KR102620070B1 - Autonomous articulation system based on situational awareness - Google Patents

Abstract

It includes a user terminal that provides situational recognition factors and receives real-time conditions, and a server that receives the real-time conditions from the user terminal, stores them, and matches the corresponding utterance, wherein the server receives the real-time conditions from the user terminal and matches the corresponding utterance. A condition detection module that receives real-time conditions and generates a condition list, a situation awareness module that compares the condition list with the real-time conditions to generate situation awareness information, and an output module that autonomously selects the situation awareness information through the user terminal. It is possible to provide an autonomous ignition system according to situation recognition, characterized in that it includes.

Description

Autonomous articulation system based on situational awareness}

The present invention relates to an autonomous speech system based on situation awareness. More specifically, the present invention relates to a situation awareness system that classifies the user's condition through the user terminal, dynamically selects a conversation topic according to the condition, and selects a conversation topic through text, mobile push, and voice. It is about an autonomous ignition system according to .

Artificial intelligence is one of the fields that is attracting attention with the advent of the 4th industrial revolution. Currently, it is limited to certain fields such as Google's Alphago, AI with intelligence close to expert, or AI that carries out user commands such as Apple's Siri or Samsung's Bixby. AI, such as secretaries, is mainly developed.

This type of assistant-like AI recognizes the user's commands, recognizes and recognizes the situation the user is in, and provides a service that actively provides the most appropriate and useful information to the user.

However, the method in which a conventionally commercialized AI, such as a secretary, can recognize a user is only possible through voice recognition for a specific word, and if it is not recognized through this method, the AI, such as a secretary, takes a passive attitude and does not take any action. There is a limit to showing.

For example, in the case of Apple's Siri or Samsung's Bixby, the system uses a system in which the user mentions a specific word that activates the AI, then adds the desired service, recognizes the appropriate user command, and makes the AI speak.

However, because this passive speech system waits for the user to call an AI such as a secretary first, there was an inconvenience in that the user had to perform the process of calling the AI every time to activate the AI.

Therefore, there was an urgent need to develop an autonomous ignition system based on situational awareness with a pre-ignition system.

As prior art, Korean Patent Publication No. 10-2071-0117364 “Method of providing artificial intelligence assistant and voice recognition equipment used therefor” is disclosed.

The present invention was invented to solve the problems described above. It recognizes the user's condition, generates situational awareness information accordingly, and selects conversation topics according to the situational awareness information, allowing the user to activate AI. In order to do this, the purpose is to provide an autonomous speech system based on situational awareness that omits the process of calling AI each time and thus increases user convenience.

To solve the above problems, an autonomous speech system based on situation awareness according to an embodiment of the present invention provides situation recognition factors, a user terminal that receives real-time conditions, and a user terminal that receives the real-time conditions from the user terminal, stores and stores them. It includes a server that matches the corresponding utterance, wherein the server includes a conversion module that converts the real-time condition into condition information, a condition detection module that receives the condition information and generates a condition list, and configures the condition list with the real-time condition. It may include a context awareness module that compares and generates situation awareness information, and an output module that autonomously selects the situation awareness information through the user terminal.

In addition, the condition detection module receives user information such as GPS, illuminance sensor, camera, microphone sensor, and stored messages mounted on the user terminal as input, converts it into a digital signal, and analyzes the digital signal to recognize the user's real-time condition. You can create the condition list by doing this.

In addition, the condition detection module is implemented every hour or minute, and may further include a scheduler that compares the pre-stored condition list and the real-time condition conditions to determine whether the real-time condition satisfies the situation awareness information generation conditions. .

Additionally, when comparing the conditions of the condition list and the real-time condition, the scheduler may divide the conditions into overall conditions and individual conditions.

Additionally, when comparing the condition list and the real-time condition, the situation awareness module may pattern the input real-time condition and template all voice and text output from the situation-aware autonomous speech system.

Additionally, when making a first speech, the output module may select a conversation topic that matches each real-time condition from the condition list based on the recognition information and make the speech.

In addition, when the output module makes a preliminary speech, the conversation topic is connected to a plurality of condition information in the condition list, and the condition information can also be connected to a plurality of conversation topics, so that even in the same condition information, other conversations are possible. A topic may arise.

Additionally, when making a first speech, if there are multiple conversation topics linked to the condition information, the output module may randomly output one of the multiple topics.

In addition, the output module may be characterized in that, when speaking, the uttered conversation topic is involved in generating the situation awareness information, thereby influencing the conversation topic uttered next.

In addition, the autonomous speech system according to the situation recognition collects data from a plurality of user terminals including the user terminal, selects the most frequently uttered conversation topic for each of the condition information, and utters the conversation topic. When doing so, it may further include a recommendation module that recommends and utters options related to the conversation topic.

