliances, the target equipment is the intelligent household appliance which is directly controlled by the control voice, and the central control module generates a first instruction for controlling the target equipment and a second instruction for controlling non-target equipment;

the authentication module is used for authenticating the mobile terminal after the mobile terminal is accessed to the control network, and activating the voice module in each intelligent household appliance if the mobile terminal is the first accessed device in the control network after the authentication is successful

The voice module is arranged in each intelligent household appliance and is used for receiving the control voice and sending the control voice to the central control module;

the cloud module is used for comparing whether the text information converted by different control voices is the same or not after converting the control voices into the text information, if yes, selecting one piece of the text information as target information, and if not, determining that the text information is correct as the target information and sending the target information to the central control module.

9. A computer storage medium, wherein the computer storage medium stores program instructions, and when the program instructions are executed, the computer storage medium is controlled to perform the intelligent household appliance control method based on voice recognition according to any one of claims 1 to 7.