CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority of U.S. Provisional Patent Application No. 63/335,900 filed on April 28, 2022, entitled “RELATIONSHIP GRAPHS FOR ARTIFICIAL INTELLIGENCE CHARACTER MODELS.” The subject matter of aforementioned application is incorporated herein by reference in its entirety for all purposes.

TECHNICAL FIELD

[0002] This disclosure generally relates to artificial intelligence (Al)-based character models. More particularly, this disclosure relates to providing interactions of an Al-based character model with users.

BACKGROUND

[0003] Virtual characters are widely used in various software applications, such as games, metaverses, social media, messengers, video communication tools, and online training tools. Some of these applications allow users to interact with virtual characters. However, existing models of virtual characters are typically developed for specific applications and do not allow integration with other applications and environments. Moreover, existing virtual character models are typically based on descriptions of specific rules and follow specific logic.

[0004] Parameters of conventional virtual character models typically remain unchanged for the entire interaction between users and virtual characters. This approach results in virtual character models that lack the ability to adapt their interactions and conversations to a specific user. Specifically, conventional virtual character models lack the ability to adjust their behavioral characteristics based on changing relationships with the specific user, between the user and other virtual characters, and between the user and other users. Example relationships include topics previously discussed by users and virtual characters, an attitude of a virtual character to a user, such as friendly or hostile, and the like. Accordingly, tools are needed that would allow virtual character models to change their interactions with users based on changing relationships between the users and the virtual characters. SUMMARY

[0005] This section is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description section. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0006] In one example embodiment, a computing platform for providing interactions of an Al-based character model with users is provided. The computing platform may include a processor and a memory storing instructions to be executed by the processor. The computing platform may be configured to receive a message from a user of a client-side computing device. The client-side computing device may be in communication with the computing platform. The computing platform may be further configured to retrieve, from a graph, information concerning relationships between the Al character model and the user. The graph may include a first node associated with the Al character model, a second node associated with the user, and an edge between the first node and the second node. The edge may be associated with the information concerning relationships between the Al character model and the user. The computing platform may be further configured to generate, based on the information concerning relationships and the message, an action associated with the Al character model. The computing platform may be further configured to cause the Al character model to perform the action in a virtual environment provided to the user via the client- side computing device.

[0007] In another example embodiment, a method for providing interactions of an Al character model with users is provided. The method may be implemented by a processor of a computing platform for providing interactions of an Al character model with users. The method may commence with receiving a message from a user of a client-side computing device. The client-side computing device may be in communication with the computing platform. The method may proceed with retrieving, from a graph, information concerning relationships between the Al character model and the user. The graph may include a first node associated with the Al character model, a second node associated with the user, and an edge between the first node and the second node. The edge may be associated with the information concerning relationships between the Al character model and the user. The method may further include generating, based on the information concerning relationships and the message, an action associated with the Al character model. The method may proceed, causing the Al character model to perform the action in a virtual environment provided to the user via the client-side computing device. [0008] According to another example embodiment, provided is a non-transitory computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the one or more processors to perform steps of the method for providing interactions of an Al character model with users.

[0009] Additional objects, advantages, and novel features of the examples will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following description and the accompanying drawings or may be learned by production or operation of the examples. The objects and advantages of the concepts may be realized and attained by means of the methodologies, instrumentalities and combinations particularly pointed out in the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements.

[0011] FIG. 1 illustrates an environment within which methods and systems for providing interactions of an Al character model with users can be implemented.

[0012] FIG. 2 is a block diagram illustrating a platform for generating an Al character model, according to an example embodiment.

[0013] FIG. 3 provides additional details for an Al character model, in accordance with an example embodiment.

[0014] FIG. 4 is an architecture diagram that shows using a surrounding architecture of an Al character model to control an output and behavior generated by large language models (LLMs), according to an example embodiment.

[0015] FIG. 5 is a detailed architecture diagram showing a surrounding architecture of an Al character model, according to an example embodiment.

[0016] FIG. 6A is a detailed architecture diagram showing a surrounding architecture of an Al character model, according to an example embodiment.

[0017] FIG. 6B is a detailed architecture diagram showing a surrounding architecture of an Al character model, according to an example embodiment.

[0018] FIG. 7A shows an architecture diagram illustrating Al character models with goal-oriented behavior, according to an example embodiment. [0019] FIG. 7B shows an architecture diagram illustrating Al character models with goal-oriented behavior, according to an example embodiment.

[0020] FIG. 8 is a block diagram illustrating a narrative structure that shows a context of scenes used to distinguish context for goals, according to an example embodiment.

[0021] FIG. 9 is a block diagram illustrating a structure of goals within scenes, according to an example embodiment.

[0022] FIG. 10 is a schematic diagram showing a relationship graph, according to an example embodiment.

[0023] FIG. 11 is a flow chart illustrating a method for providing interactions of an Al character model with users, according to an example embodiment.

[0024] FIG. 12 is a high-level block diagram illustrating an example computer system, within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein can be executed.

DETAILED DESCRIPTION

[0025] The following detailed description of embodiments includes references to the accompanying drawings, which form a part of the detailed description. Approaches described in this section are not prior art to the claims and are not admitted to be prior art by inclusion in this section. The drawings show illustrations in accordance with example embodiments. These example embodiments, which are also referred to herein as “examples,” are described in enough detail to enable those skilled in the art to practice the present subject matter. The embodiments can be combined, other embodiments can be utilized, or structural, logical, and operational changes can be made without departing from the scope of what is claimed. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope is defined by the appended claims and their equivalents.

[0026] The approaches described in this section could be pursued but are not necessarily approaches that have previously been conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.

[0027] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be apparent to one of ordinary skill in the art that the present disclosure may be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0028] Embodiments of the present disclosure are directed to a platform for generating Al character models and performing interactions between the Al character models and users. In one example embodiment, the platform may receive a description of a character and generate an Al character model capable of interacting with users verbally and through emotions, gestures, actions, and movements. The description can be provided as natural language describing a role, motivation, and environment of an Al character. The platform may utilize a common knowledge concerning the Al character to train the Al character model in order to interact with the users. The Al character model may evolve its characteristics, change emotions, and acquire knowledge based on conversations with the users.

