KR20120044911A - Affect producing servece providing system and method, and device for producing affect and method therefor - Google Patents

Abstract

PURPOSE: A system for providing an emotion direction service, a method thereof, a device thereof, and a method for directing emotions thereof are provided to use an intelligent emotion inferring module to infer the emotional state of a user using a probabilistic model to adaptively express the emotions. CONSTITUTION: A directing information storing server(120) stores emotion directing information corresponding to each emotion class classified in an emotion word dictionary. An emotion inferring server(130) infers the emotion matched to the emotion word dictionary. A direction information extracting server(140) extracts emotion directing information corresponding to the emotion from the directing information storing server.

Description

Affective Producing Servece Providing System and Method, and Device for Producing Affect and Method therefor}

An embodiment of the present invention relates to a system for providing emotional expression service, a method of providing the same, an apparatus capable of emotional expression and an emotional expression method thereof. More specifically, by using the intelligent emotional reasoning module that infers the emotional state of the user by using the probabilistic model method and adaptively expresses the emotion according to the inferred result, the emotional terminal intelligently affects the other terminal connected to the network. Emotional expression service providing system capable of directing and a method of providing the same, and an apparatus capable of emotional expression and the emotional expression method.

Recently, the spread of the Internet has been widely extended to the wireless Internet, so that users can move not only through a wired computer but also by using a mobile communication terminal such as a PDA (Personal Digital Assistant), a notebook, a mobile phone, a smartphone, and the like. In the meantime, it is possible to communicate with users of other wired or wireless communication terminals. Such wired and wireless communication is not just an exchange of voice signals or data files, but a messenger is used to communicate with other users by text or visit a blog of himself or another communication user to create a text record, image or Other activities, such as uploading videos, form a new community online.

As described above, it is often necessary to express one's emotional state to another user or to guess the emotional state of another user in the communication activity in the community formed online. To this end, online community service providers are able to express or guess the emotional state of users in various ways. For example, a community service provider using a messenger provides various emoticon selection menus corresponding to emotional states, and allows a user to select an emoticon according to his or her emotional state so that the emotional state of the user is displayed through a chat window. . In addition, the user may search for a specific word in a sentence that the user inputs through a chat window or a bulletin board, and if the specific word is found, an icon corresponding to the specific word may be displayed so that the emotional expression is automatically performed according to the input of the sentence. .

However, human emotions are not always fixed, and are generally changed every time according to a situation, a place, or an atmosphere, and it is quite cumbersome for a user to select an emoticon each time to change the emotion that changes according to a situation or an environment. It's work.

In addition, emotions or feelings have personal properties, and the psychological factors that influence human emotions can be largely divided into surprise, fear, disgust, anger, joy, happiness, and sadness. However, even in the same situation, the psychological factors of the individual may be different, and the intensity of the expressed emotion may also vary according to the individual. Nevertheless, searching for a specific word in a sentence input by the user and uniformly expressing a problem does not provide an accurate expression of the current emotional state of the individual.

An embodiment of the present invention has been devised to solve the above-described problem, using an intelligent emotional inference module to infer the emotional state of the user using a probability model method and to express the emotion adaptively according to the inferred result It is an object of the present invention to provide an emotional rendering service providing system capable of intelligently expressing emotion to another terminal connected to a network, and a method of providing the same, and a device capable of expressing emotion and a method of expressing emotion.

Emotional expression service providing system according to an embodiment of the present invention for achieving the above object, the emotional word is classified into an item containing at least one of the emotion class, similarity, positive or negative, emotional strength emotional word dictionary Word dictionary storage server for storing as; A presentation information storage server for storing emotional presentation information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a moving picture, an image, and an audio signal; An emotion inference server for inferring an emotion matching an emotion word dictionary with respect to at least one of a text and an audio signal received from a user terminal connected through a network; A production information extraction server for extracting emotion production information corresponding to the emotion from the production information storage server; And an emotion directing server for receiving information on the destination terminal from the user terminal, and transmitting at least one of the text and voice signals received from the user terminal and the extracted emotion directing information to the destination terminal.

Here, the emotional reasoning server, a sentence converter for grasping the word and phrase for the text and converts it to the basic form; Matching confirmation unit for checking a word or phrase matching the emotional word dictionary of the converted words and phrases; And a probabilistic model based on the co-occurrence of the converted words and phrases, and an emotional inference unit for inferring emotion according to the applied probabilistic model.

In addition, the emotion reasoning server stores emotions corresponding to at least one of color, image, font, emoticon, avatar, video, voice, tone, sound quality, and sound wave, corresponding to each emotion class classified in the emotional word dictionary. Emotional storage unit; A sound source detector for detecting a sound source including at least one of a tone, sound quality, and sound wave from a voice signal; And an emotion inference unit that infers a matched emotion from the emotion storage unit based on the detected sound source.

According to an embodiment of the present invention, a communication terminal capable of intelligent emotional expression according to an embodiment of the present invention may be classified into an item including emotional class, similarity, positive or negative, or emotional strength for an emotional word. A word dictionary storage for storing the word dictionary; A production information storage unit for storing emotional presentation information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a moving picture, an image, and an audio signal; An emotion inference unit for inferring an emotion matching the emotion word dictionary from at least one of the received text and voice signal; Directing information extraction unit for extracting the emotional directing information corresponding to the emotion from the directing information storage unit; And an emotion directing unit for receiving information on the destination terminal and transmitting at least one of the received text and voice signals and the extracted emotion directing information to the destination terminal.

