CN105700389A - Smart home natural language control method - Google Patents

Abstract

The invention discloses a natural language control method for a smart home, which includes the following steps: (1), a voice receiving device receives a voice control command; (2), a control unit analyzes the command; (3), executes the command; in the steps ( In 2), the following sub-steps are also included: (21). Analyze the completeness of the voice control command. If the voice control command contains the complete electrical appliance type and the action command to control the execution, go directly to step (3); otherwise, Go to step (22); (22), if the type of electrical appliance is not specified in the voice control command, locate the user's location, find out the electrical appliance closest to the user according to the user's location, and judge that the electrical appliance is the specified electrical appliance, and execute Step (3). The smart home natural language control method of the present invention adopts the method of indoor positioning of the user, and judges the information implicitly included in the user's incomplete voice control command according to the intelligent analysis of the user's position, thereby improving the recognition of the natural language by the voice control system ability.

Description

A method for natural language control of smart home

technical field

The invention relates to the technical field of smart home, in particular to a natural language control method for smart home.

Background technique

Smart home is a product of informatization and post-industrial age. In the post-industrial era, industry and informatization are closely integrated. Advanced computing methods such as computer vision, speech recognition, and artificial intelligence are widely used. Identification is gradually replacing the remote control. Voice control is the most natural and convenient way to control. At present, the voice recognition function of smart homes is fragmented and scattered. For example, users can only say voice commands that the TV understands to the TV, and voice commands that the air conditioner understands to the air conditioner. They need to remember the voice commands of different smart home appliances separately. This kind of fragmented voice control system makes users feel a kind of uncertainty. Users are always careful to use these new functions, operate them with very simple and fixed verbs or phrase instructions, and watch intently whether the target appliances follow the instructions. Your own instructions take place, the whole process is full of confusion and there is no fun of natural language interaction at all.

Contents of the invention

In order to solve the problem that the existing smart home voice control system does not support natural language control, the user’s natural voice recognition and judgment ability is poor, which leads to failure to recognize and execute, or wrong judgment leads to execution errors, or the system will send out many confirmation prompts, which instead give The user brings a lot of inconvenience, and a smart home natural language control method is proposed, which can locate the user, and judge the information implicitly contained in the user's incomplete voice control command according to the intelligent analysis of the user's location, and improve the voice control. The ability of the system to recognize natural language.

In order to solve the above-mentioned technical problems, the present invention adopts the following technical solutions to achieve:

A natural language control method for a smart home, comprising the following steps: (1), a voice receiving device receiving a voice control command; (2), a control unit analyzing the command; (3), executing the command; in the step (2), Also includes the following substeps:

(21). Analyze the completeness of the voice control command. If the voice control command contains the complete electrical appliance type and the action command to control the execution, go directly to step (3); otherwise, go to step (22);

(22) If the type of electrical appliance is not specified in the voice control command, locate the user's location, find out the electrical appliance closest to the user according to the user's location, and judge that the electrical appliance is the specified electrical appliance, and perform step (3).

Further, at the initial stage of the smart home natural language control method, it also includes the step of setting user rights, collecting voice characteristics of different users of family members, and assigning corresponding user rights levels to them, and establishing a user rights level table;

Before entering step (3) in said step (21), it also includes the step of analyzing command conflicts,

a. If voice control commands from users with different authority levels are received at the same time or within a limited time, execute the voice control commands with higher authority levels;

b. If different voice control commands of users at the same authority level conflict with each other, judge whether the conflicting voice control commands come from the same user, and if they come from the same user, send a voice prompt for user confirmation.

Still further, in the step b of analyzing command conflicts, if the conflicting voice control commands are not from the same user, continue to detect and receive voice control commands, if the conflicting voice control commands continue to appear frequently , it is judged that there is a difference in the opinions of the two users, and the method of executing the command at a discount is adopted to delay the execution of the new voice control command, and each time it is executed, the delay time is increased accordingly until there is no more contradictory voice control Order.