[0029] The Al character model may utilize a LLM in conversations with the users. The platform may apply restrictions, classifications, shortcuts, and filtering in response to user questions to form an appropriate request to the LLMs and to obtain more effective and appropriate responses to user questions and messages. For example, prior to sending a request to the LLM, the platform may classify and filter the user questions and messages to change words based on the personalities of Al characters, emotional states of Al characters, emotional states of users, context of a conversation, scene and environment of the conversation, and so forth. Similarly, the platform may adjust the response formed by the LLM by changing words and adding fillers based on the personality, role, and emotional state of the Al character. The Al character model may change emotions based on the role of the Al character and in response to emotions of the user.

[0030] The platform may include integration interfaces, such as application programming interfaces (APIs), allowing external applications to use the Al character model. The Al character models generated by the platform can be used in game applications, virtual events and conversations, corporate trainings, and so on.

[0031] Some embodiments of the present disclosure relate to a system and a method for providing interactions of an Al character model with users. The system and the method may be integrated into the platform for generating Al character models. The system and the method may track relationships between Al characters and users and generate responses of the Al characters in conversations with the users and change characteristics of the interactions of the Al characters with users based on the relationships.

[0032] In an example embodiment, the method may commence with receiving a message from a user of a client-side computing device. The client-side computing device may be in communication with the computing platform. Upon receiving the message from the user, the method may proceed with retrieving information concerning relationships between the Al character model and the user. The Al character models may be presented to the user in a form of Al characters in a virtual environment provided to the user via the client-side computing device.

[0033] The information concerning relationships may be retrieved from a graph, also referred herein to as a relationship graph. The graph may include a plurality of nodes and a plurality of edges connecting the nodes. The Al characters and users may be associated with nodes of the graph and the information concerning relationships between the Al characters and the users may be associated with edges of the graph. The stored information concerning relationships may be retrieved and used for providing interactions of the Al character models with the users in the virtual environment.

[0034] In an example embodiment, the graph may include a first node associated with the Al character model, a second node associated with the user, and an edge between the first node and the second node. The edge may be associated with the information concerning relationships between the Al character model and the user. In an example embodiment, the information concerning relationships may include data concerning historical data of conversations between the Al character model and the users, a current conversation between an Al character and the user, parameters associated with Al characters, parameters associated with the user, and the like.

[0035] Upon retrieving the information concerning relationships, the method may proceed with generating, based on the information concerning relationships and the message, an action associated with the Al character model. The method may proceed with causing the Al character model to perform the action in the virtual environment. The client-side computing device may present the action to the user.

[0036] Generally, each node of the graph represents a specific Al character or a specific user. The relationship between the user and the Al character is represented by an edge between the node associated with the user and the node associated with the Al character. The relationship represents a specific instantiation of the brain (i.e., specific data in a memory storing information associated with the Al character model) of the Al character for a specific user. The brain of the Al includes information known to the Al character in the virtual environment and used for generating interactions of the Al character with the user. The virtual environment may have a plurality of users playing a game (e.g., Alice in Wonderland) and each of users may be associated with a different instance of the brain of the same Al character. Depending on how each user interacts with the Al character, the graph of relationships of each specific user with the Al character changes.

[0037] Referring now to the drawings, FIG. 1 illustrates an environment 100 within which methods and systems for providing interactions of an Al character model with users can be implemented. The environment 100 may include a client-side computing device 102 associated with a user 104, a computing platform 106 for providing an Al character model, and a data network shown as a network 108. The computing platform 106 and client-side computing device 102 (also referred to herein as a client) may communicate via the network 108.

[0038] The client-side computing device 102 may include, but is not limited to, a smartphone, a laptop, a personal computer, a desktop computer, a tablet computer, a phablet, a personal digital assistant, a mobile telephone, a smart television set, a personal computing device, and the like. The computing platform 106 may include a processor 110 and a memory 112 storing instructions to be executed by the processor 110. [0039] The network 108 can refer to any wired, wireless, or optical networks including, for example, the Internet, intranet, a Local Area Network (LAN), a Personal Area Network, Wide Area Network (WAN), a Virtual Private Network, a Wi-Fi® network, cellular phone networks (e.g., a Global System for Mobile (GSM) communications network, a packet switching communications network, a circuit switching communications network), Bluetooth™ radio, an Ethernet network, an IEEE 802.11 -based radio frequency network, a Frame Relay network, an Internet Protocol (IP) communications network, or any other data communication network utilizing physical layers, link layer capability, or network layers to carry data packets, or any combinations of the above-listed data networks. In some embodiments, the network 108 may include a corporate network, a data center network, a service provider network, a mobile operator network, or any combinations thereof. [0040] The computing platform 106 may be associated with an Al character model (shown in detail in FIG. 2). The Al character model may be configured to generate Al-based characters, also referred herein to as Al characters. The user 104 may interact with the Al characters via the client- side computing device 102 in a virtual environment associated with the computing platform 106 and generated by the client-side computing device 102 for presenting to the user 104. The computing platform 106 is shown in detail in FIG. 2 as a platform 200.

[0041] FIG. 2 illustrates a platform 200 for generating Al character models, according to an example embodiment. The platform 200 may include a studio 204, an integration interface 206, and an Al character model 202. Al character models are also referred to herein as Al-based character models. The studio 204 and the integration interface 206 may be in communication with data sources 226. The data sources 226 may include online search services. The data sources 226 may include a set of clusters each associated with a type of a feature of an Al character.

[0042] In one example embodiment, the studio 204 may receive, via a user interface, a character description 208 of an Al character. The studio 204 may generate, based on the character description 208, an Al character model 202 corresponding to the Al character

[0043] The character description 208 can be provided using a natural human language. The character description may include a description of an Al character similar to a description of a character to be played that can be provided to a real actor. The user interface of the studio 204 may include input fields allowing a developer to enter different aspects (i.e., parameters) of the Al character. Each input field may define a part of the brain of the Al character.