Here, the apparatus capable of intelligent emotional expression, sentence conversion unit for grasping the word and phrase for the text to convert to basic form; And a matching check unit for checking a word or phrase that matches the emotional word dictionary among the converted words and phrases, and the emotional inference unit based on the co-occurrence of the converted word and phrase (Probabilistic model). ), And emotion can be inferred according to the applied probability model.

In addition, the device capable of intelligent emotional expression, corresponding to at least one of the color, image, font, emoticon, avatar, video, voice, tone, sound quality, sound waves corresponding to each emotion class classified in the emotional word dictionary Emotional storage unit for storing the; And a sound source detector that detects a sound source including at least one of a tone, sound quality, and sound wave from the voice signal, wherein the emotion inference unit may infer a matched emotion from the emotion storage unit based on the detected sound source.

Emotional expression service providing method according to an embodiment of the present invention for achieving the above object, emotional word dictionary classified by the item containing at least one of the emotion class, similarity, positive or negative, emotional strength Storing emotion directing information corresponding to each emotion class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal; Inferring an emotion matched with an emotional word dictionary for at least one of a text and an audio signal received from a user terminal connected through a network; Extracting emotion directing information corresponding to the emotion from the stored emotion directing information; Receiving information about a destination terminal from a user terminal; And transmitting at least one of the text and voice signals received from the user terminal and the extracted emotion directing information to the destination terminal.

Here, the step of identifying words and phrases with respect to the text and converting to basic form; Identifying a word or phrase that matches the emotional word dictionary among the converted words and phrases; And applying a probabilistic model based on the co-occurrence of the converted word and phrase. In this case, the inference step can infer emotion according to the applied probability model.

The method may further include storing emotions corresponding to at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave, corresponding to each emotion class classified in the emotion word dictionary; And detecting a sound source including at least one of a tone, sound quality, and sound wave from the voice signal. In this case, the matched emotion may be inferred from the emotion storage based on the detected sound source.

The intelligent emotional expression method of the device capable of emotional expression to achieve the above-mentioned object is classified as an item containing emotional class, similarity, positive or negative, emotional intensity, and stored as an emotional word dictionary. Storing emotional presentation information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal; Inferring the emotion matching the emotional word dictionary from at least one of the received text and voice signal; Extracting emotion directing information corresponding to the emotion from the stored emotion directing information; Receiving information about a destination terminal; And transmitting at least one of the text and the voice signal and the extracted emotion directing information to the destination terminal.

Preferably, the step of identifying the words and phrases for the text to the basic form; Identifying a word or phrase that matches the emotional word dictionary among the converted words and phrases; And applying a probabilistic model based on the co-occurrence of the converted word and phrase. In this case, the emotion inference step may infer emotion according to the applied probability model.

The method may further include storing emotions corresponding to at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave, corresponding to each emotion class classified in the emotion word dictionary; And detecting a sound source including at least one of a tone, sound quality, and sound wave from the voice signal. In this case, the emotion inference step may infer a matched emotion from the stored emotion based on the detected sound source.

According to an embodiment of the present invention, by using an intelligent emotional reasoning module that infers the emotional state of the user by using a probability model method and adaptively expresses emotion according to the inferred result, Emotional expression can be made intelligently.

1 is a diagram schematically showing a system for providing emotional service according to an embodiment of the present invention.
2 is a diagram illustrating an example of an emotional word dictionary according to an embodiment of the present invention.
3 is a diagram illustrating an example of the configuration of the emotional reasoning server of FIG. 1.
4 is a diagram illustrating an example of emotional log information by the emotional log storage of FIG. 3.
5 is a diagram illustrating another example of the configuration of the emotional reasoning server of FIG. 1.
6 is a view showing an example of the configuration of the apparatus capable of emotional expression according to an embodiment of the present invention.
FIG. 7 is a flowchart illustrating a method for providing emotional service by the emotional service providing system of FIG. 1.
8 is a flowchart illustrating a method of emotional expression by the apparatus capable of emotional expression of FIG. 6.

Hereinafter, some embodiments of the present invention will be described in detail through exemplary drawings. In adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are assigned to the same components as much as possible even though they are shown in different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In addition, in describing the component of this invention, terms, such as 1st, 2nd, A, B, (a), (b), can be used. These terms are only for distinguishing the components from other components, and the nature, order or order of the components are not limited by the terms. If a component is described as being "connected", "coupled" or "connected" to another component, that component may be directly connected or connected to that other component, but between components It will be understood that may be "connected", "coupled" or "connected".

1 is a diagram schematically showing a system for providing emotional service according to an embodiment of the present invention. Referring to FIG. 1, the emotion rendering service providing system 100 may be connected to a wired communication terminal 160 or a wireless communication terminal 170 through a network 150. In addition, the emotional directing service providing system 100 includes a word dictionary storage server 110, directed information storage server 120, emotional reasoning server 130, directed information extraction server 140 and emotional directed server 145. can do. Here, although the emotional rendering service providing system 100 has been described as a plurality of servers connected to each other, it may be implemented as a single device consisting of a combination of several components.

