CN107622652A - Voice control method for home appliance system and home appliance control system - Google Patents

Abstract

The present invention provides a voice control method for a household appliance system and a household appliance control system. The applicable household appliance system includes a plurality of electrical equipment, and each electrical equipment is data-connected to a cloud controller, wherein a part or all of the multiple electrical equipment are configured There is a voice collection device as the voice interaction device, and the voice control method of the home appliance system includes: determining the voice response device from the voice interaction device that has obtained the wake-up command; the voice response device uses its voice collection device to collect surrounding voice signals, And extract the vocal feature data from the collected voice signal; send the vocal feature data to the cloud controller for the cloud controller to generate control instructions for the home appliance system according to the vocal feature data, and send the specified electrical equipment to the control instruction send. Thus, the distributed voice interactive control network is used to conveniently implement the voice control of any electrical device in the household electrical appliance system, which greatly improves the user experience.

Description

Voice control method for home appliance system and home appliance control system

technical field

The invention relates to home appliance control, in particular to a voice control method for a home appliance system and a home appliance control system.

Background technique

At present, the intelligent control technology for home appliances is developing very rapidly, and voice control technology is one of the important ones.

At present, there are a large number of smart home appliances that support voice control, and these products generally integrate voice recognition modules on home appliances. When the household appliance is in the standby state, the voice recognition module picks up the surrounding voice signal and recognizes it. When it is determined that it matches its own preset wake-up signal, it wakes up from the standby state and executes the subsequent voice control instructions issued by the user. .

However, the existing voice control technology, even the far-field voice, can only meet the voice recognition within a distance of 3 to 5 meters in a complex noise environment, and the voice interaction distance is limited, which requires the user to be in the controlled household appliances. The control command can only be issued in the vicinity. If the user is in the horizontal type and wants to control the water heater in the bathroom, he must go to the bathroom to give a control command to the water heater, which brings great inconvenience to the user and greatly limits the application prospect of voice control.

Contents of the invention

An object of the present invention is to provide a voice control method and a home appliance control system for a home appliance system with good user experience.

A further object of the present invention is to enable the voice control method to accurately realize the user's control intention.

In particular, the present invention provides a voice control method for a household electrical appliance system. The applicable household electrical appliance system includes multiple electrical equipment, and each electrical equipment is connected to the cloud controller for data, wherein some or all of the multiple electrical equipment are equipped with The voice collection device is used as a voice interaction device, and the voice control method of the home appliance system includes: determining a voice response device from the voice interaction device that has obtained a wake-up instruction; the voice response device uses its voice collection device to collect surrounding voice signals, and Extract the vocal feature data from the collected voice signal; send the vocal feature data to the cloud controller for the cloud controller to generate control instructions for the home appliance system based on the vocal feature data, and send them to the electrical equipment specified by the control instruction .

Optionally, the step of the cloud controller generating control instructions for the home appliance system according to the vocal feature data includes: the cloud controller performs speech recognition on the vocal feature data and converts it into corresponding text; uses a preset semantic analysis model to The converted text is subjected to semantic analysis to determine whether the text corresponds to the preset control command in the command library; if so, extract the control command corresponding to the text.

Optionally, in the case that the text does not correspond to the preset control instructions in the instruction library, use the text as a keyword to initiate a search to the preset knowledge search engine; receive the search results returned by the knowledge search engine, and respond to the voice Provided by the device; the voice response device synthesizes a reply voice based on the search results and plays it to the user.

Optionally, the step of extracting vocal feature data from the collected speech signal includes: converting the speech signal into a digital signal, and performing front-end noise reduction processing on the digital signal; performing endpoint detection on the digital signal processed by the front-end noise reduction , to determine the start point and cut-off point of the vocal; feature extraction is performed on the digital signal between the start point and the cut-off point to obtain the vocal feature data.