[0044] The input fields may include a text field for entering a core description of the Al character. An example core description can include “Buddy is a kind young man from Argentina.” The input fields may include a text field for entering a motivation of the Al character. An example motivation may include “Buddy likes to dance.”

[0045] The input fields may also include a text field for entering common knowledge and facts that the Al character may possess. For example, the field for common knowledge may include “orcs from Mordor; orcs like to eat hobbits.”

[0046] The input fields may include fields for selecting an avatar and voice of the Al character. The input fields may include fields for defining memory and personality features of the Al character.

The input fields may also include a text field describing the scene and environment in which the Al character is placed. For example, the text field for the scene may include “savanna,” “city,” “forest,” “bar,” and so forth.

[0047] The integration interface 206 may receive a user input 210, environment parameters 212, and events 214 and generate, based on the Al character model 202, a model output 216.

[0048] The user input 210 may include voice messages of a user. The voice messages may include phrases commonly used in conversations. The integration interface 206 may generate, based on the voice messages, requests and provide the request to the Al character model 202 to generate the model output 216. In an example embodiment, the requests may include text messages verbalized by the user and an emotional state of the user.

[0049] The model output 216 may include verbal messages 218, gestures 220, emotions 222, and movements 224. The verbal messages 218 may include responses to the voice messages of the user. The gestures 220 may include specific hand and facial movements of the Al character, either accompanying the verbal messages 218 or occurring without the verbal messages 218. Gestures may include, for example, waving goodbye, nodding to indicate agreement, or pointing to indicate a direction. Gestures are typically intentional and have a specific meaning that is understood by those familiar with the culture or context in which they are used. Emotions 222 may include intonations of the voice of the Al character while uttering the verbal messages 218 or facial expressions of the Al character. Movements 224 may refer to the overall movements and postures of the body of the Al character, including the position of the arms, legs, and torso. The movements 224 can be used to convey a range of emotions and attitudes, such as confidence, relaxation, or nervousness.

Movements 224 can be both intentional and unintentional.

[0050] FIG. 3 provides additional details of an Al character model 300, in accordance with an example embodiment. The Al character model 300 may include a set of models including an avatar 302, a language model 304, a gesture model 306, an emotional model 308, a behavioral model 310, and the like. The models may include machine learning models. In some embodiments, the models can be implemented as artificial neural networks. The Al character model 300 can include runtime parameters 312 and design parameters 314.

[0051] The design parameters 314 may correspond to settings for personality and general emotions of an Al character. The design parameters 314 can be generated based on character description 208 received via the studio 204 shown in FIG. 2.

[0052] The runtime parameters 312 may correspond to an emotional state of an Al character. The emotional state can be changed based on conversations with the user, elements in the scene, the surrounding environment in which the Al character is currently present, and so forth.

[0053] The avatar 302 may include a three-dimensional body model rendering the Al character. In some embodiments, the avatar 302 can be created using applications currently available on the market.

[0054] The language model 304 can be based on a LLM. The LLM is a machine learning algorithm that can recognize, predict, and generate human languages on the basis of very large text-based data sets. The language model 304 may form a request for the LLM, receive a response from the LLM, and process the response from the LLM to form a response to voice messages of the user. The request for the LLM can include classification and adjustment of the text requests from the integration interface 206, according to the current scene, environmental parameters, an emotional state of the Al character, an emotional state of the user, and current context of the conversation with the user. Processing of the response from the LLM may include filtering of the response to exclude unwanted words, verifying relevancy of the response, changing the words in the response, and adding fillers to phrases according to the personality of Al characters. In other embodiments, the language model 304 may also retrieve data from available sources, such as Wikipedia® or Game Wikipedia®, to generate the response.

[0055] The gesture model 306 may generate a movement of the body of the Al character based on the response to the user, an emotional state of the Al character, and current scene parameters. For example, the Al character may turn to the user and raise a hand in response to a greeting from the user. The greeting gestures can differ based on scenes and environments.

[0056] The emotional model 308 may track the emotional state of the Al character based on the context of the conversation with the user, an emotional state of the user, a scene and environmental parameters, and so forth. [0057] The behavioral model 310 may track and change behavioral characteristics of the Al character as a result of conversations with users or changes in the environment and scenes during a predetermined time period.

[0058] In general, the LLM can statistically suggest a continuation to any input provided to the LLM. If a conversation is started by using the LLM, the LLM may propose the next step for the conversation. For example, if a conversation includes a story related to some topic, the LLM may propose the next line for the story. One of the key characteristics of LLMs is the fact that LLMs are large. In particular, the LLMs are trained on vast amounts of data. When used in conversations, the LLMs can statistically suggest some text determined by the LLMs to be meaningful in the next step of the conversation. Therefore, the LLMs conventionally build the conversation based on the text itself.

[0059] FIG. 4 is an architecture diagram 400 that shows using a surrounding architecture of an Al character model to control an output and behavior generated by LLMs, according to an example embodiment. The main steps implemented to control the output and behavior of Al characters using the Al character model include an input step 402 (step A), a transformation step 404 (step B), an orchestration step 406 (step C), and a generation step 408 (step D). The input step 402 includes providing a connection with a client and performing input streaming. The transformation step 404 includes pre-processing and transforming an incoming data stream. The orchestration step 406 and the generation step 408 include processing and transforming an incoming data stream. Steps A-D are shown in detail in FIG. 5, FIG. 6A, and FIG. 6B.