The word dictionary storage server 110 stores the sentiment word as an emotional word dictionary by classifying the sentiment word into an item including at least one of an emotion class, similarity, positive or negative, and sentiment strength. Emotion is a property of feelings of stimulus or change of stimulus, and depends on psychological factors such as surprise, fear, disgust, anger, joy, happiness, and sadness. However, even for the same stimulus, the emotions may be different for each individual, and the intensity of the emotions may also be different. In view of such a property, the word dictionary storage server 110 classifies each emotional class for emotional words such as happy, embarrassing, and depriving, and similarity, positive or negative, sensitivity to the classified emotional classes. Classify the intensity and store it as an emotional word dictionary. Here, the emotion class classifies the internal emotional states of human beings such as satisfaction, longing, and happiness, and according to an embodiment of the present invention, the emotion class may be classified into a total of 77 emotional classes, and among them, the emotional class to which the corresponding word belongs may be matched. Here, the number of emotion classes is only an example of a sortable emotion, but is not limited thereto. Similarity represents the similarity between the word and the item in the emotion class, and can be expressed as a numerical value within a certain range. Positive or negative degree is a degree indicating whether the attribute of the word is a positive emotion or a negative emotion, can be expressed as a negative or positive number within a certain range based on zero. Emotional strength indicates the strength of emotion among attributes of the word, and may be expressed as a numerical value within a certain range. 2 is a view showing an example of the emotional word dictionary according to an embodiment of the present invention, where the similarity is expressed as a numerical value in the range of 0 to 10, positive or negative is expressed as 0, 1 or -1, Intensity is expressed as a number from 0 to 10. However, such a numerical value is not limited to the illustrated range and various modifications are possible. For example, positive or negative may be expressed as a numerical value of 0.1 unit within the range of -1 to 1, similarity or emotional strength may also be expressed as a numerical value of 0.1 unit within the range of 0 to 1. Also, the emotional word storage unit 102 may classify a plurality of emotional classes for the same emotional word, such as tickle, warm, or lumpy, and in this case, similarity, At least one of positive or negative or emotional strength may be classified and stored as an emotional word dictionary. In addition, even if the same emotional words, emotional class, similarity, positive or negative, emotional intensity is different or gender, age for each user according to environmental information including at least one of input time, place, and weather of sentences logged for each user. Emotion class, similarity, positive or negative, and emotional strength may vary according to profile information including personality, personality, and occupation.Emotion class, similarity, positive or negative, emotional strength in consideration of user-specific environment information and profile information If at least one is inferred differently, an emotional word dictionary for each user may be set and stored based on the emotional log information for each user.

The directed information storage server 120 stores the emotional directed information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal. That is, the directing information storage server 120 may store the emotional directing information corresponding to a total of 77 emotional classes classified in the emotional word dictionary, and the emotional directing information is avatar, emoticon, flash image, video, image, sound. It can be represented as a signal. Here, the avatar refers to a visual image representing an alter ego in an online virtual society. An emoticon is a compound word of emotion and icon, which means emotion. It is also called 'Graphics' in Korean. In the virtual space online, a combination of letters, symbols, and numbers from computer keyboards Say an expression that represents your intention. Flash images refer to animations created using flash software. The video refers to an image photographed by a user to express his emotion. An image is a visual representation of a scene that appears on a screen by a photographic process, and refers to a single picture composed of a constant screen size and camera angle. The sound signal is a sound signal. The sound of animals such as dogs, cats, and lions, the sounds of objects such as ships, trains, cars, and musical instruments, or the sounds of nature, such as streams, waves, and water drops, and laughter, crying, and angry sounds. It includes the sound of a person such as the back. In this case, the direction information storage server 120 may store the emotion direction information registered by the user as well as the emotion direction information provided by the emotion directing service provider for each user. That is, the user may match and store the avatar, emoticon, flash image, video, image, and audio signal selected by the user with at least one of a corresponding emotional class, similarity, positive or negative, and emotional strength. In addition, even in the case of a video, it is possible to use the 3D image or the like to edit and store your own unique emotional presentation information. Accordingly, the emotion directing service providing system 100 may provide a unique emotion directing corresponding to each user according to a user's selection or editing. At this time, the direction information storage server 120 may be implemented to exchange, distribute or share the direction information stored for each user between users.

The emotion inference server 130 infers the emotion matched to the emotional word dictionary for at least one of the text and voice signals received from the user terminals 160 and 170 connected through the network 150. The emotional reasoning server 130 may be implemented as an emotional reasoning apparatus 300 as shown in FIG. 3.

3 is a diagram schematically showing the configuration of the emotional reasoning server 130 of FIG. Hereinafter, it will be described on the assumption that the emotional reasoning apparatus 300 is used as the emotional reasoning server 130 of the emotional expression service providing system 100. Referring to FIG. 3, the emotional reasoning server 300 may include a sentence converter 310, a matching checker 320, an emotional reasoner 330, an emotional log storage 340, and a log information searcher 350. It may include.

The sentence converting unit 310 detects words and phrases with respect to texts received from the user terminals 160 and 170 connected through the network 150 and converts them into basic forms. That is, the sentence converting unit 310 may first divide the text received from the user terminals 160 and 170 into a plurality of words, and then convert the text into a basic form. The combination of words allows you to identify phrases and convert them to basic forms.

The matching confirming unit 320 compares each word and phrase converted by the sentence converting unit 310 with the emotional word dictionary stored in the word dictionary storage server 110 to identify a matching word or phrase.