Optionally, the step of determining the voice response device from the voice interaction device that has obtained the wake-up instruction includes: multiple voice interaction devices respectively use their own voice collection devices to collect surrounding voice signals; identify the collected voice signals, To determine whether the voice signal matches the wake-up instruction; determine the voice response device from the voice interaction devices that receive the voice signal that matches the wake-up instruction.

Optionally, the step of determining the voice response device from the voice interaction devices that receive the voice signal that matches the wake-up instruction includes: if there are multiple voice interaction devices that receive voice signals that match the wake-up instruction within the first set time voice signal, then extract the signal parameters of the voice signals received by multiple voice interactive devices, set the response priority of multiple voice interactive devices according to the signal parameters, and determine the voice interactive device with the highest priority as the voice response equipment.

According to another aspect of the present invention, a home appliance control system is also provided, which is used to control a home appliance system including a plurality of electrical equipment, and a part or all of the multiple electrical equipment is equipped with a voice collection device as a voice interaction device , and the home appliance control system includes: a cloud controller, which is respectively connected to a plurality of electrical appliances in data, and is configured to determine the voice response device from the voice interaction device that has obtained the wake-up command; and the voice response device is configured to utilize its voice The device collects the voice signal around it, extracts the voice feature data from the collected voice signal, and sends the voice feature data to the cloud controller, so that the cloud controller can generate the control system for the home appliance based on the voice feature data. Control command, and send to the electrical equipment specified by the control command.

Optionally, the cloud controller is also configured to: conduct speech recognition on the vocal feature data and convert it into corresponding text; use a preset semantic analysis model to perform semantic analysis on the converted text to determine whether the text is consistent with the instruction library Correspond to the preset control command; if so, extract the control command corresponding to the text; if the text does not correspond to the preset control command in the command library, use the text as a keyword to initiate a search to the preset knowledge search engine ; Receive the search results returned by the knowledge search engine and provide them to the voice response device; and the voice response device is also configured to synthesize a reply voice according to the search results and play it to the user.

Optionally, the voice response device is further configured to: convert the voice signal into a digital signal, and perform front-end noise reduction processing on the digital signal; perform endpoint detection on the digital signal after the front-end noise reduction processing, to determine the starting point and cut-off point; feature extraction is performed on the digital signal between the start point and the cut-off point to obtain vocal feature data.

Optionally, the voice interaction device is also configured to collect surrounding voice signals using respective voice collection devices; the cloud controller is also configured to identify the collected voice signals to determine whether the voice signals match the wake-up instruction , determine the voice response device from the voice interaction devices that receive the voice signal matching the wake-up instruction.

In the voice control method of the household appliance system of the present invention, a voice collection device is installed in some or all of the electrical equipment in the household electrical appliance system, and the voice control signal sent by the user is extracted from the surrounding voice signals by using these distributed electrical equipment, and the voice control signal is obtained by The cloud controller processes and recognizes the voice control signal, and generates corresponding control instructions for execution by the corresponding electrical equipment in the home appliance system. It can interact with users through voice interaction devices within its processing range, and use the cloud controller Coordinated control enables users to conveniently control any electrical equipment in the home appliance system, greatly improving the convenience of use.

Furthermore, in the household appliance control system of the present invention, it is only necessary to arrange voice collection devices in some electrical equipment for voice collection and preliminary processing, and the complex voice recognition is performed by the cloud controller, which saves the hardware cost of electrical equipment.

Furthermore, in the voice control method of the home appliance system of the present invention, multiple voice interactive devices can use the same wake-up word or wake-up instruction, and the cloud controller automatically determines the voice interaction device that receives the voice signal matching the wake-up instruction. Voice response equipment prevents users from memorizing complex wake-up words, forms a distributed voice control network, and improves user experience.

Those skilled in the art will be more aware of the above and other objects, advantages and features of the present invention according to the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings.

