WO2020228208A1

WO2020228208A1 - User smart device and emoticon processing method therefor

Info

Publication number: WO2020228208A1
Application number: PCT/CN2019/106033
Authority: WO
Inventors: 易欣; 周凡贻; 尚国强
Original assignee: 深圳传音控股股份有限公司
Priority date: 2019-05-13
Filing date: 2019-09-16
Publication date: 2020-11-19
Also published as: CN110222210A

Abstract

A user smart device and an emoticon processing method therefor. The method comprises: a user smart device obtains input information of a user; performs, according to the input information, emotion recognition on the user; obtains corresponding key information according to the emotion recognition result; and according to the key information, generates a corresponding emoticon. Emotion recognition can be performed according to the input information of the user, and emotion pictures, texts or languages, etc. can be generated into corresponding emoticons, which achieves the quick input effect of multiple input channels, performs emotion recognition on long input information, meets individual needs, and achieves the artificial intelligence to a certain extent.

Description

User intelligent equipment and its emotional icon processing method

This patent application requires the priority of the Chinese patent application number 201910395252.1 filed on May 13, 2019, the applicant is Shenzhen Transsion Holdings Co., Ltd., and the invention title is "User's Smart Device and Its Emotional Symbol Processing Method". This application The full text of is incorporated into this application by reference.

Technical field

This application relates to the field of information processing technology, and in particular to an emotion icon processing method, and a user's smart device applying the emotion icon processing method.

Background technique

With the rapid development of personal mobile terminal technology, personal mobile communication functions are becoming more and more perfect. Correspondingly, the emoticons (pictures or combinations of various strings) that need to be used in communication are gradually becoming individual化.

For example, when a user enters a message in instant messaging software or social platforms, if he wants to insert an emoticon, he will select an emoticon that he wants to insert from the existing emoticons, or enter an emoticon that corresponds to the preset emoticon Text, and then select one of the emoji recommendation results. Among them, "existing emoticons" generally come from presets in application software or downloaded from the Internet, and "emoticons" generally only refer to emoji and static or dynamic graphics. Of course, users can also take pictures or videos and add text to generate static or dynamic emoticons.

However, the user can only use the designed emoji, and can only obtain the corresponding emoji recommendation results by manually clicking to select or text input, but the messages input by the user are various, not only text but also voice, pictures, and videos Etc. Therefore, this solution cannot meet other information input scenarios. Moreover, sometimes the input text can only be compared with the name of the existing emoji to obtain the emoji recommendation results related to the input text. When the input text has emotional meaning but is not completely equal to the name of the existing emoji, Unable to get emoji recommendation results. For example, if you enter "I am so happy", you cannot get the emoji recommendation related to "happy", you must enter "happy" to get it, which does not conform to the user's real habits. Furthermore, the emojis that have been designed are not personalized enough to meet the individual needs of current users.

Similarly, users generate static or dynamic emoticons by shooting images or videos, and also need to add text and other editing to indicate their emotional state. Although it meets individual needs, it is not smart and portable, which increases the user's burden, and the user experience is poor.

In view of the various deficiencies of the prior art, the inventor of the present application has conducted in-depth research and proposed a user smart device and a method for processing emotional icons.

Summary of the invention

technical problem

The solution to the problem

Technical solutions

The purpose of this application is to provide a user's smart device and its emotional icon processing method, which can recognize emotions based on user input information, generate corresponding emotional icons from emotional pictures, text, or language, and realize multiple input channels The fast input effect can also realize the emotion recognition of long input information, which meets the individual needs, and to a certain extent realizes artificial intelligence.

In order to solve the above technical problem, the present application provides an emotion icon processing method. As one of the implementation manners, the emotion icon processing method includes the steps:

The user's smart device obtains the user's input information;

Perform emotion recognition on the user according to the input information;

Obtain corresponding key information according to the result of emotion recognition;

The corresponding emotion icon is generated according to the key information.