Still further, in the step b of analyzing command conflicts, if the conflicting voice control commands are not from the same user, and the repetition frequency of the conflicting voice control commands is lower than the set limit value, it is judged as One of the users has mischief control, then locate the user's position, find out the user who is closest to the controlled electrical appliance according to the user's position, and execute the user's voice control command.

Preferably, in the step (22), the method for locating the user's location is: using IPV6 address positioning, binding the IPV6 address of each electrical appliance with the actual space address where it is located, and combining the IPV6 address of each voice receiving device The address is bound to the actual space address where it is located, and the location of the user is judged according to the IPV6 address of the voice receiving device receiving the voice control signal, and then the electrical appliance closest to the user's location is controlled accordingly to execute the corresponding control command.

Further, in the step (22), if multiple voice receiving devices receive the voice control command at the same time, the voice receiving device with the strongest strength of the received voice signal is compared according to the intensity of the voice signal received by each voice receiving device, And judge the location of the user with the IPV6 address of the voice receiving device.

Preferably, before the step (1), a step of presetting a trigger word for triggering the voice control operation is also included, and only when the user utters the voice of the trigger word, step (1) or the steps after it are executed.

Preferably, in the step (3), if the voice command does not contain specific parameter information, the user command is executed according to default parameters, and each electrical appliance is set with default parameters.

Compared with the prior art, the advantages and positive effects of the present invention are: the intelligent home natural language control method of the present invention adopts the method of indoor positioning for the user, and judges the incomplete voice control command of the user according to the intelligent analysis of the user's position. The information contained implicitly improves the recognition ability of the voice control system for natural language, and avoids the inconvenience caused by the system not performing actions because it cannot recognize or asking more confirmation questions for users to confirm because it cannot recognize accurately Problems, and help to improve the user's fun and experience of using the smart home system.

Other features and advantages of the present invention will become more apparent after reading the detailed description of the embodiments of the present invention in conjunction with the accompanying drawings.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a kind of flow chart of the embodiment of the intelligent family natural language control method proposed by the present invention;

Fig. 2 is a schematic diagram of discount execution steps of the smart home natural language control method proposed by the present invention.

detailed description

Voice control is the most natural and convenient control method, and it is also a hot spot in technological development in recent years. The Siri voice control function of Apple mobile phones is considered to have opened a new round of artificial intelligence technology revolution. Through this intelligent voice system, users can read text messages, introduce restaurants, ask about the weather, and set alarm clocks by voice through their mobile phones. Siri can also call applications such as weather forecasts, schedules, and search materials that come with the system, and can continuously learn new voices. and intonation, to provide conversational responses.

Domestic voice technology manufacturer Ke launched the voice assistant "Lingxi". Say "Central Set" to Lingxi, and the TV channel will automatically switch to Central Set; say "Make me a cup of coffee" to Lingxi, and the coffee machine will start working; Lingxi said, "Turn down the temperature by two degrees", and the air conditioner will start to adjust the temperature... At present, many home appliance manufacturers are cooperating with Internet hardware manufacturers to explore the application of voice technology in smart home scenarios. Some simple voice control recognition can already be realized, but In the smart home industry, due to the wide variety of controlled appliances involved, combined with the current application environment of user control, many scenarios and many possibilities are involved. Therefore, it brings great difficulties and challenges to the voice recognition system. Challenges. The current voice recognition control needs to issue clear and complete voice control commands to specific controlled objects, or fuzzy recognition of some incomplete commands. However, it is impossible to fully realize the recognition and judgment of the user's natural language. Based on the above, the present invention proposes a smart home natural language control method.

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Embodiment 1, this embodiment proposes a smart home natural language control method, including the following steps:

S1. The voice receiving device receives voice control commands;

S2, the control unit analysis command;

S3, execute the command;

In said step S2, the following sub-steps are also included:

S21. Analyze the completeness of the voice control command. If the voice control command includes the complete electrical appliance type and the action command for control execution, then directly enter step S3; otherwise, enter step S22;

S22. If the type of electrical appliance is not specified in the voice control command, locate the user's position, find out the electrical appliance closest to the user according to the user's location, and judge that the electrical appliance is the specified electrical appliance, and execute step S3.