Description of drawings

Hereinafter, some specific embodiments of the present invention will be described in detail by way of illustration and not limitation with reference to the accompanying drawings. The same reference numerals in the drawings designate the same or similar parts or parts. Those skilled in the art will appreciate that the drawings are not necessarily drawn to scale. In the attached picture:

Fig. 1 is a schematic architecture diagram of a home appliance system suitable for a home appliance control system according to an embodiment of the present invention;

Fig. 2 is a schematic architecture diagram of a home appliance control system according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a voice control method for a home appliance system according to an embodiment of the present invention; and

Fig. 4 is an optional execution flow chart of the voice control method of the home appliance system according to one embodiment of the present invention.

detailed description

Embodiments of the present invention provide a voice control method for a home appliance system and a home appliance control system, which are suitable for voice control of home appliance systems arranged within a certain area. For example, the smart home appliances arranged in the home are uniformly controlled to form a distributed voice control network, enabling users to conveniently and flexibly control the electrical equipment within their management range.

Fig. 1 is a schematic architecture diagram of a home appliance system to which a home appliance control system 10 according to an embodiment of the present invention is applied. The home appliance system includes various electrical equipment 100 with remote control functions arranged in a certain characteristic area, such as kitchen appliances (hoods, dishwashers, microwave ovens), lamps, air conditioners, and televisions arranged in each room of the home appliance. , refrigerators, water heaters, washing machines, purifiers and other electrical equipment 100, some of these electrical appliances may be placed in the same room, and some may be placed in different rooms, and the specific space layout is determined according to the user's needs. These electrical appliances 100 all have transmission means (which can be wired network connection, wireless network connection, local area network connection, etc.), and can be connected to cloud controller 200 or other centralized control devices through network transmission equipment 300 (such as routers, home gateways, etc.). The electrical equipment 100 can execute the remote control command sent by the cloud controller 200 or the centralized control device to complete the corresponding function. For example, adjust the indoor temperature, turn on the lights, etc. The way of data connection between the electrical equipment 100 and the cloud controller 200 includes but not limited to Wi-Fi, zigbee and other wireless or wired connection ways.

The cloud controller 200 can be a network server set on the network side, and in some optional embodiments, it can also be replaced by other centralized control devices for household appliances that can be connected to the Internet.

The home appliance control system 10 of an embodiment of the present invention is aimed at the above-mentioned home appliance system, and completes the automatic issuing of control instructions through voice control. Fig. 2 is a schematic architecture diagram of the home appliance control system 10 according to an embodiment of the present invention. In the household appliance control system 10 , part or all of the above-mentioned plurality of electrical appliances 100 are equipped with a voice collection device 112 , which can be configured as a voice interaction device 110 . In some embodiments, the entire space where the home appliance system is located can be divided into several sub-areas, and a voice interaction device 110 equipped with a voice collection device 112 is arranged in each sub-area, for example, in each room of the home. One voice interaction device 110 can also be equipped with multiple voice interaction devices 110 in a larger room at the same time. In some other embodiments, all the electrical equipment 100 in the household electrical appliance system can be configured with the voice collection device 112 , so that all the electrical equipment 100 can be used as the voice interaction equipment 110 .

The speech collection device 112 may include a microphone array or a single microphone device, an analog-to-digital conversion device, a front-end noise reduction device, a speech preprocessing device, and the like. The microphone can pick up the surrounding voice signal, the analog-to-digital conversion device converts the analog voice signal into a digital signal, the noise reduction device can filter out noise, and the voice preprocessing device can perform voice endpoint detection and feature extraction. The preprocessed voice data can be sent to the cloud controller 200 through the transmission device of the voice interaction device 110 where the voice collection device 112 is located.

The cloud controller 200 may determine the voice response device 114 from the voice interaction devices 110 that have obtained the wake-up instruction. In some optional embodiments, the process of determining the voice response device 114 may be as follows: the voice interaction device 110 uses its own voice collection device 112 to collect surrounding voice signals; the cloud controller 200 recognizes the collected voice signals to determine Whether the voice signal matches the wake-up instruction, the voice response device 114 is determined from the voice interaction devices 110 that receive the voice signal that matches the wake-up instruction. After the cloud controller 200 selects the voice response device 114, it may issue a control command to the voice response device 114 to enter the voice response state.