As one of the implementation manners, the step of acquiring the user's input information by the user smart device specifically includes:

The user's smart device captures the user's body language content through the camera device;

Preprocessing the body language content as the input information.

As one of the implementation manners, the step of preprocessing the body language content as the input information specifically includes:

Perform action recognition on the body language content;

Obtain the target part whose motion amplitude change value is greater than the preset threshold according to the result of motion recognition;

Use the movement of the target part as the input information.

As one of the implementation manners, the body language content is eyes, mouth, hands, head, torso, feet, or any combination of two or more actions.

The user's smart device obtains the user's voice content through the microphone;

Preprocessing the language content as the input information.

As one of the implementation manners, the step of performing emotion recognition on the user according to the input information specifically includes:

Emotion recognition is performed on the data of the voice content by using voice emotion analysis technology.

The user’s smart device obtains the user’s text through the text message input box;

The text is preprocessed as the input information.

Using text-based sentiment analysis technology to perform sentiment recognition on the text data.

As one of the implementation manners, after the step of generating and obtaining the corresponding emotion icon according to the key information, the method further includes:

Prompt the user to preview the emotion icon;

Determine whether the user's confirmation operation is obtained;

If the user's confirmation operation is obtained, the emotion icon is sent and/or stored.

As one of the implementation manners, the step of generating and obtaining corresponding emotion icons according to the key information specifically includes:

Obtaining a corresponding emotion expression vector according to the key information;

The corresponding emotion icon is generated according to the emotion expression vector.

As one of the embodiments, the emotion expression carrier is an emotion picture, an emotion text, an emotion voice, or any combination of the three.

As one of the implementation methods:

When the input information is body language content, the key information is emotional action information composed of eyes, mouth, hands, head, torso, feet, or any combination of two or more;

When the input information is language content, the key information is emotional voice information composed of sentences, tone, words, scenes, or any combination of two or more;

When the input information is text, the key information is emotional text information composed of sentences, words, scenes, context, or any combination of two or more;

Wherein, the emotional icon is composed of the eyes, mouth, hands, head, torso, feet, or any two or more combinations of emotional action information, sentence, tone, word, scene or any two Or the emotional speech information formed by the above combination, and/or the emotional text information formed by the sentence, word, scene, context or any combination of two or more.

As one of the implementation manners, the emotion icon is an expression static image, an expression video, and an expression dynamic image including pictures, text, and/or voice for communication.

As one of the implementation manners, when the emotion icon is sent to the receiver, it is also used to generate an interactive effect according to the on-demand action of the receiver.

In order to solve the above technical problems, this application also provides a user smart device. As one of the implementation manners, the user smart device includes a processor, and the processor is configured to execute a computer program to implement the emotion icon as described above. Approach.

The beneficial effects of the invention

Beneficial effect

In the user smart device and its emotional icon processing method provided in the present application, the user smart device obtains user input information, performs emotion recognition on the user according to the input information, and obtains corresponding key information according to the result of the emotion recognition, and according to the key The information is generated to obtain the corresponding emotion icon. This application can perform emotion recognition based on user input information, generate corresponding emotion icons from emotion pictures, text or language, etc., realize the quick input effect of multiple input channels, and can also realize emotion recognition for longer input information, satisfying In order to meet individual needs, artificial intelligence has been realized to a certain extent.

Brief description of the drawings

Description of the drawings

FIG. 1 is a schematic flowchart of an embodiment of the emotion icon processing method of this application.

Fig. 2 is a schematic diagram of modules of an implementation manner of a user smart device of this application.

Invention embodiment

Embodiments of the invention

In order to further illustrate the technical means and effects adopted by the application to achieve the intended application purpose, the application will be described in detail below with reference to the drawings and preferred embodiments.

Through the description of the specific implementation manners, it is possible to gain a deeper and specific understanding of the technical means and effects adopted by this application to achieve the intended purpose. However, the attached drawings are only for reference and explanation purposes, and are not used to describe the application. limit.