A significant challenge to speech recognition in the smart home is the ellipsis of language. There are a lot of known preconditions hidden in our natural language. For example, does "turn down a little bit" mean the temperature is lowered, or the curtains are lowered, or the brightness of the screen is lowered? In daily life, people may know which device the speaker wants to turn down through a glance, but smart home appliances are not yet able to perceive this expressive expression. The smart home natural language control method of the present invention adopts the method of indoor positioning of the user, and judges the information implicitly included in the user's incomplete voice control command according to the intelligent analysis of the user's position, thereby improving the recognition of the natural language by the voice control system The ability avoids the system not performing actions due to inability to recognize or raises more confirmation questions for users to confirm due to inability to recognize accurately, which brings inconvenience to users and reduces the user's enjoyment and feeling of the smart home system. According to the judgment of incomplete voice control commands based on the user's positioning method, the hidden Markov model is actually used to solve the problem of language simplification. With the help of indoor positioning information, if the user is closer to the air conditioner, the command is considered to be sent to the air conditioner. ; If the user is closer to the window, the command is considered to be addressed to the window. In fact, this is also completely in line with the normal usage habits of human beings, which helps to improve the accuracy of speech judgment.

So how does the air conditioner know that "bigger" means "cooler"? Or "hotter"? Or a little more ventilation? It is not enough to only rely on the user's location information, but also need to use the hidden Markov model algorithm to judge.

According to people's logical reasoning method, we can infer the real intention of the user to say "bigger" in this way:

1. If the window has just been opened, then "bigger" should refer to the window being opened larger.

2. If you have just turned on the air conditioner, it also depends on whether the air conditioner is cooling or heating. When cooling, it means "cooler"; when heating, it means "hotter". Generally speaking, the possibility of adjusting the ventilation volume of the air conditioner is not high.

3. If the user has just adjusted the ventilation volume of the air conditioner, the possibility of requiring the ventilation volume to be adjusted again becomes very high. "Bigger" at this point most likely means a little more ventilation.

Turning this process into something a computer can do requires the use of Hidden Markov Models. Hidden Markov model is a mathematical model like the following, also known as "five-tuple":

S: set of hidden states;

N: observation state set;

A: Transition probability matrix between hidden states;

B: output matrix (that is, the probability of hidden state to output state);

PI: initial probability distribution (the initial probability distribution of the hidden state);

Taking the bedroom as an example, we can define a quintuple as follows:

S={window opening, cold air conditioner, hot air conditioner, air volume of air conditioner}

N={

Window opening: {cooler: 0.0, warmer: 0.0, larger: 0.5, smaller: 0.5},

Air-conditioning refrigeration: {Colder: 0.4, hotter: 0.0, larger: 0.3, smaller: 0.3},

Air conditioning heating: {colder: 0.0, warmer: 0.4, larger: 0.3, smaller: 0.3},

Air-conditioning air volume: {cooler: 0.0, hotter: 0.0, larger: 0.5, smaller: 0.5},

}

A={

Window opening: {window opening: 0.7, air conditioning cooling: 0.1, air conditioning heating: 0.1, air conditioning air volume: 0.1},

Air conditioner cooling: {window opening: 0.1, air conditioner cooling: 0.5, air conditioner heating: 0.3, air volume: 0.1},

Air conditioning heating: {window opening: 0.1, air conditioning cooling: 0.3, air conditioning heating: 0.5, air conditioning air volume: 0.1},

Air-conditioning air volume: {window opening: 0.1, air-conditioning cooling: 0.2, air-conditioning heating: 0.2, air-conditioning air volume: 0.5},

}

PI={window opening: 0.0, air conditioner cold: 0.0, air conditioner heat: 0.0, air volume: 1.0}

Among them, in the N of the quintuple, the observation state set represents the probability of each operation occurring. If the user wants to operate the opening of the window, he must not say "cooler" or "hotter", so these two The probability is 0, and the user has half the possibility to say "bigger" or "smaller".