[0060] FIG. 5 is a detailed architecture diagram 500 showing a surrounding architecture of an Al character model, according to an example embodiment. The input step (step A) may include establishing a connection between a client and a server, as shown in block 502. In an example embodiment, the client may include a user device associated with a user. The user may use the client device to interact with Al characters in a virtual environment using an application running on the user device. To establish the connection between the system of the present disclosure and the client, a server (e.g., a web server), a game client, and an application running on the user device may be provided. The server, the game client, and the application may be set up based on predetermined rules to enable streaming multimodal inputs from the client to the server, as shown in block 504. The inputs are shown in detail in FIG. 6A.

[0061] FIG. 6A and FIG. 6B show a detailed architecture diagram 600 that illustrates a surrounding architecture of an Al character model, according to an example embodiment. The connection established between the client and the server via predetermined protocols enables collecting a plurality of streams of inputs from the client. Each stream may be associated with one of multiple modalities. In an example embodiment, the modality may include a type of data. As shown in FIG. 6A, the inputs collected from the client may include text 602, audio 604, visuals 606, events 608, actions 610, gestures (not shown), and so forth.

[0062] Referring again to FIG. 5, the transformation step (step B) may include pre-processing the incoming streams of data in block 506. The streams of inputs may be pre-processed differentially based on the specific modality. The pre-processing may include converting the received inputs into a singular format. The pre-processing is shown in detail in FIG. 6A.

[0063] As shown in FIG. 6A, the text 602 is in the form of a natural language and may need no preprocessing. The audio 604 may be pre-processed using a speech to text conversion 612, in the course of which the audio input may be transformed into text. The visuals 606 may be pre- processed using a machine vision 614 based on object classification, environment understanding, and so forth.

[0064] The events 608 may include any event received from the client. An example event may include a button click in a game, an Al character moving a sword in a game, a button click in a web application, and so forth. The actions 610 may be received from an environment of Al characters with which the user interacts. An example action may include reacting to a horse riding by in an application, calling a web hook to retrieve information, and so forth. The events 608 and the actions 610 may be processed into client triggers 616. Based on the pre-processing, all inputs may be transformed into text and/or embeddings 618. The embeddings (also referred to as word embeddings) are word representations, in which words with similar meaning have a similar representation. Thus, a pre-processed data stream in the form of text and/or embeddings 618 may be obtained upon pre-processing of the received inputs.

[0065] Referring again to FIG. 5, the transformation step (step B) may further include running the pre-processed data through a series of machine learning models that represent different elements of cognition and producing intermediate outputs, as shown in block 508. Processing the data using the series of machine learning models is shown in detail in FIG. 6A.

[0066] As shown in FIG. 6A, the text and/or embeddings 618 may be passed through a plurality of machine learning models shown as heuristics models 620. The processing of the text and/or embeddings 618 using the heuristics models 620 may include passing the text and/or embeddings 618 through a goals model 622, a safety model 624, an intent recognition model 626, an emotion model 628, an events model 630, and a plurality of further heuristics models 632.

[0067] The goals model 622 may be configured to process the text and/or embeddings 618 and recognize, based on what was said by the user or the Al character, what goals need to be activated. The safety model 624 may be configured to process the text and/or embeddings 618 and filter out unsafe responses. The intent recognition model 626 may be configured to process the text and/or embeddings 618 and determine what a player (i.e., a user) intends to do and use an intent to trigger one or more events at a later point of interaction of the player with Al characters in the game.

[0068] The emotion model 628 may be configured to process the text and/or embeddings 618 and update, based on what the player said, the emotions of the Al character. The events model 630 may be configured to process the text and/or embeddings 618 and determine the events. The events may act as triggers for performing an action based on predetermined rules. For example, a predetermined rule may include a rule according to which when the player steps into a specific location (the event) near the Al character, the Al character takes a predetermined action.

[0069] Upon the processing of the data, the heuristics models 620 may provide intermediate outputs. Each of the intermediate outputs provided by the heuristics models 620 may be a differential element. Specifically, the goals model 622, the safety model 624, the intent recognition model 626, the emotion model 628, and the events model 630 may each provide a specific sort of a separate element. The separate elements need to be orchestrated by composing together into a specific templated format.

[0070] Referring again to FIG. 5, the orchestration step (step C) may include composing the intermediate outputs received from the heuristics models into templated formats for ingestion by LLMs and animation, gesture, and action models in block 510. Upon composing the intermediate outputs into a template, the composed outputs may be fed into primary models representing elements of multimodal expression, as shown in block 512. The orchestration step (step C) is further shown in detail in FIG. 6B.

[0071] As shown in FIG. 6B, the orchestration step (step C) may include formatting and representation 634 of the intermediate outputs received from the heuristics models. Upon being formatted, the composed data may be sent to another series of Al models. Specifically, the composed data received in block 510 shown in FIG. 5 may include dialogue prompts 636, active goals and actions 638 (i.e., what goals and actions need to be active based on what was said or done by the user or the Al character), animation and gesture state 640 (i.e., what gestures or animations need to be active depending on the emotional state and the goal), narrative triggers 642, voice parameters 644, and so forth. The dialogue prompts 636 may be provided to a LLM 646. The active goals and actions 638 may be provided to a goals and actions model 648, the narrative controller 650, and the animation and gesture model 652. The animation and gesture state 640 may be provided to the goals and actions model 648, the narrative controller 650, and the animation and gesture model 652.

[0072] The narrative triggers 642 may be provided to the goals and actions model 648, the narrative controller 650, and the animation and gesture model 652. An example of the narrative triggers 642 may include words "I want to be in the investigation" said by the player. The goals and actions model 648, the narrative controller 650, and/or the animation and gesture model 652 may receive this narrative trigger and change the storyline and progress forward in the game.

[0073] The voice parameters 644 may be used for enacting the voice in the virtual environment. For example, if the Al character is angry, the voice parameter “angry” may be used to change the voice of the Al character in the game. If the state of the Al character changes to very forceful, the state can be shown by changing the voice of the Al character.

[0074] Referring again to FIG. 5, the generation step (step D) may include using primary models and systems to generate final behavior-aligned data outputs in block 514. The generation step (step D) may further include streaming outputs through predetermined protocols to the client and applying final transformations in block 516. The generation step (step D) is further shown in detail in FIG. 6B.