Please refer to Figure 1. Figure 1 is a schematic flow chart of an embodiment of the emotional icon processing method of this application. The emotional icon processing method of this embodiment can be applied to user smart devices such as mobile phones, laptops, tablets, or wearable devices. on.

It should be noted that, as shown in FIG. 1, the emotion icon processing method described in this embodiment may include but is not limited to the following steps.

Step S101, the user's smart device obtains the user's input information.

Step S102: Perform emotion recognition on the user according to the input information.

Step S103: Obtain corresponding key information according to the result of emotion recognition.

Step S104: Generate and obtain corresponding emotion icons according to the key information.

In this embodiment, the step of generating and obtaining a corresponding emotion icon based on the key information specifically includes: obtaining a corresponding emotion expression vector according to the key information; generating a corresponding emotion image according to the emotion expression vector symbol.

It should be noted first that the emotion expression carrier in this embodiment can be an emotion picture, an emotion text, an emotion voice or any combination of the three.

In this embodiment, when the input information is body language content, the key information is emotional action information composed of eyes, mouth, hands, head, torso, feet, or any combination of two or more; When the input information is language content, the key information is emotional speech information composed of sentences, mood, words, scenes, or any combination of two or more; when the input information is text, the key information The information is emotional text information composed of sentences, words, scenes, contexts, or any combination of two or more. Wherein, in one embodiment, the emotional icon is composed of the eyes, mouth, hands, head, torso, feet, or any combination of two or more of emotional action information, sentences, tone, and words , Scenes, or any two or more combinations of emotional speech information, and/or, sentences, words, scenes, context, or any two or more combinations of emotional text information.

It should be noted that the emotion icon in this embodiment is an expression static image, an expression video, and an expression dynamic image including pictures, text, and/or voice for communication.

In addition, when the emotion icon described in this embodiment is sent to the recipient, it is also used to generate an interaction or stress response effect according to the on-demand action of the recipient.

Among them, the stress response effect described in this embodiment refers to, for example, after the emotional icon is sent to the recipient, it is displayed as "angry", and when the recipient orders the emotional icon from different positions, it can produce different results. For example, if it is a double-click to apologize, it can display a stress response of "forgive you"; or if it is detected that the receiver does not make any on-demand actions, it can display "I am very angry", "I am very The mood of “very angry” deepens and strengthens the effect.

It should be noted that the step of acquiring the user’s input information by the user’s smart device in S101 of this embodiment may specifically include: the user’s smart device photographing the user’s body language content through a camera; and preprocessing the body language content to As the input information.

For example, the user's smart device can capture the user's body language content through multiple cameras such as front and rear cameras, and it can also perform continuous shooting and three-dimensional shooting for comparison and screening.

Further, the step of preprocessing the body language content as the input information in this embodiment may specifically include: performing action recognition on the body language content; obtaining the change in the motion amplitude according to the result of the action recognition A target part with a value greater than a preset threshold; taking the action of the target part as the input information.

For example, if the user wants to express surprise, the largest change in the amplitude of the action is usually the eyes or mouth. For example, if the eyes are opened very wide, or the mouth is opened very wide, there may be no obvious changes in the face. Therefore, the most effective The expression of emotions will be the movements of the eyes or mouth.

Of course, considering that the user can express different emotions through a variety of actions, the body language content in this embodiment is eyes, mouth, hands, head, torso, feet, or any combination of two or more actions.