The emotion inference unit 330 applies a probabilistic model based on the co-occurrence of the converted words and phrases, and infers emotion according to the applied probability model. For example, assuming that the word 'bucked up' among the words converted into the basic form by the sentence converter 310 matches the emotional class of 'emotion' of the emotional word dictionary, the emotional reasoning unit 330 is 'bucked up'. A probability model may be applied based on a combination of the word and other words or phrases converted into a basic form, and the emotion may be inferred according to the applied probability model. Here, the probability model is an algorithm for calculating the probability of belonging to a specific emotion by using the frequency of a specific word or phrase in the entire corpus, and based on this, the probability of a new word belonging to a specific emotion may be calculated. For example, as shown in Equation 1, inferring the similarity of emotion for a new word by calculating the frequency of the combination of the new word W and a specific emotional C in a sentence in the corpus for the total frequency of the new word W in the corpus. can do.

In addition, PMI (Pointwise Mutual Information) may be used to obtain Co-occurrence similarity. In this case, PMI may be calculated as in Equation 2.

Dice coefficients Dice (W, C) may be used in a manner similar to PMI.

The formula for inferring emotional similarity is not limited to the presented equation, and various modifications are possible.

The emotion reasoning unit 330 infers the emotions for <word + word>, <word + phrase>, and <phrase + phrase> in this manner, and then combines each of the inferred emotions to provide emotion for the whole sentence. It can be inferred.

The emotional log storage unit 340 may store an emotional log including a word and a word, a word and a phrase, a phrase, and a phrase based on the word or phrase checked by the matching verifier 320. That is, the emotion log storage unit 340 may store sentences logged by the user as a combination of meaningful words and words, words and phrases, phrases, and phrases to infer emotions for new words. For example, the emotion log storage unit 110 uses the basic type word "love" that is confirmed as having a feeling of "love" by the matching confirmation unit 320, and "completion" which is a word without emotion in the sentence. And "friends", respectively, and two emotional logs of <slow-love> and <friend-love> can be generated and stored. In this case, as shown in FIG. 4, the emotional log may store time information together. In this case, the information stored together with the emotion log information is not limited to time information, and weather information, location information of the user, and the like may be stored together.

The log information search unit 350 may search whether there is log information that is greater than or equal to a value set for the log information stored in the emotional log storage unit 340. That is, the log information search unit 350 searches whether the number of log information stored in the emotional log storage unit 340 is greater than or equal to a set value. In this case, the emotional reasoning unit 330 may be associated with the phrase or sentence combined with the word or phrase only when a specific word or phrase is stored more than the number of times set in the emotional log storage unit 340 through the log information search unit 350. It can be implemented to infer emotion.

5 is a diagram illustrating another configuration example of the emotional reasoning server 130 of FIG. 1. The emotional reasoning server 130 may be implemented as an emotional reasoning apparatus 300 as shown in FIG. 5. Hereinafter, it will be described on the assumption that the emotional reasoning apparatus 300 is used as the emotional reasoning server 130 of the instant messenger service providing system 100.

The emotion reasoning server 500 may include an emotion storage unit 510, a sound source detector 520, and an emotion reasoning unit 530.

The emotion storage unit 510 stores emotions represented by at least one of color, image, font, emoticon, avatar, video, voice, tone, sound quality, and sound wave corresponding to each emotion class classified in the emotion word dictionary. That is, the emotion storage unit 510 may express colors, images, fonts, emoticons, avatars, videos, voices, tones, sound quality, or sound waves corresponding to each of the 77 emotional classes in the whole classified in the emotional word dictionary. It may store the emotions corresponding to the tone, sound quality or sound waves matched with each emotion class.

Here, the tone refers to the sensory characteristic that appears in the difference of the components of the sound. Even if the sound of the same height is raised to the same size, there is a difference in the sensory properties of the sound depending on the difference in the pronunciation and the vibration method, which is determined by what intensity of the partial sound is generated by the vibration. For example, it is for this reason that the sounds of the piano and violin are completely different even though the pitch is the same.

In addition, the sound quality means the degree of sharpness or softness of the sound, also called tone. In addition, sound waves vibrate the eardrum in the form of a longitudinal wave by partially changing the pressure in the medium (air) where the vibration of the object was uniform, and when the vibration of the string or the object moves in an arbitrary direction, it partially The high and low pressures appear at that time, and the pressure difference causes the air particles to move from the high pressure to the low pressure, so that the sound waves spread out, and when the air pressure changes reach people's ears, the eardrum Will vibrate.

The sound source detector 520 detects a sound source including at least one of a tone, sound quality, and sound wave with respect to a voice signal received from the user terminals 160 and 170 connected through the network 150. In this case, the sound source detector 520 preferably detects the same sound source as the sound source stored in the emotion storage unit 510. That is, when the tone is stored in the emotion storage unit 510, the sound source detector 520 may detect the tone in response to the voice signal received from the user terminal 160.

The emotion inference unit 530 infers a matched emotion from the emotion storage unit 510 based on the sound source detected by the sound source detector 520. That is, the emotion inference unit 530 infers the corresponding emotion by comparing the tone, sound quality or sound wave detected by the sound source detection unit 520 with the tone, sound source or sound wave stored in the emotion storage unit 510.

The directed information extraction server 140 of FIG. 1 extracts the emotional directed information corresponding to the emotion inferred by the emotional inference server 130 from the directed information storage server 120.