In addition, when the conversation topic is uttered, the recommendation module can differentiate it based on the number of utterances, assign a recommendation grade to the options, grade them, and display them on the user terminal.

The autonomous ignition system based on situation awareness according to an embodiment of the present invention can increase user convenience by igniting based on situation awareness information.

In addition, better quality service providers can be presented to users through a recommendation method based on the number of utterances.

1 is a configuration diagram of an autonomous ignition system based on situation recognition according to an embodiment of the present invention.
Figure 2 is a perspective view showing a real-time condition collection method.
Figure 3 is a configuration diagram showing the server of Figure 1.
Figure 4 is a configuration diagram showing a situation awareness information generation method.
Figure 5 is an example diagram showing the selection method of the output module.
Figure 6 is a configuration diagram of an autonomous speech system based on situation recognition according to an embodiment of the present invention including a recommendation module.
Figure 7 is an example diagram showing selection creation through a recommendation module according to an embodiment of the present invention.
Figure 8 is a perspective view showing a recommendation rating displayed on a user terminal.

Hereinafter, the description of the present invention with reference to the drawings is not limited to specific embodiments, and various changes may be made and various embodiments may be possible. In addition, the content described below should be understood to include all conversions, equivalents, and substitutes included in the spirit and technical scope of the present invention.

In the following description, the terms first, second, etc. are terms used to describe various components, and their meaning is not limited, and is used only for the purpose of distinguishing one component from other components.

Like reference numerals used throughout this specification refer to like elements.

As used herein, singular expressions include plural expressions, unless the context clearly dictates otherwise. In addition, terms such as “comprise,” “provide,” or “have” used below are intended to designate the presence of features, numbers, steps, operations, components, parts, or a combination thereof described in the specification. It should be construed and understood as not precluding the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with the meanings they have in the context of the related technology, and should not be interpreted as having ideal or excessively formal meanings, unless explicitly defined in the present application. No.

In addition, when describing with reference to the accompanying drawings, identical components will be assigned the same reference numerals regardless of the reference numerals, and overlapping descriptions thereof will be omitted. In describing the present invention, if it is determined that a detailed description of related known technologies may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying FIGS. 1 to 8.

FIG. 1 is a configuration diagram of an autonomous ignition system based on situation recognition according to an embodiment of the present invention, FIG. 2 is a perspective view showing a real-time condition collection method, FIG. 3 is a configuration diagram showing the server of FIG. 1, and FIG. 4 is a configuration diagram showing the situation awareness information generation method, and Figure 5 is an example diagram showing the selection method of the output module.

First, referring to FIGS. 1 to 3, an autonomous speech system based on situation recognition according to an embodiment of the present invention may include a user terminal 100 and a server 200.

The user terminal 100 may be equipped with GPS, an illumination sensor, a camera, a microphone sensor, etc. that can receive external stimuli as input.

For this reason, the user terminal 100 is preferably formed as a PC or smartphone type terminal equipped with various types of hardware sensors and capable of processing them with software, but is not limited thereto.

At this time, the user terminal 100 can receive input of the real-time condition 110.

Specifically, referring to FIG. 2, the real-time condition 110 includes both flexible and fixed conditions such as the user's age, gender, current date, time, weather, illumination, and the user's facial expression, which are input through the user terminal 100. It can be included.

In addition, the real-time condition 110 can determine the user's current situation by identifying fluid conditions in real time.

For example, the real-time condition 110 may be recognized as a different type of input depending on changes in GPS and input into the user terminal 100.

Accordingly, the real-time condition 110 may be recognized and input into the user terminal 100 in different formats according to changes in flexible factors such as location, weather, and time.

At this time, the real-time condition 110 can be determined through multiple conditions.

For example, the real-time condition 110 can be formed by combining multiple conditions, such as ‘sea in rainy weather’, ‘sea in snowy weather’, and ‘sea in sunny weather’.

At this time, the real-time condition 110 may be formed by combining various conditions as well as the weather and location mentioned above.

In addition, the real-time condition 110 can be input through manual and automatic methods.

At this time, the real-time condition 110 can be automatically input into the user terminal 100 by continuously observing the user's condition.

Additionally, the real-time condition 110 can be directly input through the user terminal 100 through a manual method.

Here, the method of directly inputting the real-time condition 110 can be interpreted as the user inputting text or uttering a voice through text or voice.

The server 200 can receive the real-time condition 110 from the user terminal 100, store it, and match the corresponding utterance.

Specifically, referring to FIG. 3 , the server 200 may further include a conversion module 210, a condition detection module 220, a situation recognition module 230, and an output module 240.

The conversion module 210 can convert the real-time condition 110 into condition information 211.