This implementation manner performs emotion recognition on the user based on the input information, which may specifically include: firstly, modeling the joint points of the human body. The human body is regarded as a rigid system with internal connections, which includes bones and joint points. The relative movement with the joint points constitutes the change of the human body’s posture, that is, the usual description action. Among the many joints of the human body, according to the degree of emotional impact, the fingers and toes are ignored, and the human spine is abstracted into the neck, chest and Three joints of the abdomen, sum up a human body model, in which the upper body includes the head, neck, chest, abdomen, two upper arms and two forearms, and the lower body includes two thighs and two lower legs; for a variety of emotional states selected , The human body is selected to express each emotional state under normal conditions, and the body response is analyzed in detail; since the human body is abstracted into a rigid body model, the first is the movement of the human center of gravity, which is divided into forward, backward and natural states ; In addition to the movement of the center of gravity, followed by the rotation of the joint points, the human body changes in motion, and the joint points related to emotions include the head, chest, shoulders and elbows. The corresponding actions are head bending, chest rotation, The swing and extension direction of the upper arm, and the bending of the elbow, these parameters combined with the movement of the upper center of gravity, including a total of seven degrees of freedom of movement, express a person's upper body movement. In this way, emotion recognition can be performed based on the user's body language content.

For example, a photo or video containing a face taken by the user through the front or rear camera, when the user finishes the shooting, collect the user's facial image data; among them, the collection action can occur when the facial movement is the largest and most obvious, it can be Recognize and judge automatically.

Next, this embodiment adopts image emotion analysis technology to perform emotion recognition on facial images and output emotion results. The emotional result of this embodiment can include two levels of determination: the first level determines whether it contains emotional information, if it does not, the emotional result is no emotional information; if the first level determines that it contains emotional information, then enter the second level Judgment refers to a certain type of emotion, where the emotion type includes, but is not limited to, happiness, sadness, depression, anger, fear, tension, etc.

This embodiment can determine whether to obtain the corresponding emotion expression carrier according to the emotion result. When the emotion result is no emotion, the emotion icon is not generated; when the emotion result is a certain emotion type, it will be combined with the existing emotion at the same time. The icons are compared with the preset text library to provide users with emotional icons related to this type of emotion.

Furthermore, on the basis of filtering out the emotion icons related to a certain emotion type, the user selects the emotion icon he wants to use and adds it to the facial photo or video containing the face, and then a new emotion icon preview can be generated At this time, the user can choose to save the new emotion icon to the existing emotion icon set, or use it immediately by sending a message.

It should be noted that the step of acquiring the user’s input information by the user’s smart device in S101 of this embodiment may specifically include: the user’s smart device acquiring the user’s voice content through a microphone; and preprocessing the language content as述input information.

For example, when the user uses the device, he can directly perform voice input "Very annoying". At this time, this embodiment can acquire "Very annoying" through a microphone as input information.

Correspondingly, the step of performing emotion recognition on the user according to the input information in S102 of this embodiment may specifically include: using a voice emotion analysis technology to perform emotion recognition on the data of the voice content.

For example, when the user enters the voice information in the voice message input box through the microphone, when the user completes the input, the voice data input by the user is collected; then, the voice data based on voice emotion analysis technology is used to perform emotion recognition on the voice data and output the emotion result. Among them, the emotional result of this embodiment may include two levels of determination: the first level determines whether it contains emotional information, if it does not, the emotional result is no emotional information; if the first level determines that it contains emotional information, enter the second The second-level judgment refers to a certain type of emotion, where the emotion type includes but is not limited to happy, sad, depressed, angry, scared, nervous, etc.

Next, this embodiment can determine whether to obtain the corresponding emotion expression vector according to the emotion result. When the emotion result is no emotion, the emotion icon is not generated; when the emotion result is a certain type of emotion, it can be compared with the existing Emotion icons are compared to provide users with emotion icons related to this type of emotion; finally, emotion icons can be generated on the basis of existing emotion icons, and provided to users for preview and selection before use.

It should be noted that the step of acquiring the user’s input information by the user’s smart device in S101 of this embodiment may specifically include: the user’s smart device acquiring the user’s text through a text message input box; and preprocessing the text to As the input information.

Correspondingly, the step of performing emotion recognition on the user according to the input information in S102 of this embodiment may specifically include: performing emotion recognition on the data of the text text by using a text emotion analysis technology.

For example, the text information entered by the user in the text message input box, when the user completes the input, the text text data entered by the user is collected; then, the text sentiment analysis technology is used to perform emotion recognition on the text and the emotion result is output.