The emotion directing server 145 receives information on the destination terminal 170 from the user terminal 160, and transmits the text and / or voice signal received from the user terminal 160 and the extracted emotional direction information to the destination terminal 170. To send). The information on the destination terminal 170 may include a phone number of the destination terminal, a name of the destination terminal user, and the like. Here, it is assumed that the wired terminal 160 is a user terminal and the wireless terminal 170 is a destination terminal. However, the present invention is not limited thereto and vice versa or may be implemented in the same type of terminal. The same applies to the following.

In this case, the emotion directing server 145 may forcibly transmit the emotion directing information to the destination terminal 170 in a push up manner. In addition, if the emotional directing server 145 wants to provide emotional directing after a certain time after the text or voice signal is transmitted, or other additional information is to be transmitted together, the transmission condition information for the user from the user Receiving, and accordingly sent emotion directing information to the destination terminal 170. As a result, the destination terminal 170 may receive the text and / or the voice signal from the user terminal 160 and check the emotional direction information on the text and / or the voice signal on the screen.

6 is a view showing an example of the configuration of a device capable of intelligent emotional expression according to an embodiment of the present invention.

Referring to FIG. 6, the apparatus 600 capable of emotional expression includes a word dictionary storage unit 610, a presentation information storage unit 620, an emotional reasoning unit 630, a sentence converter 640, and a matching confirmation unit 650. ), An emotion storage unit 660, a sound source detector 670, a direction information extractor 680, and an emotion director 690.

Here, the device 600 capable of emotional expression refers to various devices capable of transmitting and receiving various data via a communication network () according to a user's key manipulation. Preferably, the device 600 may be a smart phone. It doesn't happen. That is, the device 600 capable of emotional expression includes a tablet PC, a laptop, a personal computer, a personal digital assistant, and a mobile communication terminal. Terminal) and the like, as well as may be implemented as a home appliance, such as a smart TV. On the other hand, the device 600 capable of emotional expression can be linked to a cloud computing device that supports cloud computing (Cloud Computing) that can use services such as data reading and writing and storage, network, content use through the network 150 Can be. That is, the apparatus 600 capable of emotional expression is a device for performing voice or data communication using the network 150, a memory for storing a program or protocol for communicating with an external device via the network 150, Means a terminal equipped with a microprocessor for executing and controlling the program. That is, the device 600 capable of emotional expression can be any device as long as it can communicate with an external device and can include any communication computing device such as a notebook computer, a mobile communication terminal, a PDA, and a smart TV. Concept.

The word dictionary storage 610 classifies the emotional word into an item including at least one of an emotion class, similarity, positive or negative, and emotional strength, and stores the emotional word as an emotional word dictionary.

The direction information storage unit 620 stores emotion direction information corresponding to each emotion class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal. That is, the directing information storage unit 620 may store emotional directing information corresponding to a total of 77 emotional classes classified in the emotional word dictionary, and the emotional directing information is avatar, emoticon, flash image, video, image, sound. It can be represented as a signal. Here, the direction information storage unit 620 may store the emotional direction information generated or edited by the user as well as the emotional direction information previously stored in the apparatus 600 capable of emotional expression. That is, the user matches the avatar, emoticon, flash image, video, image, and audio signal selected by the user with at least one of a corresponding emotional class, similarity, positive or negative, emotional intensity, and outputting information storage unit 620. Can be stored in In addition, even in the case of a video, it is possible to use the 3D image or the like to edit and store your own unique emotional presentation information. Accordingly, the apparatus 600 capable of emotional expression can provide a unique emotional presentation corresponding to each user according to a user's selection or editing.

The emotion inference unit 630 infers the emotion matched in the emotional word dictionary on at least one of the text and voice signals received from the user terminal 160 connected through the network 150.

In this case, the sentence converter 640 may identify the word and the phrase with respect to the text received from the user terminal 160 connected through the network 150 and convert the word and phrase into a basic form. In addition, the matching checker 650 may check a word or phrase that matches the emotional word dictionary among the words and phrases converted by the sentence converter 640. In this case, the emotional reasoning unit 630 applies a probability model based on the co-occurrence of the word or phrase converted by the sentence converter 640, and applies the emotional model according to the result value of the applied probability model. Can deduce.

Meanwhile, the emotion storage unit 660 may store emotions represented by at least one of color, image, font, emoticon, avatar, video, voice, tone, sound quality, and sound wave corresponding to each emotion class classified in the emotion word dictionary. have. That is, the emotion storage unit 510 may express colors, images, fonts, emoticons, avatars, videos, voices, tones, sound quality, or sound waves corresponding to each of the 77 emotional classes in the whole classified in the emotional word dictionary. The emotion corresponding to the tone, sound quality or sound wave matched with each emotion class may be stored together. In addition, the sound source detector 670 detects a sound source including at least one of a tone, sound quality, and sound wave with respect to a voice signal received from the user terminals 160 and 170 connected through the network 150. In this case, the sound source detector 670 may detect the same sound source as the sound source stored in the emotion storage unit 660. That is, when a tone is stored in the emotion storage unit 510, the sound source detection unit 670 may receive a voice signal received from the user terminal 100 (in this case, the device 600 capable of emotion generation serves as a user terminal). It is preferable to detect the tone with respect to. In this case, the emotion inference unit 630 may infer a matched emotion from the emotion storage unit 660 based on the sound source detected by the sound source detector 670. That is, the emotion inference unit 630 may infer a corresponding emotion by comparing the tone, sound quality, or sound wave detected by the sound source detection unit 670 with the tone, sound source, or sound wave stored in the emotion storage unit 660.