Since it may happen that multiple voice interaction devices 110 receive voice signals matching the wake-up instructions at the same time, the home appliance control system 10 of this embodiment also designs a method to realize a method of determining the voice signal from voice interaction devices 110. Answering device 114 . The method is as follows:

A plurality of voice interaction devices 110 are respectively configured to use their respective voice collection devices 112 to collect surrounding voice signals, identify the collected voice signals, extract signal parameters of the voice signals, and determine whether the voice signals match the preset wake-up signals .

The voice interaction device 110 that receives the voice signal matching the preset wake-up signal sends the signal parameters of the received voice signal to the cloud controller 200 . The cloud controller 200 selects the voice response device 114 from the voice interaction device 110 according to the parameters of the voice signal.

When the cloud controller 200 receives the signal parameters sent by multiple voice interaction devices 110 within the first set time (for example, 0.5s, which can be considered as receiving the voice signal sent by the user at the same time), the cloud control The device 200 can also set the response priority of multiple voice interaction devices 110 according to the signal parameters, and set the voice interaction device 110 with the highest response priority as the voice response device 114, and issue a control command to enter the voice response state .

The above-mentioned signal parameter items include one or more of the following: signal-to-noise ratio, voice signal strength, voice feature clarity, and receiving time. The cloud controller 200 can extract the numerical values of each item in the signal parameters; input the numerical values of each item in the signal parameters into the preset multi-dimensional voice quality evaluation model to obtain the voice quality judged by the multi-dimensional voice quality evaluation model; Set the response priority of multiple voice interaction devices 110. This way of determining the response priority can accurately reflect the distance between the user and the voice interaction device 110 and the voice transmission quality, which is beneficial to selecting the optimal voice response device 114 .

The cloud controller 200 can also wait for the voice response device 114 to respond to the confirmation message of the control instruction for entering the voice response state; , the voice response device 114 is used to receive the user's control voice; if no confirmation message is received within the second set time, the next response priority voice interaction device 110 is set as the voice response device 114 .

When the cloud controller 200 receives a signal parameter sent by a voice interaction device 110 within the first set time, the cloud controller 200 can issue a control command to the voice interaction device 110 to enter the voice response state, so as to It is set as the voice response device 114; after receiving the confirmation message of the voice response device 114 responding to the control instruction of entering the voice response state, the voice response device 114 is used to receive the user's control voice.

Using the above method, multiple voice interaction devices 110 can use the same wake-up word or wake-up instruction, and the cloud controller 200 automatically determines the voice response device 114 from the voice interaction devices 110 that receive the voice signal that matches the wake-up instruction, avoiding Users memorize complex wake-up words, forming a distributed voice control network, which improves user experience.

After the voice response device 114 is determined, the voice response device 114 can use the voice signal around it collected by its voice collection device 112, and extract the vocal characteristic data from the collected voice signal, and send the human voice data to the cloud controller 200. Acoustic feature data, for the cloud controller 200 to generate a control command for the home appliance control system 10 according to the human voice feature data, and send it to the electrical device 100 specified by the control command.

An optional execution process for the cloud controller 200 to generate control instructions is: conduct speech recognition on the vocal feature data and convert it into corresponding text; use a preset semantic analysis model to perform semantic analysis on the converted text to judge the text Whether it corresponds to the preset control command in the command library; if so, extract the control command corresponding to the text; The knowledge search engine initiates a search; the search result returned by the knowledge search engine is received and provided to the voice response device 114 . The voice response device 114 can synthesize a reply voice according to the search result, and play it to the user. The user can conveniently and flexibly issue control instructions, and can also use the voice response device 114 to intelligently obtain network information.

An optional way for the voice response device 114 to extract the vocal feature data from the collected voice signal may be: convert the voice signal into a digital signal, and perform front-end noise reduction processing on the digital signal; Endpoint detection is performed on the processed digital signal to determine the start point and cut-off point of the human voice; feature extraction is performed on the digital signal between the start point and the cut-off point to obtain vocal feature data.