In A of the quintuple, the transition probability matrix between hidden states refers to the probability of these four states recurring after each state in the S set. Taking the window opening as an example, it is known that the user’s last command was to adjust If the window opening is changed, the probability that this command will adjust the window opening is 70%, and the probability of adjusting the cooling, heating, and air volume of the air conditioner is very low, all of which are 10%.

PI is the initial probability distribution, which refers to the last judgment result. We assume that the last calculation result is to adjust the air volume of the air conditioner.

After the above three arrays are defined, then the Viterbi algorithm (Viterbi algorithm) can be used to obtain the calculation result. The Viterbi algorithm is an algorithm commonly used in hidden Markov models, and details of the algorithm will not be repeated in this embodiment.

For example, to set the input sequence:

Input={bigger, smaller, cooler, bigger}

After calculation by the Viterbi algorithm, the output results are as follows:

Output={Air Conditioning Air Volume, Air Conditioning Air Volume, Air Conditioning Refrigeration, Air Conditioning Refrigeration}

The calculation result is completely consistent with the previous logical reasoning result.

As a preferred embodiment, at the beginning of the smart home natural language control method, it also includes the step of setting user rights, collecting the voice characteristics of different users of family members, and assigning corresponding user rights levels to them, and establishing user rights grade table;

Before entering step (3) in said step (21), it also includes the step of analyzing command conflicts,

a. If voice control commands from users with different authority levels are received at the same time or within a limited time, execute the voice control commands with higher authority levels;

b. If different voice control commands of users at the same authority level conflict with each other, judge whether the conflicting voice control commands come from the same user, and if they come from the same user, send a voice prompt for user confirmation. This step is to prevent users from having a tendency to joke about smart home control. Some users may use voice commands too casually out of curiosity. degree" command. If the voice recognition system strictly follows the user's command, it may damage the air conditioning equipment. Therefore, the system needs to issue a prompt for the user to confirm, which is conducive to prompting the user to return to rationality and improving the seriousness of voice control.

This step increases the degree of intelligence by assigning different permissions to different users. For example, the system administrator has the highest authority to completely control the system, and the parent user is set by the system administrator. At the same time, the parent user can also designate a new parent user. , the power of guardian users is less than that of parent users. When the orders of guardian users and parent users conflict, the system will give priority to executing the orders of parent users. Guest users are mainly guests who visit the home for a short time, or housekeeping service personnel. The smart home system provides visitors with Minimal convenience utilities and lighting services. Of course, the voice feature information of the guest user may not be collected and stored in advance, and if the system cannot find a user matching the voice feature information in the stored data, it will be collectively classified as a guest user. By limiting the level of authority, for example, guardian users are generally young, unsound, drunk, in a bad mood, or people with mental illness, can limit their control authority to 1, and cannot use commands that violate the rules, such as homework time Ask to turn on the TV or open the door to go out and play. 2. Commands that affect neighbors cannot be used, such as turning the room stereo to the maximum volume at night. For another example, visitors are generally guests, housekeeping service personnel, etc., and their control rights can be limited to: 1. Voice commands related to family privacy cannot be used, such as opening the family mailbox. 2. Cannot set or modify system parameters. 3. Except for lights, air conditioners, and water heaters, no other household appliances are allowed. By adopting the above restrictions, the management and control of the main user is facilitated, and at the same time, corresponding services can be intelligently provided, which brings great convenience to the user.