[0075] As shown in FIG. 6B, the LLM 646 is a model used to generate a dialogue output 654. The goals and actions model 648 and the narrative controller 650 both decide what needs to be sent to the client side. The client side may be represented by a client engine, a game engine, a web application running on a client-side computing device, and the like. The goals and actions model 648 and the narrative controller 650 may decide what needs to be enacted on the client side. The animation and gesture model 652 may decide what animations or gestures need to be activated on the client side to enact the behavior of Al characters. Therefore, the goals and actions model 648, the narrative controller 650, and the animation and gesture model 652 provide client-side narrative triggers 656 and animation controls 658. The dialogue output 654, the client-side narrative triggers 656, and the animation controls 658 provide the dialogue, the events, the client-side triggers, and the animations that need to be enacted on the client side. [0076] The dialogue output 654, the client-side narrative triggers 656, the animation controls 658, and the voice parameters 644 may be processed using text to speech conversion 660. The output data obtained upon applying the text to speech conversion 660 are sent as a stream to the client 662. The game engine animates the Al character based on the received data to provide the generative behavior of the Al character. The animating may include, for example, instructing the Al character on what to say, how to move, what to enact, and the like.

[0077] FIG. 7A and FIG. 7B show an architecture diagram 700 illustrating Al character models with goal-oriented behavior, according to an example embodiment. The Al character models may include generative models configured to follow sequential instructions for dialog and actions that are driven by a specific purpose or intent for Al-driven characters. FIG. 7A shows possible user inputs 702 and input impact for goals model 704. The possible user inputs 702 include fields that are exposed to the user and can be changed by the user in the studio. The input impact for goals model 704 includes impacts of each user input on the goals model.

[0078] Compared to general language models that provide general goals for Al characters, the goals model enables providing specific goals. FIG. 7A shows that each type of configuration caused by the possible user inputs 702 may influence the goals and actions of the Al character. More specifically, the Al character personality and background description 706 selected by the user has an impact on the constitution of Al character personality and style, which biases the reason for which, and manner in which, the Al character pursues goals, as shown in block 708. Therefore, the Al character personality and background description 706 may influence how the Al character enacts its goals. For example, if the Al characters are Alice in Wonderland versus Jack Sparrow, the Al characters may have the exact same goal (e.g., to show their house to a player). However, the Al characters may show their houses in completely different ways because the Al characters represent two different people.

[0079] The motivations 710 received from the user may structure top-level motivations that underlie the reasoning for all Al character behavior and directions, as shown in block 712. Therefore, the motivations 710 may effectively determine why this Al character is pursuing this goal, i.e., determine the top-level motivation of the Al character. For example, the motivation of Alice in Wonderland is to get home. The goals of Alice are to ask the Mad Hatter what he knows about Wonderland. These goals may be determined and provided to the top level motivation. [0080] Flaws and challenges 714 selected by the user allow establishment of flaws and challenges for the Al character, which may influence, motivate, or hinder goal enactment by the Al character, as shown in block 716.

[0081] An identity profile 718 selected by the user may specify elements of an Al character (e.g., role, interests) which may have an influence on how the Al character pursues goals (e.g., a policeman trying to uncover information differently from a salesperson), as shown in block 720. The flaws and challenges 714 and the identity profile 718 are ways of enacting so as to influence the goal more contextually. For example, the Al character is Indiana Jones and his flaw is that he is scared of snakes. The goal of the Al character is to cross a cavern covered in snakes. Therefore, based on the flaw, the Al character may say, "Oh, I'm so scared of snakes," and then achieve the goal. Therefore, the flaws and challenges 714 are used to add a context to the goal-oriented behavior of the Al character. The identity profile 718 is used similarly to further contextualize the goal-oriented behavior of the Al character. For example, the Al characters may include a police person (a first identity) and a salesperson (a second identity) both trying to uncover information, but the salesperson may do it very differently than the police person.

[0082] An emotional profile 722 received from the user may be used to establish an emotional profile of an Al character, such that the emotional profile may influence expression of goals, as shown in block 724. The emotional profile 722 may include the expression. For example, the introvertedness of the Al character may be turned up to make the Al character introverted, in which case if the Al character had to sell something or the Al character had to say something to someone, the Al character may be more nervous than if the Al character was extroverted.

[0083] Various parts of memories, such as a personal memory 726, world knowledge 730, and contextual knowledge 734 provide information that may be relevant to the pursuit of a goal. Specifically, the personal memory 726 may be used to provide an Al character with personal memories that may be brought up during the pursuit of a goal, as shown in block 728. For example, if the Al character remembers that the Al character recently was bitten by a dog and the goal is to go in and tie up a dog, the Al character may express fear or angst and say, "Oh, I can do that, but I'm really scared, I had this bad experience." Therefore, changing the behavior of the Al character based on the personal memory 726 makes the behavior more realistic.

[0084] The world knowledge 730 may be used to integrate information about the world to contextualize pursuit of the goal, as shown in block 732. The world knowledge 730 may be used to further contextualize the behavior of the Al character. For example, in a specific science fiction world, the Al character knows that all the police are corrupt in an area and working for an evil overlord. Therefore, the Al character may be scared or show more cautious when pursuing an investigation.

[0085] The contextual knowledge 734 may be processed to include information about an environment or context to contextualize pursuit of the goal, as shown in block 736. For example, if a volcano has just exploded and the Al character is asked to carry a girl to safety, the Al character may show more hurriedness and may be forceful to the girl, versus if that was not true, the Al character might pursue the goal differently.

[0086] Voice configuration 738 may be used to determine the configuration of voice in real-time, which can allow Al characters to show different expressions when pursuing a goal, as shown in block 740. For example, if the Al character is a fireman who is saving someone, it may be extremely loud in a burning building; therefore, the voice of the Al character may be made loud and forceful. The Al character may pursue the goal differently as compared, for example, the case when the Al character was doing the same actions in a courtroom.