The voice control method of the home appliance system provided by the embodiment of the present invention can be executed by the home appliance control system 10 of any of the above-mentioned embodiments to realize distributed voice control of home appliances and build a home voice network based on the electrical equipment 100, so as to meet the needs of users. the needs of life. FIG. 3 is a schematic diagram of a voice control method for a home appliance system according to an embodiment of the present invention. The voice control method for a home appliance system may generally include:

Step S302, determining the voice response device 114 from the voice interaction device 110 that has obtained the wake-up instruction;

Step S304, the voice response device 114 uses its voice collection device 112 to collect surrounding voice signals, and extracts vocal feature data from the collected voice signals;

Step S306, sending the vocal feature data to the cloud controller 200;

In step S308, the cloud controller 200 generates a control command for the home appliance system according to the human voice feature data, and sends it to the electrical device 100 specified by the control command.

Wherein, an optional manner of step S302 is: multiple voice interactive devices 110 respectively use their respective voice collection devices 112 to collect surrounding voice signals; identify the collected voice signals to determine whether the voice signals are consistent with the wake-up instruction; Matching: determine the voice response device 114 from the voice interaction devices 110 that receive the voice signal that matches the wake-up instruction.

It is determined that the voice response device 114 can be performed by the cloud controller 200 through the analysis of the voice signal. For example, multiple voice interaction devices 110 are respectively configured to use their respective voice collection devices 112 to collect surrounding voice signals, and analyze the collected voice The signal is identified, the signal parameters of the voice signal are extracted, and it is judged whether the voice signal matches the preset wake-up signal.

The voice interaction device 110 that receives the voice signal matching the preset wake-up signal sends the signal parameters of the received voice signal to the cloud controller 200 . The cloud controller 200 selects the voice response device 114 from the voice interaction device 110 according to the parameters of the voice signal.

When the cloud controller 200 receives the signal parameters sent by multiple voice interaction devices 110 within the first set time (for example, 0.5s, which can be considered as receiving the voice signal sent by the user at the same time), the cloud control The device 200 can also set the response priority of multiple voice interaction devices 110 according to the signal parameters, and set the voice interaction device 110 with the highest response priority as the voice response device 114, and issue a control command to enter the voice response state .

The above-mentioned signal parameter items include one or more of the following: signal-to-noise ratio, voice signal strength, voice feature clarity, and receiving time. The cloud controller 200 can extract the numerical values of each item in the signal parameters; input the numerical values of each item in the signal parameters into the preset multi-dimensional voice quality evaluation model to obtain the voice quality judged by the multi-dimensional voice quality evaluation model; Set the response priority of multiple voice interaction devices 110. This way of determining the response priority can accurately reflect the distance between the user and the voice interaction device 110 and the voice transmission quality, which is beneficial to selecting the optimal voice response device 114 .

The cloud controller 200 can also wait for the voice response device 114 to respond to the confirmation message of the control instruction for entering the voice response state; , the voice response device 114 is used to receive the user's control voice; if no confirmation message is received within the second set time, the next response priority voice interaction device 110 is set as the voice response device 114 .

When the cloud controller 200 receives a signal parameter sent by a voice interaction device 110 within the first set time, the cloud controller 200 can issue a control command to the voice interaction device 110 to enter the voice response state, so as to It is set as the voice response device 114; after receiving the confirmation message of the voice response device 114 responding to the control instruction of entering the voice response state, the voice response device 114 is used to receive the user's control voice.

Using the above method, multiple voice interaction devices 110 can use the same wake-up word or wake-up instruction, and the cloud controller 200 automatically determines the voice response device 114 from the voice interaction devices 110 that receive the voice signal that matches the wake-up instruction, avoiding Users memorize complex wake-up words, forming a distributed voice control network, which improves user experience.