The director information extracting unit 680 extracts emotion director information corresponding to the emotion inferred by the emotion reasoning unit 630 from the director information storage unit 620.

The emotion directing unit 690 receives information on the destination terminal 170 from the user terminal 160, and transmits the text and / or voice signal received from the user terminal 160 and the extracted emotion directing information to the destination terminal 170. To send. The information on the destination terminal 170 may include a phone number of the destination terminal, a name of the destination terminal user, and the like. In this case, the emotion directing unit 690 may forcibly transmit the emotion directing information to the destination terminal 170 in a push up manner. In addition, the emotion directing unit 690 receives the transmission condition information from the user when the user wants to provide the emotional direction after a certain time after the text or voice signal is transmitted, or other additional information is to be transmitted together. In addition, the emotion directing information may be transmitted to the destination terminal 170 accordingly. As a result, the destination terminal 170 may receive the text and / or the voice signal from the user terminal 160 and check the emotional direction information on the text and / or the voice signal on the screen.

FIG. 7 is a flowchart illustrating a method for providing emotional service by the emotional service providing system of FIG. 1.

1 and 7, the word dictionary storage server 110 stores the emotional word as an emotional word dictionary by classifying it into an item including at least one of an emotional class, similarity, positive or negative, and emotional strength ( S701).

In addition, the direction information storage server 120 stores emotion direction information corresponding to each emotion class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal (S703). ). That is, the directing information storage server 120 may store the emotional directing information corresponding to a total of 77 emotional classes classified in the emotional word dictionary, and the emotional directing information is avatar, emoticon, flash image, video, image, sound. It can be represented as a signal. In the drawing, although the emotional word dictionary storing step and the emotional directing information storing step are configured as separate steps, the emotional word dictionary storing step and the emotional directing information storing step may be implemented as one step.

The emotional storage unit 510 of the emotional reasoning server 130 corresponds to at least one of color, image, font, emoticon, avatar, video, voice, tone, sound quality, and sound wave corresponding to each emotional class classified in the emotional word dictionary. The corresponding emotion may be stored (S705). That is, the emotion storage unit 510 may match a tone, sound quality, or sound wave corresponding to all 77 emotion classes classified in the emotional word dictionary, and correspond to the tone, sound quality, or sound wave matched with each emotion class. You can save your emotions together.

When the signal received from the user terminal 160 connected through the network 150 is a text signal (S707), the sentence converter 310 of the emotional reasoning server 130 grasps words and phrases with respect to the text, and then converts the text into a basic form. Can be converted (S709). That is, the sentence converting unit 310 may first divide the text received from the user terminals 160 into a plurality of words and then convert the text into a basic form, and words and words used conventionally among the segmented words. You can use the combination of to figure out the syntax and convert it to the basic form.

In addition, the matching checker 320 may check the words or phrases matched by comparing each word and phrase converted by the sentence converter 310 with the emotional word dictionary stored in the word dictionary storage server 110 ( S711).

The emotion inference unit 330 applies a probabilistic model based on the co-occurrence of the converted words and phrases, and infers emotion according to the applied probability model (S713). For example, assuming that the word 'bucked up' among the words converted into the basic form by the sentence converter 310 matches the emotional class of 'emotion' of the emotional word dictionary, the emotional reasoning unit 330 is 'bucked up'. A probability model may be applied based on a combination of the word and other words or phrases converted into a basic form, and the emotion may be inferred according to the applied probability model.

When the signal received from the user terminal 160 connected through the network 150 is a voice signal (S715), the sound source detector 520 may generate a tone from the voice signal received from the user terminal 160 connected through the network 150. The sound source including at least one of sound quality and sound waves may be detected (S717). In this case, the sound source detector 520 preferably detects the same sound source as the sound source stored in the emotion storage unit 510. That is, when the tone is stored in the emotion storage unit 510, the sound source detector 520 may detect the tone from the voice signals received from the user terminals 160.

The emotion inference unit 530 may infer a matched emotion from the emotion storage unit 510 based on the sound source detected by the sound source detection unit 520 (S719). That is, the emotion inference unit 530 may infer a corresponding emotion by comparing the tone, sound quality, or sound wave detected by the sound source detection unit 520 with the tone, sound source, or sound wave stored in the emotion storage unit 510.

The directed information extraction server 140 extracts emotional directed information corresponding to the emotion inferred by the emotional inference server 130 from the directed information storage server 120 (S721).

The emotion directing server 145 receives the information on the destination terminal 170 from the user terminal 160 (S723), and receives the text and / or voice signal and the extracted emotion directing information received from the user terminal 160 as a destination. The terminal 170 transmits the data to the terminal 170 (S725).

In this case, the emotion directing server 145 may forcibly transmit the emotion directing information to the destination terminal 170 in a push up manner. As a result, the destination terminal 170 may receive the text and / or the voice signal from the user terminal 160 and check the emotional direction information on the text and / or the voice signal on the screen.

8 is a flowchart illustrating an intelligent emotional expression method by the apparatus capable of emotional expression of FIG. 6. Referring to FIGS. 6 and 8, the word dictionary storage unit 610 classifies the emotional word into an item including at least one of an emotional class, similarity, positive or negative, and emotional strength, and stores the emotional word as an emotional word dictionary. S801).