[0087] Dialogue style controls 742 may be used to control a dialogue style of an Al character. The dialogue style may influence the manner and style of speech of the Al character, as shown in block 744. For example, the user may set the dialog style to be a modern day New York dialogue style or a Wild West style. In each of the styles, the Al character may use different words. For example, a Wild West bartender may use slang when selling a drink.

[0088] Goals and actions 746 received from the user may be processed to specify the goals that an Al character has per scene, and then set up the actions that the Al character has available to pursue the goal, as shown in block 748. Therefore, the goals and actions 746 specify the goals for the scene in which the Al character is currently present, the sequence of goals, and actions that the Al characters have to do to pursue the goals.

[0089] Animation triggers and controls 750 may include animations and gestures, which may determine which actual physical movements the Al character can take to pursue the goal, as shown in block 752. For example, the Al character is selling an item and needs to take the item off the shelf and show it to the player when selling.

[0090] The input impact for goals model 704 are provided to a plurality of Al models to generate a consequent behavior 754 due to goal configurations, as shown in FIG. 7B. More specifically, the LLM may determine what the Al character needs to say to enact the goal, as shown in block 756. The goals and actions model shown in block 758 is the controller for determining which goals need to be pursued and in which order, when is the goal confirmed as complete, and the like.

[0091] The narrative controller determines how the narrative progresses depending on how the Al character pursues the goal (the goal is successful or failed) and if the narrative shifts as a result of a success or a failure, as shown in block 760. For example, in a game an Al character is supposed to save a girl, but the Al character fails, and the girl dies. This failure to complete the goal may change the narrative. The narrative controller may send a trigger to change the behavior of the Al character based on this failure to the game engine.

[0092] The text to speech conversion model determines how the Al character speaks his lines (audio) to pursue the goal, as shown in block 762. The parameters to be changed may also include, for example, the dialogue style and voice configuration.

[0093] The animation and gesture model may determine what actual actions, animations, or gestures the Al character enacts to pursue the goal (e.g., smiling and taking an item off the shelf, picking up a girl to save her from a burning building), as shown in block 764.

[0094] The outputs obtained in blocks 756-764 may include a dialogue output (audio or text) 766, client side narrative triggers 768, and animation controls 770. The dialogue output (audio or text) 766, the client side narrative triggers 768, and the animation controls 770 may be provided to a client 772 (e.g., a client engine, a game engine, a web application, and the like).

[0095] FIG. 8 is a block diagram 800 illustrating a narrative structure that shows a context of scenes used to distinguish context for goals, according to an example embodiment. The narrative structure may include world/narrative settings 802 and world knowledge 804 (world knowledge for all Al characters in all scenes). The world/narrative settings 802 and the world knowledge 804 may be used to transition from one scene to another in a story. Therefore, a story or an experience associated with an Al character may happen as a series of scenes and transitions.

[0096] In an example embodiment, an Al character may exist in a scene 806. Based on the world/narrative settings 802 and the world knowledge 804, the scene 806 may be transitioned in block 808 into a scene 810 and a scene 812. The scene 810 may be transitioned in block 814 and the scene 812 may be transitioned in block 816 into a scene 818, a scene 820, and a scene 822.

[0097] FIG. 9 is a block diagram 900 illustrating a structure of goals within scenes, according to an example embodiment. Within each scene, for each specific Al character, there is a goal that the Al character has to pursue. A scene 902 may be driven by a plurality of parameters. The parameters may include scene and location knowledge 904, which may include world knowledge for all Al characters. The parameters may further include historical knowledge 906, which may include knowledge from previous scenes and from transition between the previous scene and the current scene. The parameters may further include relationships 908, which determine relations between Al characters 910, 920, and 922. Each of the Al characters 910, 920, and 922 may have contextual knowledge 912, i.e., scene-specific knowledge. Each of the Al characters 910, 920, and 922 may further have a goal set 914. The goal set 914 may include a plurality of goals 916. Each of the goals 916 may be associated with a plurality of actions 918 to be taken by the Al character to pursue the goals 916.

[0098] In an example embodiment, the scene 902 is a scene in which the Al character 910 is Indiana Jones who enters a cave (scene and location knowledge 904). The context is as follows: the Al character 910 knows that he is scared of snakes (contextual knowledge 912), but he is running away from enemies (contextual knowledge 912) and the Al character 910 now has the first goal 916 to run through the cave and escape the snakes. Therefore, the Al character 910 has actions 918 available to pursue the goal 916. The actions 918 may include running, asking for help, and the like. The next goal 916 of the Al character 910 may be to find the buried treasure. The last goal 916 may be to escape. For each of those goals 916, the Al character 910 has specific actions 918 that are available for the Al character 910 to pursue.

[0099] FIG. 10 is a schematic diagram showing a relationship graph 1000, according to an example embodiment. The relationship graph 1000 (also referred herein to as a graph) may include nodes and edges that connect the nodes. The nodes may be associated with Al character models and users. The edges may include information concerning relationships between the Al character models and the users. In an example embodiment, the information concerning relationships may include historical data of previous conversations between the Al character model and the users, a current conversation between an Al character and a user, parameters associated with Al characters, parameters associated with the user, and the like. In some embodiments, the relationships may be associated with previous actions performed by a user with respect to one or more Al characters, previous actions performed by an Al character with respect to a user, and so forth.

[00100] In an example embodiment, the relationship graph 1000 may include a first node associated with an Al character model 1002 and a second node associated with a user A 1004. The relationship graph 1000 may further have an edge between the first node and the second node. The edge may be associated with the information concerning relationships between the Al character model 1002 and the user A 1004. The information concerning relationships is shown as relationship data 1006. The relationship data 1006 may include information concerning relationships between the user A 1004 and other Al character models, for example Al character model 1008. The relationship data 1006 may also include information concerning relationships of the user A 1004 and user B 1012 and relationship of the user A 1004 and user C 1018.