Emotional results can include two levels of judgment: the first level determines whether it contains emotional information, if not, the emotional result is no emotional information; if the first level determines that it contains emotional information, then it enters the second level of determination, that is, it is determined as A certain emotion type. In this embodiment, the emotion type may include, but is not limited to, happy, sad, depressed, angry, scared, nervous, and so on.

Furthermore, this embodiment judges whether to obtain the corresponding emotion expression vector based on the emotion result. For example, when the emotion result is no emotion, no emotion icon is generated; when the emotion result is a certain emotion type, the user will be provided with emotions related to this emotion type by comparing with existing emotion icons Icon. For example, when the emotional result is "angry", through comparison, an "angry" emotional icon will be generated based on the existing emotional icon. Finally, the result of the emotion icon can be generated on the basis of the existing emotion icon, and the result is provided to the user for preview and selection before use.

It is worth mentioning that after the step of generating the corresponding emotion icon according to the key information in this embodiment, it may further include: prompting the emotion icon to the user for preview; judging whether the user's confirmation is obtained Operation; if the user's confirmation operation is obtained, send and/or store the emotion icon.

For example, when the emotional result is "angry", the comparison will filter out all the "angry" expressions in the existing emotional icons, and filter out all the "angry" related content in the preset text/text library such as "The baby is angry." The user can directly select the emotion icon or text/text content by touching or clicking with the mouse, and then adding it to the original facial photo or video containing the face to generate an expressive emotion icon.

It should be noted that the emotion icon processed by the emotion icon processing method of this embodiment can be sent to a third party through an APP third-party application for communication and interaction, or stored locally or in the cloud.

Please refer to FIG. 2 next. This application also provides a user smart device. As one of the implementation manners, the user smart device includes a processor 21, and the processor 21 is configured to execute a computer program to implement the above Emotion icon processing method.

Specifically, the processor 21 is used to obtain input information of the user; the processor 21 is used to perform emotion recognition on the user according to the input information; the processor 21 is used to obtain corresponding information according to the result of the emotion recognition Key information; the processor 21 is configured to generate corresponding emotion icons according to the key information.

It should be noted that the processor 21 in this embodiment is used to obtain user input information, which may specifically include: the processor 21 is used to capture the user's body language content through a camera device; Processing as the input information.

For example, the processor 21 is configured to capture the user's body language content through multiple cameras such as front and rear cameras, and it may also perform continuous shooting and three-dimensional shooting for comparison and screening.

Further, the processor 21 in this embodiment is configured to preprocess the body language content as the input information, which may specifically include: performing action recognition on the body language content; acquiring according to the result of the action recognition A target part whose motion amplitude change value is greater than a preset threshold; the motion of the target part is used as the input information.

The processor 21 in this embodiment is used to recognize the user's emotions according to the input information, which may specifically include: firstly, modeling the joint points of the human body, and treating the human body as a rigid system with internal connections. Including bones and joints, the relative movement of bones and joints constitutes the change of the human body’s posture, that is, the usual description of actions. Among the many joints of the human body, according to the degree of emotional impact, the fingers and toes are ignored. The spine is abstracted into three joints of the neck, chest and abdomen. A human body model is summarized. The upper body includes the head, neck, chest, abdomen, two arms and two forearms, and the lower body includes two thighs and two lower legs; For the selected multiple emotional states, the human body is selected to express each emotional state under normal conditions, and the body response is analyzed in detail; since the human body is abstracted into a rigid body model, the first is the movement of the human center of gravity, which is divided into Forward, backward and natural state; in addition to the movement of the center of gravity, followed by the rotation of the joint points, the human body changes in motion, and the joint points related to emotions include the head, chest, shoulders and elbows. The corresponding motion is the head The bending of the chest, the rotation of the chest cavity, the swing and extension direction of the upper arm, and the bending of the elbow, these parameters combined with the movement of the upper center of gravity, including a total of seven degrees of freedom of movement, express a person's upper body movement. In this way, emotion recognition can be performed based on the user's body language content.