In this embodiment, a method of how to deal with voice control command conflicts from the same authority level is also provided. In the step b of analyzing command conflicts, if the conflicting voice control commands do not come from the same user, Then continue to detect and receive voice control commands, if the voice control commands that conflict with each other continue to occur at a high frequency, then it is judged that there are differences in the opinions of the two users, then the method of executing commands at a discount is adopted to delay the implementation of new voice control commands, and Each time it is executed, the delay time is increased accordingly until conflicting voice control commands no longer appear. This step can solve the problem of conflict between parents and users. For example, what should I do if the host and hostess quarrel? The hostess says "turn on the air conditioner", but the hostess says "turn off the air conditioner", what should I do? In this embodiment, the "best-for-good" solution is adopted—executing the completely opposite order at a discount until one party gives up. The schematic diagram of "discounted execution" is shown in Figure 2. In short, the more "tossing" between A and B, the smaller the slope of the line segment representing the command execution process, which means that the response of the system is getting slower and slower. Parameters are discounted to an increasing degree. After A and B stop arguing, the system will return to the normal state of rapid response and thorough execution when the system delays for 10 seconds to confirm that there are no more opposing orders.

In addition, this embodiment also provides the step of judging a prank. In the step b of analyzing command conflicts, if the conflicting voice control commands do not come from the same user, and the conflicting voice control commands If the command repetition frequency is lower than the set limit, it is judged that one of the users has mischievous control, then the user's position is located, and the user who is closest to the controlled appliance is found according to the user's position, and the user's voice control command is executed . For example, the hostess is taking a shower in the bathroom, and the host is shouting "stop supplying hot water" outside. At this time, it can be judged whether to execute this order according to the position of the host and hostess. Indoor positioning information is often the best decision support data for judging whether an act is mischievous or not. This embodiment proposes the simplest indoor positioning method, by binding the IPV6 address of the electrical appliance with its spatial location into a table, as shown in Table 1:

ID Appliance IPV6 address Appliance Type Location information of home appliances 1 2001:0DB8::1428:0950 television Master bedroom, east wall, center 2 2001:0DB8::1428:0951 air conditioner Master bedroom, east wall, south side 3 2001:0DB8::1428:0952 Electric window Master bedroom, south wall, central ... ... ... ...

Table 1 IPV6 address and location comparison table of smart home appliances

Since there are enough IPV6 addresses, there is no need to worry about address exhaustion. Generally speaking, the main home appliances are installed against the wall, ceiling or floor, so it is not necessary to use a Cartesian coordinate system to accurately describe their spatial positions, which can greatly reduce the amount of data.

The present invention is not limited to this indoor positioning method. The positioning information formed by other indoor positioning methods, such as smart floor, Bluetooth beacon positioning, sound source positioning, and wifi positioning, can be used to participate in the calculation based on the above position information.

Preferably in this embodiment, in the step S22, the method for locating the user's position is: use IPV6 address positioning, bind the IPV6 address of each electrical appliance with the actual space address where it is located, and connect each voice receiving device Bind the IPV6 address with the actual space address where it is located, judge the location of the user according to the IPV6 address of the voice receiving device receiving the voice control signal, and then correspondingly control the electrical appliance closest to the user's location to execute the corresponding control command. The binding manner can be referred to as shown in Table 1.

Further, since multiple voice receiving devices may be installed indoors, in the step S22, if multiple voice receiving devices receive the voice control command at the same time, then according to the intensity of the voice signal received by each voice receiving device, compare the received voice signal The voice receiving device with the strongest strength, and judge the location of the user based on the IPV6 address of the voice receiving device.

Preferably, before the step S1, a step of presetting a trigger word for triggering the voice control operation is further included, and only when the user utters the voice of the trigger word, step S1 or the steps after it are executed. This is to prevent the user's daily conversations from being separated from the voice control commands sent to the smart home. Only when the user enables the voice control function, the system will analyze and execute the voice.

In daily life, it is possible that the voice control command sent by the user does not contain specific parameter information, for example, "increase the temperature of the air conditioner", the system only receives the command to increase the temperature of the air conditioner, but does not know how to adjust it specifically. How many degrees, for another example, "turn on the TV", the system does not know which station should be turned on, therefore, in order to be able to handle the above situation, in the step S3, if the voice command does not contain specific parameter information, then according to the default Parameters execute user commands, and each appliance is set with default parameters. Taking TV as an example, the default parameter of "Watch TV" is to watch the station where the TV was last turned off. If that station has no signal, choose the station with the highest viewing frequency. Taking an air conditioner as an example, the default parameter is to add 1 degree or subtract 1 degree. Other types of smart home appliances or smart devices should set appropriate default parameters.