[00101] The relationship graph 1000 may further include a third node associated with a further Al character model 1008 and a further edge between the second node (the user A 1004) and the third node. The further edge may be associated with further information concerning relationships between the further Al character model 1008 and the user A 1004. The further information concerning relationships is shown as relationship data 1010. The relationship data 1010 may include information concerning relationships between the user A 1004 and other Al character models, for example Al character model 1002. The relationship data 1010 may also include information concerning relationships of user A 1004 and user B 1012 and relationship of user A 1004 and user C 1018.

[00102] The relationship graph 1000 may further include a further node associated with a further user (shown as a user B 1012) and a further edge between the first node (the Al character model 1002) and the further node. The further edge may be associated with further information concerning relationships between the Al character model 1002 and the user B 1012. The further information concerning relationships is shown as relationship data 1016. The relationship graph 1000 may further have an edge between the third node (the further Al character model 1008) and the further node (the user B 1012). The edge may be associated with further information concerning relationships between the further Al character model 1008 and the user B 1012. The further information concerning relationships is shown as relationship data 1014.

[00103] The relationship graph 1000 may further include a further node associated with another user (shown as a user C 1018) and a further edge between the first node (the Al character model 1002) and the further node. The further edge may be associated with further information concerning relationships between the Al character model 1002 and the user C 1018. The further information concerning relationships is shown as relationship data 1020. The relationship graph 1000 may further have an edge between the third node (the further Al character model 1008) and the further node (the user C 1018). The edge may be associated with further information concerning relationships between the further Al character model 1008 and the user C 1018. The further information concerning relationships is shown as relationship data 1022. [00104] The Al character model 1002 may be presented to the user A 1004, the user B 1012, and the user C 1018 in the form of an Al character in a virtual environment. The further Al character model 1008 may be presented to the user A 1004, the user B 1012, and the user C 1018 in the form of a further Al character in the virtual environment.

[00105] Thus, the Al characters and users may be stored as nodes of the relationship graph 1000 and the information concerning relationships between the Al characters and the users may be stored as edges of the relationship graph 1000. The stored information concerning relationships may be retrieved and used for providing interactions of the Al character models with the users in a virtual environment.

[00106] Al character model 1002 may adjust conversations with user A 1004 based not only on relationship data 1006, but also on relationship data 1016 and relationship data 1020. Similarly, Al character model 1008 may adjust conversations with user A 1004 based not only on relationship data 1010, but also on relationship data 1014 and relationship data 1022. For example, Al character model 1002 may analyze relationship data 1012 and relationship 1020 to retrieve opinions of user B 1012 and user C 1018 on user A 1004, a specific characteristic of user A 1004, specific facts that user A 1004 may know, specific actions that user A 1004 performed previously, and so forth. Al character model 1002 may update, based on information retrieved from relationship data 1016 and relationship data 1020, relationship data 1006 and adjust the conversation with user A 1004 based on the updated relationship data 1006.

[00107] In an example embodiment, the Al character model may be associated with an Al character that is an onboarding character in a game. The onboarding character may be responsible for onboarding a user in the game and may be configured to introduce themselves; talk about the world; ask about user preferences, a name, or other user data; and answer any additional questions of the user. The Al character may be configured to have different relationships with other Al characters or have an attitude toward objects. The relationships of the Al character may depend on the personality of the user, interactions of the user or the Al character with other users and Al characters, motivation of the user or the Al character, and so on. In addition to the general memory of the Al character, a portion of the memory of the Al character may be based on information retrieved from a graph of relationships (shown as the relationship graph 1000) between the users and the Al characters. The relationship graph 1000 may be used to store the information concerning relationships of the user with the Al character and other Al characters and generate relevant responses of the Al character to the user and other Al characters during interaction of the Al character with the user. For example, the Al character model may store the information about another Al character acting aggressively with the Al character. If the Al character does not like another Al character, this information may be fed to the Al character model so that the Al character may respond accordingly.

[00108] In an example graph, one node represents a user and a plurality of further nodes represent Al characters with which the user interacts in a virtual environment. The node associated with the user is connected by an edge with each of the nodes associated with the Al characters. Depending on the interactions between the user and the Al characters, a state of the edge (i.e., information concerning relationships) between the two nodes may be updated. The information concerning relationships may indicate whether the user and the Al character are friends or enemies or have a neutral relationship, or indicate a continuous state of information concerning relationships, e.g., how friendly or angry the relationship is, and the like. A variable may be assigned to each type of information concerning relationships. The information concerning relationships may include a zero to a hundred score across each of the variables.

[00109] Accordingly, an edge relationship graph 1000 is a relationship between at least two nodes and includes information on who the Al characters (or a user and an Al character) are and on a specific relationship status between the Al characters and the users. Therefore, the edge of the graph may include a pairing between the nodes and may represent a status of the relationship between a user and an Al character (or between two Al characters).

[00110] The edge may contain both a continuous representation of the relationship between the user and the Al character and a discreet natural language description of the relationship in terms of historical data or historical descriptions. More specifically, the information concerning relationships may further include the description or a history of the relationship. For example, in the initial input data, the relationship is described as friendly and, in the previous conversation, a user and Alice (the Al character) spoke about their favorite animal. The information on the relationship being friendly and the information on the favorite animal may be included in the edge (i.e., stored in the memory associated with the edge) that represents the relationship between the user and the Al character.

[00111] The graph may be updated based on historical conversation between Al characters and users. The updates may be triggered based on emotional parameters of the conversations, actions or events that trigger a change in relationship, the content (e.g., the natural language content) updated based on the interaction, and the like. 1 [00112] In an example embodiment, events that can trigger updates of the edges of the graph may include a result of a conversation between a user and an Al character, appearance of a new Al character in a scene, or a result of a conversation between the user and the new Al character. For example, a user may say a phrase to a second Al character about a first Al character and, as a result, the relationship between the first Al character and the user may change.

[00113] In an example embodiment, in which a user plays the Alice in Wonderland game, the user may show up to see the Queen of Hearts and the Queen of Hearts may be angry at the user.