The condition detection module 220 receives user information 221 such as GPS, illuminance sensor, camera, microphone sensor, and stored messages mounted on the user terminal 100 as input and converts it into a digital signal 222 to create a condition list ( 223) can be created.

At this time, the user information 221 may be formed in various forms such as voice, light amount, sound wave, and text.

The digital signal 222 is generated by converting the user information 221 through an existing modulation method, using a method similar to that performed through software or hardware of existing smartphone terminals and PC terminals. Detailed explanation will be omitted.

The condition list 223 may receive a real-time condition 110 and include a plurality of condition information 211.

At this time, the condition list 223 can convert the input real-time condition 110 into condition information 211, add it to the existing condition list 223, and store it.

Accordingly, the condition list 223 can continuously increase the number of items in the list.

Meanwhile, the condition detection module 220 is implemented every hour or minute, and may further include a scheduler 224 that compares the conditions of the pre-stored condition list 223 and the real-time condition 110.

delete

delete

delete

Accordingly, when the scheduler 224 is executed every hour or minute, accuracy and reliability can be increased when comparing the condition list 223 and the real-time condition 110.

The situation awareness module 230 may generate situation awareness information 231 by comparing the condition list 223 with the real-time condition 110.

At this time, when comparing the condition list 213 and the real-time condition 110, the situation recognition module 220 patterns the input real-time condition 110 and all the voices and voices output by the situation-aware autonomous speech system 1. Text output can be templated.

Accordingly, the situation recognition module 230 can improve comparison speed by patterning the input real-time condition 110 and comparing and analyzing it with condition information 211 of a similar pattern stored in the condition list 223.

Additionally, the situation awareness module 230 can output frequently used voices and texts more quickly by templateizing the output.

At this time, the template represents a form used to serve as a guide when collecting frequently used commands or creating something in a program, and is not difficult for a person skilled in the art to understand, so a detailed explanation will be omitted.

Meanwhile, referring to FIG. 4, when comparing the condition list 223 and the real-time condition 110, the situation awareness information 231 determines whether and to what degree it matches the condition information 110 stored in the condition list 223. It may be created differently depending on the

For example, when detecting GPS movement due to a change in real-time conditions 110, the situation awareness information 231 may identify condition information 211 having similar conditions among the condition list 223.

At this time, the situation awareness information 231 may be formed in different forms by determining whether the situation is the same as the stored condition information 211, a similar situation, or a different situation based on the input real-time condition 110.

At this time, the situation awareness information 231 may be generated through one of the stored condition information 211 when the situation is the same.

Additionally, the situation awareness information 231 may be formed through the remaining condition information 211 excluding the condition information 211 for the same situation in similar and different situations.

The output module 240 can autonomously select the situation awareness information 231 through the user terminal 100.

At this time, the output module 240 can turn the pre-ignition function ON or OFF.

It is preferable that the output module 240 utters the conversation topic through voice when the pre-speech function is turned on, and transmits it in the form of text through the notification function of the user terminal 100 when the pre-speech function is turned off.

In detail, referring to FIG. 5, the output module 240 may select and utter a conversation topic 241 that matches the situation awareness information 231 when making the first utterance.

Additionally, when the output module 240 makes a selection, the conversation topic 241 may be connected to a plurality of situation awareness information 231.

In addition, the output module 240 may be connected to the situation awareness information 231 and a plurality of conversation topics 241 when making a first speech.

Accordingly, the output module 240 can select a different conversation topic 241 even in the same real-time condition 110.

At this time, when making the first speech, the output module 240 may randomly output one of the plurality of conversation topics 241 connected to the situation awareness information 231.

As a result, when speaking through the output module 240, the number of speech options increases, allowing more diverse conversations to be held with the user.

In addition, when the output module 240 utters a conversation, the uttered conversation topic 241 participates in the creation of the situation awareness information 231, thereby influencing the next uttered conversation topic 241.

Figure 6 is a configuration diagram of an autonomous speech system based on situation recognition according to an embodiment of the present invention including a recommendation module, and Figure 7 is an exemplary diagram showing selection generation through a recommendation module according to an embodiment of the present invention. Figure 8 is a perspective view showing a recommendation level displayed on a user terminal.

First, referring to FIGS. 6 and 7 , the autonomous speech system 1 based on situation recognition according to an embodiment of the present invention may further include a recommendation module 300.

The recommendation module 300 may collect data from a plurality of user terminals 100, including the user terminal 100.

At this time, the recommendation module 300 may collect and store data of users using the autonomous speech system 1 according to situation recognition according to an embodiment of the present invention through a network connection method.

Additionally, the recommendation module 300 may select the most frequently uttered conversation topic 241 for each condition information 211 through the collected user data.

At this time, when uttering the most selected conversation topic 241, the recommendation module 300 may recommend and utter an option 310 related to the conversation topic.