For example, the aforementioned voice interaction device 110 has uniformly set "Haier activation" as the wake-up word, and the user utters the voice of "Haier activation" in the bedroom. If the wake-up instruction matches the voice signal, the air conditioner is set as the voice response device 114, which can play the response voice of "started, waiting for control", and send the control to enter the voice response state to the cloud controller 200 The confirmation message of the instruction, and then wait for the user's next instruction.

If the user speaks the voice of "Haier start" in the bedroom, and it is received by the air conditioner as the voice interaction device 110 in the bedroom and the TV set as the voice interaction device 110 in the adjacent living room, then the air conditioner and the TV The machine will extract the signal parameters of the voice signals collected respectively, such as signal-to-noise ratio, voice signal strength, voice feature clarity, etc. Cloud controller 200 can extract the numerical values of each item in the signal parameters; input the numerical values of each item in the signal parameters into the preset multi-dimensional voice quality evaluation model to obtain the voice quality judged by the multi-dimensional voice quality evaluation model, if the voice quality of the air conditioner is excellent Regarding the voice quality of the TV, the cloud controller 200 uses the air conditioner as the voice response device 114 with the first priority, and uses the TV as the voice response device 114 with the second priority. The cloud controller 200 issues a control command to the air conditioner to enter the voice response state. If the cloud controller 200 receives the confirmation message of the air conditioner within the second set time (for example, 5s, which can be set according to the communication method and reliability), Then use the air conditioner to receive the user's control voice, if the cloud controller 200 does not receive the confirmation message of the air conditioner within the second set time (for example, 5s, which can be set according to the communication method and reliability), then issue a command to the TV set. Enter the control command of the voice response state, and wait for the confirmation message from the TV.

The above are examples. In specific use, multi-level priorities can be set according to the number of voice interaction devices 110 that have received the wake-up instructions, such as three levels, four levels, etc., and the cloud controller 200 can send them in sequence according to the priority. The control command of the voice response state until the voice interaction device 110 responds correctly. In this way, the user can only interact with one optimal voice interaction device 110, which is convenient for the user and avoids resource waste caused by multiple devices working at the same time.

In step S304, the voice response device 114 may process the voice signal as follows: convert the voice signal into a digital signal, and perform front-end noise reduction processing on the digital signal; The start point and cut-off point; feature extraction is performed on the digital signal between the start point and the cut-off point to obtain vocal feature data. The voice response device 114 can only be equipped with preliminary voice processing devices such as a microphone array or a single microphone device, an analog-to-digital conversion device, a front-end noise reduction device, a voice pre-processing device, etc., which does not require powerful voice processing capabilities, saving hardware costs, and complicated The data processing is performed by the cloud controller 200, and by virtue of its powerful data processing capability, accurate processing of voice is realized.

The execution flow of the cloud controller 200 in step S308 may be as follows: the cloud controller 200 performs speech recognition on the vocal feature data and converts it into corresponding text; uses a preset semantic analysis model to perform semantic analysis on the converted text to judge Whether the text corresponds to a preset control command in the command library; if yes, the control command corresponding to the text is extracted. In this step, the cloud controller 200 converts speech to text, and the converted text can be further subjected to semantic analysis. In addition, these texts can also be used as learning targets. Through user feedback, etc., the learning library can be enriched to optimize speech recognition. The model provides the means. In addition, the semantic analysis through text can improve the accuracy of semantic analysis with the help of existing semantic analysis tools.

For example, if the recognized text is "turn on the water heater, the temperature is 60 degrees", the cloud controller 200 can directly send the control command "turn on, set the temperature to 60 degrees" to the water heater for execution by the water heater. The water heater feeds back the execution result to the cloud controller 200, and the cloud controller 200 can feed back the execution result to the voice response device 114, so that the voice response device 114 synthesizes the execution result into the voice "the water heater is turned on" and plays it to the user. Thus completing the entire control process.