Next, in this embodiment, image emotion analysis technology may be used to perform emotion recognition on facial images and output emotion results. The emotional result of this embodiment can include two levels of determination: the first level determines whether it contains emotional information, if it does not, the emotional result is no emotional information; if the first level determines that it contains emotional information, then enter the second level Judgment refers to a certain type of emotion, where the emotion type includes, but is not limited to, happiness, sadness, depression, anger, fear, tension, etc.

It should be noted that the processor 21 in this embodiment is configured to obtain user input information, and specifically may further include: the processor 21 is configured to obtain the user's voice content through a microphone; and preprocess the language content to As the input information.

Correspondingly, the processor 21 in this embodiment is configured to perform emotion recognition on the user according to the input information, and may specifically include: the processor 21 is configured to perform emotion recognition on the data of the voice content by using voice emotion analysis technology. Recognition.

For example, the voice information entered by the user in the voice message input box through the microphone, when the user completes the input, the voice data input by the user is collected; then, voice-based emotion analysis technology is used to perform emotion recognition on the voice data and output the emotion result. Among them, the emotional result of this embodiment may include two levels of determination: the first level determines whether it contains emotional information, if it does not, the emotional result is no emotional information; if the first level determines that it contains emotional information, enter the second The second-level judgment refers to a certain type of emotion, where the emotion type includes but is not limited to happy, sad, depressed, angry, scared, nervous, etc.

It should be noted that the processor 21 in this embodiment is configured to obtain user input information, which may specifically include: the processor 21 is configured to obtain the user's text and text through a text message input box; Processing as the input information.

Correspondingly, the processor 21 of this embodiment is configured to perform emotion recognition on the user according to the input information, and may specifically include: the processor 21 is configured to perform emotion recognition on the data of the text text by using a text emotion analysis technology. Recognition.

For example, the text information entered by the user in the text message input box, when the user completes the input, the text text data entered by the user is collected; then, the processor 21 is used to perform emotion recognition on the text text by using text emotion analysis technology And output emotional results.

Furthermore, this embodiment judges whether to obtain the corresponding emotion expression vector based on the emotion result. For example, when the emotion result is no emotion, no emotion icon is generated; when the emotion result is a certain emotion type, the user will be provided with emotions related to this emotion type by comparing with existing emotion icons Icon. For example, when the emotional result is "angry", through the comparison, the "qi" emotional symbol will be generated based on the existing emotional symbol. Finally, the result of the emotion icon can be generated on the basis of the existing emotion icon, and the result is provided to the user for preview and selection before use.

In this embodiment, when the input information is body language content, the key information is emotional action information composed of eyes, mouth, hands, head, torso, feet, or any combination of two or more; When the input information is language content, the key information is emotional speech information composed of sentences, mood, words, scenes, or any combination of two or more; when the input information is text, the key information The information is emotional text information composed of sentences, words, scenes, contexts, or any combination of two or more. Preferably, the emotional icon is composed of the eyes, mouth, hands, head, torso, feet, or any two or more combination of emotional action information, sentence, tone, word, scene or any two Emotional speech information composed of one or more combinations, and/or, emotional text information composed of sentences, words, scenes, contexts, or any two or more combinations

It is worth mentioning that the processor 21 in this embodiment is configured to generate and obtain the corresponding emotion icon according to the key information, and then may also be used to prompt the user to preview the emotion icon; to determine whether the user is obtained If the user's confirmation operation is obtained, send and/or store the emotion icon.

In addition, the present application also provides a storage medium that stores a computer program, and when the computer program is executed by a processor, it is used to implement the emotion icon processing method described in the above embodiment.