At this time, it is desirable to recommend the choice 310 in the form of a place such as a restaurant, amusement facility, cafe, or tourist attraction.

Additionally, the options 310 may change according to real-time conditions 110, such as time, weather, and date, as the recommendation module 300 analyzes the collected user data.

At this time, it is preferable to use a large website platform or a database (D) provided by companies as the database of choice 310, but it is not limited to this and a database built directly can be used.

Next, referring to FIG. 8, when the conversation topic 241 is uttered, the recommendation module 300 differentiates it based on the number of utterances, assigns a recommendation grade 320 to the option 310, grades it, and sends it to the user terminal. It can be written as (100).

The recommendation grade 320 may be generated by combining the amount of speech of the conversation topic 241 and user satisfaction.

At this time, user satisfaction is determined by detecting changes in the real-time condition 110 and determining whether to visit the choice 310, and a notification window may be generated on the user terminal 100 of the user visiting the choice 310.

In addition, user satisfaction is displayed on the user terminal 100 in the form of a notification window, so that the user can directly determine the recommendation level 320 of the option after using the option 310 and score it.

At this time, the recommendation grade 320 is formed in a form with excellent visibility, like a star model, and is located on one side of the selection 310 so that the selection 310 and the recommendation grade 320 can be seen at a glance. desirable.

Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art can realize that the present invention can be implemented in other specific forms without changing the technical idea or essential features of the present invention. You will be able to understand it. Therefore, the embodiments described above are all

It is illustrative in that respect and is not limited.

1: Autonomous ignition system based on situation awareness
100: user terminal
110: Real-time condition
200: server
210: conversion module
211: Condition information
220: Condition detection module
221: User information
222: digital signal
223: Condition list
224: Scheduler
230: Situation awareness module
231: Situational awareness information
240: output module
241: Topic of conversation
300: Recommended module
310: Options
320: Recommended rating
D: database

Claims (11)

It provides situational awareness factors and is equipped with GPS, illuminance sensor, camera, and microphone sensors to provide real-time information including dynamic and fixed conditions, including the user's age, gender, current date, time, weather illuminance, and the user's facial expression. A user terminal that receives input of conditions;
A server that receives the real-time condition from the user terminal, stores it, and matches the corresponding utterance, and
Recommendation that collects data from a plurality of user terminals, including the user terminal, selects the most frequently spoken conversation topic for each condition information, and recommends and utters options related to the conversation topic when the conversation topic is uttered. Contains modules,
The server is,
a conversion module that converts the real-time condition into condition information;
a condition detection module that receives the condition information and generates a condition list;
A situation awareness module that generates situation awareness information by comparing the condition list with the real-time conditions, and
It includes an output module that autonomously selects the situation awareness information through the user terminal,
The output module is,
When speaking first, a conversation topic is selected and uttered based on the above situational awareness information,
The recommended module is,
When the conversation topic is uttered, it is differentiated based on the number of utterances, a recommendation grade is given to the options, a grade is given, and this is displayed on the user terminal,
The above options are:
It is in the form of a place such as a restaurant, amusement facility, cafe, or tourist attraction, and changes according to the real-time conditions,
The above recommendation level is,
An autonomous speech system based on situational awareness, characterized in that it is generated by combining the amount of speech of the conversation topic and user satisfaction that determines whether to visit the option by detecting changes in the real-time condition.
According to clause 1,
The condition detection module is,
Receives user information such as GPS, illuminance sensor, camera, microphone sensor, and stored messages mounted on the user terminal as input, converts them into digital signals, and analyzes the digital signals to recognize the user's real-time condition to generate the condition list. An autonomous ignition system based on situation recognition, characterized by:
According to clause 1,
The condition detection module is,
It is performed every hour and minute, and further includes a scheduler that compares the pre-stored condition list and the real-time condition conditions and determines whether the real-time condition satisfies the situation awareness information generation conditions. Autonomous speech according to situation awareness. system.
delete According to clause 1,
The situation awareness module is,
When comparing the condition list and the real-time condition, the input real-time condition is patterned and all voice and text output from the situation-aware autonomous speech system is templated.
delete According to clause 1,
The output module is,
When speaking first, the conversation topic is connected to multiple situation awareness information, and the situation awareness information can also be connected to multiple conversation topics, so even when speaking through the same situation awareness information, differences in the conversation topic occur. An autonomous ignition system based on situation recognition, characterized by:
According to clause 1,
The output module is,
When speaking first, if there are multiple conversation topics connected to the situation awareness information, an autonomous speech system based on situation awareness, characterized in that one of the plurality is randomly output.
According to clause 1,
The output module is,
An autonomous speech system based on situation awareness, characterized in that when speaking, the uttered conversation topic is involved in generating the situation awareness information, thereby influencing the conversation topic uttered next.
delete delete

Images