In the case that the text does not correspond to the preset control instructions in the command library, the cloud controller 200 can also use the text as a keyword to initiate a search to a preset knowledge search engine; receive the search results returned by the knowledge search engine, and send Provided by the voice response device 114; the voice response device 114 synthesizes a reply voice according to the search results and plays it to the user. For example, if the recognized text is "how is the weather today", the cloud controller 200 can also initiate a search to a preset knowledge search engine to obtain local weather conditions and provide them to the voice response device 114; The received weather conditions are synthesized into a reply voice and played to the user. Or the recognized text is "the result of yesterday's game", and the cloud controller 200 can also initiate a search to a preset knowledge search engine to obtain the game result related to the user, and provide it to the voice response device 114; 114 synthesizes the reply voice according to the received game result, and plays it to the user.

Therefore, the voice control method of the home appliance system in this embodiment can not only implement the voice control of the electrical equipment 100 but also provide the user with a voice interaction function, which greatly improves the intelligence of the user's life.

Fig. 4 is an optional execution flowchart of the voice control method of the household appliance system according to an embodiment of the present invention, the optional execution flow of the voice control method of the household appliance system may include:

Step S402, each voice interactive device 110 respectively collects surrounding voice signals;

Step S404, judging whether any voice interaction device 110 has received a voice signal matching the wake-up instruction, if not, proceed to step S402;

Step S406, if a voice interaction device 110 receives a voice signal matching the wake-up instruction, determine the voice response device 114 from the voice interaction device 110 that has obtained the wake-up instruction, if the voice interaction device 110 that has obtained the wake-up instruction is Multiple, the priority of selecting the voice response device 114 can be set according to the quality of the voice signal, if there is only one voice interaction device 110 that has received the wake-up instruction, the voice interaction device 110 can be directly used as the voice response device 114;

Step S408, the voice response device 114 responds to the user and receives the user's voice signal;

Step S410, the voice response device 114 preliminarily processes the voice signal and uploads it to the cloud controller 200;

Step S412, the cloud controller 200 performs speech recognition on the vocal characteristic data;

Step S414, the cloud controller 200 performs semantic analysis on the recognized text;

Step S416, judging whether the voice analysis result is a preset control command in the command library, if so, execute step S420 to step S424 in sequence, if not, execute step S430 to step S434 in sequence;

Step S420, extracting control instructions corresponding to the text;

Step S422, providing the control instruction to the electrical equipment 100 specified by the control instruction, so that the electrical equipment 100 executes it;

Step S424, feedback the execution result of the controlled electrical device 100 to the customer through the voice response device 114;

Step S430, initiate a search to a preset knowledge search engine;

Step S432, feeding back the search result to the voice response device 114;

Step S434, the voice response device 114 synthesizes a reply voice according to the search result, and plays it to the user.

Utilizing the home appliance control system 10 and the voice control method of the home appliance system in the above embodiment, a distributed voice interaction network can be established, so that users can conveniently implement voice control of any electrical device 100 in the home appliance system at any time, which greatly improves the user experience. use experience.

So far, those skilled in the art should appreciate that, although a number of exemplary embodiments of the present invention have been shown and described in detail herein, without departing from the spirit and scope of the present invention, the disclosed embodiments of the present invention can still be used. Many other variations or modifications consistent with the principles of the invention are directly identified or derived from the content. Accordingly, the scope of the present invention should be understood and deemed to cover all such other variations or modifications.

Claims (10)