In addition, the direction information storage unit 620 stores emotion direction information corresponding to each emotion class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a video, an image, and an audio signal (S803). ). That is, the directing information storage unit 620 may store emotional directing information corresponding to a total of 77 emotional classes classified in the emotional word dictionary, and the emotional directing information is avatar, emoticon, flash image, video, image, sound. It can be represented as a signal. In the drawing, although the emotional word dictionary storing step and the emotional directing information storing step are configured as separate steps, the emotional word dictionary storing step and the emotional directing information storing step may be implemented as one step.

The emotion storage unit 660 may store emotions corresponding to at least one of color, image, font, emoticon, avatar, video, voice, tone, sound quality, and sound wave corresponding to each emotion class classified in the emotion word dictionary. (S805). That is, the emotion storage unit 660 may match the tone, sound quality, or sound waves corresponding to the total 77 emotion classes classified in the emotional word dictionary, and correspond to the tone, sound quality, or sound waves matched with each emotion class. You can save your emotions together.

When the signal received from the apparatus 600 capable of emotional expression is a text signal (S807), the sentence converter 640 may identify a word and a phrase with respect to the text and convert the text into a basic type (S809). That is, the sentence converting unit 640 may first divide the text received from the apparatus 600 capable of emotional expression into a plurality of words, and then convert the text into a basic form, and words conventionally used among the segmented words The combination of and words can be used to identify a phrase and convert it to a basic form. The matching checker 650 may check the matched word or phrase by comparing each converted word and phrase with the emotional word dictionary stored in the word dictionary storage 610 (S811). In this case, the emotion inference unit 630 applies a probabilistic model based on the co-occurrence of the converted words and phrases, and infers emotion according to the applied probability model (S813). For example, assuming that the word 'bucked up' among the words converted into the basic form by the sentence converter 310 matches the emotional class of 'emotion' of the emotional word dictionary, the emotional reasoning unit 330 is 'bucked up'. A probability model may be applied based on a combination of the word and other words or phrases converted into a basic form, and the emotion may be inferred according to the applied probability model.

When the signal received from the user of the apparatus 600 capable of emotional expression is a voice signal (S815), the sound source detector 670 may generate a sound source including at least one of a tone, sound quality, and sound wave with respect to the voice signal received from the user. It can be detected (S817). In this case, the sound source detector 670 may detect the same sound source as the stored sound source. That is, when the tone is stored, it is preferable that the sound source detector 670 detects the tone in response to the voice signal received from the apparatus 600 capable of emotional expression. In this case, the emotion reasoning unit 630 may infer a matched emotion based on the detected sound source (S819). That is, the direction information extractor 630 may infer a corresponding emotion by comparing the detected tone, sound quality or sound wave with the stored tone, sound source or sound wave.

The director information extracting unit 680 extracts emotion director information corresponding to the emotion inferred by the emotion reasoning unit 630 from the director information storage unit 620 (S821).

The emotion directing unit 690 receives information on the destination terminal 170 from the user of the apparatus 600 capable of emotional presentation (S823), and receives the text and / or voice signal received from the user and the extracted emotional direction information. It transmits to the terminal 170 (S825). The information on the destination terminal 170 may include a phone number of the destination terminal, a name of the destination terminal user, and the like. In this case, the emotion directing unit 690 may forcibly transmit the emotion directing information to the destination terminal 170 in a push up manner. As a result, the destination terminal 170 may receive the text and / or the voice signal from the user terminal 160 and check the emotional direction information on the text and / or the voice signal on the screen.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. In other words, within the scope of the present invention, all of the components may be selectively operated in combination with one or more. In addition, although all of the components may be implemented in one independent hardware, each or all of the components may be selectively combined to perform some or all functions combined in one or a plurality of hardware. It may be implemented as a computer program having a. Codes and code segments constituting the computer program may be easily inferred by those skilled in the art. Such a computer program may be stored in a computer readable storage medium and read and executed by a computer, thereby implementing embodiments of the present invention. The storage medium of the computer program may include a magnetic recording medium, an optical recording medium, a carrier wave medium, and the like.

In addition, the terms "comprise", "comprise" or "having" described above mean that the corresponding component may be inherent unless specifically stated otherwise, and thus excludes other components. It should be construed that it may further include other components instead. All terms, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. Terms used generally, such as terms defined in a dictionary, should be interpreted to coincide with the contextual meaning of the related art, and shall not be interpreted in an ideal or excessively formal sense unless explicitly defined in the present invention.

The above description is merely illustrative of the technical idea of the present invention, and those skilled in the art to which the present invention pertains may make various modifications and changes without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention but to describe the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas falling within the scope of the same shall be construed as falling within the scope of the present invention.