The above are only the preferred embodiments of the application, and do not limit the application in any form. Although the application has been disclosed as the preferred embodiments, it is not intended to limit the application. Anyone familiar with the profession The technical personnel, without departing from the scope of the technical solution of the present application, can use the technical content disclosed above to make slight changes or modification into equivalent embodiments with equivalent changes, provided that the content of the technical solution of the present application is not deviated from the technical solution of the present application. Any simple modifications, equivalent changes and modifications made to the above embodiments by the technical essence still fall within the scope of the technical solutions of the present application.

Claims

An emotion icon processing method, characterized in that the emotion icon processing method includes the steps:

The user's smart device obtains the user's input information;

Perform emotion recognition on the user according to the input information;

Obtain corresponding key information according to the result of emotion recognition;

The corresponding emotion icon is generated according to the key information.
The emotional icon processing method according to claim 1, wherein the step of obtaining the user's input information by the user smart device specifically includes:

The user's smart device captures the user's body language content through the camera device;

Preprocessing the body language content as the input information.
The emotion icon processing method according to claim 2, wherein the step of preprocessing the body language content as the input information specifically comprises:

Perform action recognition on the body language content;

Obtain the target part whose motion amplitude change value is greater than the preset threshold according to the result of motion recognition;

Use the movement of the target part as the input information.
The emotional icon processing method according to claim 3, wherein the body language content is eyes, mouth, hands, head, torso, feet, or any combination of two or more actions.
The emotional icon processing method according to claim 1, wherein the step of obtaining the user's input information by the user smart device specifically includes:

The user's smart device obtains the user's voice content through the microphone;

Preprocessing the language content as the input information.
The emotion icon processing method according to claim 5, wherein the step of performing emotion recognition on the user according to the input information specifically comprises:

Using voice-based emotion analysis technology to perform emotion recognition on the data of the voice content.
The emotional icon processing method according to claim 1, wherein the step of obtaining the user's input information by the user smart device specifically includes:

The user’s smart device obtains the user’s text through the text message input box;

The text is preprocessed as the input information.
8. The emotion icon processing method according to claim 7, wherein the step of performing emotion recognition on the user according to the input information specifically comprises:

Using text-based sentiment analysis technology to perform sentiment recognition on the text data.
8. The emotion icon processing method according to any one of claims 1-8, wherein after the step of generating and obtaining the corresponding emotion icon according to the key information, the method further comprises:

Prompt the user to preview the emotion icon;

Determine whether the user's confirmation operation is obtained;

If the user's confirmation operation is obtained, the emotion icon is sent and/or stored.
8. The emotion icon processing method according to any one of claims 1-8, wherein the step of generating and obtaining corresponding emotion icons according to the key information specifically comprises:

Obtaining a corresponding emotion expression vector according to the key information;

The corresponding emotion icon is generated according to the emotion expression vector.
10. The emotional icon processing method according to claim 10, wherein the emotional expression carrier is an emotional picture, an emotional text, an emotional voice, or any combination of the three.
The emotion icon processing method according to any one of claims 1-8, wherein:

When the input information is body language content, the key information is emotional action information composed of eyes, mouth, hands, head, torso, feet, or any combination of two or more;

When the input information is language content, the key information is emotional speech information composed of sentences, mood, words, scenes, or any combination of two or more;

When the input information is text, the key information is emotional text information composed of sentences, words, scenes, context, or any combination of two or more;

Wherein, the emotional icon is composed of the eyes, mouth, hands, head, torso, feet, or any two or more combinations of emotional action information, sentence, tone, word, scene or any two Or the emotional speech information formed by the above combination, and/or the emotional text information formed by the sentence, word, scene, context or any combination of two or more.
The emotional icon processing method according to any one of claims 1-8, wherein the emotional icon is a static image of an expression, a small expression video, and an expression including pictures, text and/or voice used for communication. Dynamic emoticons.
The method for processing emotion icons according to claim 13, wherein when the emotion icons are sent to the receiver, they are also used to generate interaction or stress response effects according to the on-demand actions of the receiver.
A user smart device, wherein the user smart device includes a processor, and the processor is configured to execute a computer program to implement the emotion icon processing method according to any one of claims 1-14.