1. A voice control method for a home appliance system, the home appliance system comprising a plurality of electrical equipment, each electrical equipment is data-connected to a cloud controller, wherein a part or all of the plurality of electrical equipment is equipped with a voice collection device, As a voice interaction device, and the voice control method of the household appliance system includes: Determine the voice response device from the voice interaction devices that have obtained the wake-up instruction; Collect surrounding voice signals by the voice response device using its voice collection device, and extract vocal feature data from the collected voice signals; and Sending the voice feature data to the cloud controller, so that the cloud controller can generate a control command for the home appliance system according to the voice feature data, and send it to the electrical equipment specified by the control command. 2. The method according to claim 1, wherein the step of the cloud controller generating a control instruction for the home appliance system according to the vocal feature data comprises: performing speech recognition on the vocal feature data by the cloud controller, and converting it into corresponding text; Using a preset semantic analysis model to perform semantic analysis on the converted text to determine whether the text corresponds to a preset control instruction in the instruction library; and If yes, extract the control instruction corresponding to the text. 3. The method of claim 2, wherein, In the case that the text does not correspond to any of the preset control instructions in the instruction library, using the text as a keyword to initiate a search to a preset knowledge search engine; receiving the search results returned by the knowledge search engine and providing them to the voice response device; and The voice response device synthesizes a reply voice according to the search result, and plays it to the user. 4. The method according to claim 1, wherein, the step of extracting vocal feature data from the collected speech signal comprises: Converting the voice signal into a digital signal, and performing front-end noise reduction processing on the digital signal; Endpoint detection is performed on the digital signal processed by the front-end noise reduction to determine the starting point and the ending point of the human voice; and Feature extraction is performed on the digital signal between the start point and the cut-off point to obtain the vocal feature data. 5. The method according to claim 1, wherein the step of determining the voice response device from the voice interaction devices that have obtained the wake-up instruction comprises: A plurality of said voice interaction devices use their respective voice collection devices to collect surrounding voice signals; Recognizing the collected voice signal to determine whether the voice signal matches the wake-up instruction; and The voice response device is determined from the voice interaction devices that have received the voice signal matching the wake-up instruction. 6. The method according to claim 5, wherein the step of determining the voice response device from the voice interaction devices that receive the voice signal matching the wake-up instruction comprises: If there are multiple voice interaction devices that receive voice signals that match the wake-up instruction within the first set time, then extract the signal parameters of the voice signals received by each of the multiple voice interaction devices, according to the The signal parameters set the response priorities of the plurality of voice interaction devices, and determine the voice interaction device with the highest response priority as the voice response device. 7. A home appliance control system, used to control a home appliance system including a plurality of electrical equipment, part or all of the plurality of electrical equipment is equipped with a voice collection device as a voice interaction device, and the home appliance control system include: The cloud controller is connected to the plurality of electrical devices respectively, and is configured to determine the voice response device from the voice interaction device that has obtained the wake-up instruction; and The voice response device is configured to use the surrounding voice signals collected by its voice collection device, and extract vocal feature data from the collected voice signals, and send the vocal feature data to the cloud controller, The cloud controller generates a control command for the home appliance control system according to the voice feature data, and sends it to the electrical device specified by the control command. 8. The home appliance control system according to claim 7, wherein The cloud controller is also configured to: performing speech recognition on the vocal feature data, and converting it into corresponding text; Using a preset semantic analysis model to perform semantic analysis on the converted text to determine whether the text corresponds to a preset control instruction in the instruction library; If so, extracting a control instruction corresponding to the text; In the case that the text does not correspond to any of the preset control instructions in the instruction library, using the text as a keyword to initiate a search to a preset knowledge search engine; receiving the search results returned by the knowledge search engine and providing them to the voice response device; and The voice response device is further configured to synthesize a reply voice according to the search result, and play it to the user. 9. The household appliance control system according to claim 7, wherein the voice response device is further configured to: Converting the voice signal into a digital signal, and performing front-end noise reduction processing on the digital signal; Endpoint detection is performed on the digital signal processed by the front-end noise reduction to determine the starting point and the ending point of the human voice; and Feature extraction is performed on the digital signal between the start point and the cut-off point to obtain the vocal feature data. 10. The home appliance control system according to claim 7, wherein, The voice interaction device is further configured to collect surrounding voice signals by using the respective voice collection devices; and The cloud controller is also configured to identify the collected voice signal to determine whether the voice signal matches the wake-up instruction, and the voice interaction device that receives the voice signal that matches the wake-up instruction Identify the Voice Response Device in .