100: emotional directing service providing system
110: word dictionary storage server 120: directing information storage server
130: emotional reasoning server 140: directed information extraction server
150: emotional production server

Claims (12)

A word dictionary storage server for classifying an emotional word into an item including at least one of an emotional class, similarity, positive or negative, and emotional strength, and storing the emotional word as an emotional word dictionary;
A presentation information storage server for storing emotional presentation information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a moving picture, an image, and an audio signal;
An emotional reasoning server for inferring an emotion matching the emotional word dictionary with respect to at least one of a text and an audio signal received from a user terminal connected through a network;
A production information extraction server for extracting emotion production information corresponding to the emotion from the production information storage server; And
Emotion directing server for receiving information on the destination terminal from the user terminal, and transmits at least one of the text and voice signal received from the user terminal and the extracted emotional direction information to the destination terminal.
Emotional expression service providing system comprising a.
The method of claim 1, wherein the emotional reasoning server,
A sentence converter for grasping words and phrases from the text and converting them into basic forms;
A matching checker for checking a word or phrase that matches the emotional word dictionary among the converted words and phrases; And
Emotion reasoning unit applying a probabilistic model based on the co-occurrence of the converted word and phrase, and inferring emotion according to the applied probability model
Emotional expression service providing system comprising a.
The method of claim 1, wherein the emotional reasoning server,
An emotion storage unit for storing emotions represented by at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave corresponding to each emotion class classified in the emotion word dictionary;
A sound source detector for detecting a sound source including at least one of a tone, sound quality, and sound wave from the voice signal; And
Emotion inference unit that infers a matched emotion from the emotion storage unit based on the detected sound source
Emotional expression service providing system comprising a.
A word dictionary storage unit classifying an emotional word into an item including at least one of an emotion class, similarity, positive or negative, and emotional strength, and storing the emotional word as an emotional word dictionary;
A production information storage unit for storing emotional presentation information corresponding to each emotional class classified in the emotional word dictionary using at least one of an avatar, an emoticon, a flash image, a moving picture, an image, and an audio signal;
An emotional inference unit for inferring an emotion matching the emotional word dictionary from at least one of a received text and an audio signal;
Directing information extraction unit for extracting the emotional directing information corresponding to the emotion from the directing information storage unit; And
Emotion directing unit for receiving information on the destination terminal, and transmits at least one of the received text and voice signal and the extracted emotion directing information to the destination terminal
Apparatus capable of intelligent emotional expression comprising a.
The method of claim 4, wherein
A sentence converter for grasping words and phrases from the text and converting them into basic forms; And
Matching confirmation unit for confirming a word or phrase matching the emotional word dictionary of the converted words and phrases
More,
The emotional reasoning unit applies a probabilistic model based on the co-occurrence of the converted words and phrases, and induces emotional emotion according to the applied probability model. Device.
The method of claim 4, wherein
An emotion storage unit for storing emotions represented by at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave corresponding to each emotion class classified in the emotion word dictionary; And
A sound source detection unit for detecting a sound source including at least one of a tone, sound quality, sound waves from the voice signal
More,
And the emotional inference unit infers the emotions matched from the emotional storage unit based on the detected sound source.
The emotional word is classified into an item including at least one of emotional class, similarity, positive or negative, and emotional strength, and stored as an emotional word dictionary, and at least one of an avatar, an emoticon, a flash image, a video, an image, an audio signal Storing emotion directing information corresponding to each emotion class classified into the emotion word dictionary using the emotion word dictionary;
Inferring an emotion matching the emotional word dictionary with respect to at least one of a text and an audio signal received from a user terminal connected through a network;
Extracting emotion directing information corresponding to the emotion from the stored emotion directing information;
Receiving information on a destination terminal from the user terminal; And
Transmitting at least one of the text and voice signals received from the user terminal and the extracted emotion directing information to the destination terminal;
Emotional expression service providing method comprising a.
The method of claim 7, wherein the emotional reasoning step,
Identifying a word and a phrase with respect to the text and converting the same into a basic form;
Identifying a word or phrase that matches the emotional word dictionary among the converted words and phrases; And
Applying a probabilistic model based on the co-occurrence of the converted words and phrases
Including;
Emotional expression service providing method comprising inferring the emotion according to the applied probability model.
The method of claim 7, wherein the emotional reasoning step,
Storing emotions represented by at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave, corresponding to each emotion class classified in the emotion word dictionary; And
Detecting a sound source including at least one of a tone, sound quality, and sound wave from the voice signal
Including;
Based on the detected sound source inferred emotion matching service from the emotional storage unit.
The emotional word is classified into an item including at least one of emotional class, similarity, positive or negative, and emotional strength, and stored as an emotional word dictionary, and at least one of an avatar, an emoticon, a flash image, a video, an image, an audio signal Storing emotion rendering information corresponding to each emotion class classified by the emotion word dictionary using the emotion word dictionary;
Inferring an emotion matching the emotional word dictionary from at least one of the received text and voice signal;
Extracting emotion directing information corresponding to the emotion from the stored emotion directing information;
Receiving information about a destination terminal; And
Transmitting at least one of the text and voice signal and the extracted emotion directing information to the destination terminal;
Intelligent emotional expression method of the apparatus capable of emotional expression comprising a.
The method of claim 10,
Identifying a word and a phrase with respect to the text and converting the same into a basic form;
Identifying a word or phrase that matches the emotional word dictionary among the converted words and phrases; And
Applying a probabilistic model based on the co-occurrence of the converted words and phrases
More,
The emotional reasoning step is a method for intelligent emotional expression of the emotion-expressing device, characterized in that inferring emotion according to the applied probability model.
The method of claim 10,
Storing emotions represented by at least one of a color, an image, a font, an emoticon, an avatar, a video, a voice, a tone, a sound quality, and a sound wave, corresponding to each emotion class classified in the emotion word dictionary; And
Detecting a sound source including at least one of a tone, sound quality, and sound wave from the voice signal
More,
The emotional reasoning step is an intelligent emotional expression method of the emotion-expressing device, characterized in that it infers a matched emotion from the stored emotion based on the detected sound source.

Images