In summary, the smart home natural language control method of this embodiment, combined with indoor positioning technology and hidden Markov principle, can judge incomplete voice control commands according to people's normal habits and logical thinking, improving the accuracy of judgment rate, and can judge a variety of different application control environments, avoiding user mischief and causing damage to electrical appliances, and at the same time giving a solution to user command conflicts, making smart home control more intelligent and more in line with human Habits of logical thinking and living habits, and at the same time, there are very few links in the whole control method where the system needs to ask the user questions, which will not bring more inconvenience to the user.

Of course, the above descriptions are not intended to limit the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention shall also belong to protection scope of the present invention.

Claims (8)

1. a wired home natural language control method, comprises the following steps: (1), pronunciation receiver receive voice control command；(2), control unit analysis order；(3), order is performed；It is characterized in that, in described step (2), also include following sub-step:

(21), the integrity degree of analyzing speech control command, if voice control command comprising complete appliance type and controlling the action command performed, be then directly entered step (3)；Otherwise, step (22) is entered；

(22) if not specified appliance type in voice control command, then the position of user is positioned, find out the electrical equipment of distance users minimum distance according to customer location, and judge that this electrical equipment is as specifying electrical equipment, and perform step (3)。

2. wired home natural language control method according to claim 1, it is characterized in that, when described wired home natural language control method is initial, also include the step setting user right, gather the phonetic feature of kinsfolk's different user, and distribute corresponding user right grade for it, set up user right table of grading；

Described step (21) enters before step (3), also includes the step analyzing command collision,

If the while of a or when receiving the voice control command with different rights class user in limiting time, then perform the voice control command that Permission Levels are high；

If b is positioned at the different phonetic control command of same Permission Levels user when colliding with each other, it is judged that whether this conflicting voice control command comes from same user, if coming from same user, then sending voice message, allowing user confirm。

3. wired home natural language control method according to claim 2, it is characterized in that, in the step b of described analysis command collision, if this conflicting voice control command is not from same user, then continue detection and receive voice control command, occur if described conflicting voice control command continues high frequency, then it is judged as that two consumers' opinions exist difference, then adopt the exectorial method that gives a discount, time delay performs new voice control command, and often perform once, the time of time delay accordingly increases, until conflicting voice control command no longer occurs。

4. wired home natural language control method according to claim 3, it is characterized in that, in the step b of described analysis command collision, if this conflicting voice control command is not from same user, and described conflicting voice control command repetition rate is lower than setting limit value, then it is judged as that one of them user exists trick control, then the position of user is positioned, find out, according to customer location, the user that distance controlled electric appliance is closest, and perform the voice control command of this user。

5. the wired home natural language control method according to any one of claim 1-4, it is characterized in that, in described step (22), the method that the position of user is positioned is: adopt IPV6 address location, the real space address of the IPV6 address of each electrical equipment Yu its place is bound, and the real space address of the IPV6 address of each pronunciation receiver Yu its place is bound, IPV6 address according to the pronunciation receiver receiving audio controls judges user's location, then the electrical equipment that corresponding this customer location of command range is nearest performs corresponding control command。

6. wired home natural language control method according to claim 5, it is characterized in that, in described step (22), if multiple pronunciation receiver are simultaneously received the words of voice control command, the intensity of voice signal is then received according to each pronunciation receiver, compare and receive the pronunciation receiver that voice signal intensity is the strongest, and judge user's location with the IPV6 address of this pronunciation receiver。

7. the wired home natural language control method according to any one of claim 1-4, it is characterized in that, before described step (1), also include the preset step triggering vocabulary for triggering voice-controlled operations, when user sends the voice of this triggering vocabulary, just perform step (1) or step after。

8. the wired home natural language control method according to any one of claim 1-4, it is characterized in that, in described step (3), if in voice command without design parameter information time, then performing user command according to default parameters, each electrical equipment is provided with default parameters。