The user may have an interaction with the Queen of Hearts in which the user may be rude and abusive and may perform cruel actions with respect to the Queen of Hearts. Based on the rude and abusive interaction of the user with the Al character, a friendliness score or a kindness score associated with the relationship between the user and the Al character may be decreased.

[00114] In an example embodiment, the virtual environment may have a non-player character being a peasant in a village and the user may attack the peasant. Based on such aggressive actions of the user, the friendliness score associated with the relationship between the user and the Al character may be decreased.

[00115] In an example embodiment, a user may have a discussion with an Al character. The discussion may be a part of information relating to the relationship between the user and the Al character. Depending on the conversation between the user and Alice (the Al character), the information concerning relationships may be updated to include the information that the user and Alice discussed a specific topic (e.g., favorite animal).

[00116] FIG. 11 is a flow chart of a method 1100 for providing interactions of an Al character model with users, according to an example embodiment. In some embodiments, the operations may be combined, performed in parallel, or performed in a different order. The method 1100 may also include additional or fewer operations than those illustrated. The method 1100 may be performed by processing logic that may comprise hardware (e.g., decision making logic, dedicated logic, programmable logic, and microcode), software (such as software run on a general-purpose computer system or a dedicated machine), or a combination of both. The method 1100 may be implemented with a processor of a computing platform configured to provide the Al character model.

[00117] The method 1100 may commence in block 1102 with receiving, by the processor, a message from a user of a client-side computing device. The client-side computing device may be in communication with the computing platform. [00118] In block 1104, the method 1100 may include retrieving, by the processor, from a graph, information concerning relationships between the Al character model and the user. The graph may include a first node associated with the Al character model, a second node associated with the user, and an edge between the first node and the second node. The edge may be associated with the information concerning relationships between the Al character model and the user. The example information concerning relationships may include one of the following: “the Al character model is feeling angry with the user,” “the Al character model is friendly with the user,” “the Al character model is neutral with the user,” and so forth. In an example embodiment, the information concerning relationships may include a score. A lowest level of the score may correspond to a state in which the Al character model is hostile to the user. A highest level of the score may correspond to a further state in which the Al character model is friendly to the user. In some example embodiments, the information concerning relationships may include a topic of a previous conversation between the Al character model and the user.

[00119] In block 1106, the method 1100 may include generating, by the processor and based on the information concerning relationships and the message, an action associated with the Al character model. In block 1108, the method 1100 may proceed with causing, by the processor, the Al character model to perform the action in a virtual environment provided to the user via the client- side computing device.

[00120] In an example embodiment, the graph may include a third node associated with a further Al character model and a further edge between the second node and the third node. The further edge may be associated with further information concerning relationships between the further Al character model and the user. The action may be determined based on the further information concerning relationships.

[00121] In some example embodiments, the graph may include a third node associated with a further user and a further edge between the first node and the third node. The further edge may be associated with further information concerning relationships between the Al character model and the further user. The action may be determined based on the further information concerning relationships.

[00122] In an example embodiment, the action may include a verbal response provided by the Al character model to the user via the client-side computing device. In some example embodiments, the action may include a gesture executed by the Al character model in the virtual environment provided to the user via the client-side computing device. In an example embodiment, the action may include a motion of the Al character model in the virtual environment provided to the user via the client-side computing device. In some example embodiments, the action may be associated with one of the following: an angry sentiment, a happy sentiment, a neutral sentiment, and so forth.

[00123] In an example embodiment, the method 1100 may further include updating the information concerning relationships between the Al character model and the user based on data concerning an interaction between the Al character model and the user.

[00124] FIG. 12 is a high-level block diagram illustrating an example computer system 1200, within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein can be executed. The computer system 1200 may include, refer to, or be an integral part of, one or more of a variety of types of devices, such as a general-purpose computer, a desktop computer, a laptop computer, a tablet computer, a netbook, a mobile phone, a smartphone, a personal digital computer, a smart television device, and a server, among others. Notably, FIG. 12 illustrates just one example of the computer system 1200 and, in some embodiments, the computer system 1200 may have fewer elements/modules than shown in FIG. 12 or more elements/modules than shown in FIG. 12.

[00125] The computer system 1200 may include one or more processor(s) 1202, a memory 1204, one or more mass storage devices 1206, one or more input devices 1208, one or more output devices 1210, and a network interface 1212. The processor(s) 1202 are, in some examples, configured to implement functionality and/or process instructions for execution within the computer system 1200. For example, the processor(s) 1202 may process instructions stored in the memory 1204 and/or instructions stored on the mass storage devices 1206. Such instructions may include components of an operating system 1214 or software applications 1216. The software applications may include the studio 204, the integration interface 206, and the Al character model 202. The computer system 1200 may also include one or more additional components not shown in FIG. 12, such as a housing, a power supply, a battery, a global positioning system (GPS) receiver, and so forth.

[00126] The memory 1204, according to one example, is configured to store information within the computer system 1200 during operation. The memory 1204, in some example embodiments, may refer to a non-transitory computer-readable storage medium or a computer- readable storage device. In some examples, the memory 1204 is a temporary memory, meaning that a primary purpose of the memory 1204 may not be long-term storage. The memory 1204 may also refer to a volatile memory, meaning that the memory 1204 does not maintain stored contents when the memory 1204 is not receiving power. Examples of volatile memories include random access memories (RAM), dynamic random access memories (DRAM), static random access memories (SRAM), and other forms of volatile memories known in the art. In some examples, the memory 1204 is used to store program instructions for execution by the processor(s) 1202. The memory 1204, in one example, is used by software (e.g., the operating system 1214 or the software applications 1216). Generally, the software applications 1216 refer to software applications suitable for implementing at least some operations of the methods for providing interactions of an Al character model with users.

[00127] The mass storage devices 1206 may include one or more transitory or non-transitory computer-readable storage media and/or computer-readable storage devices. In some embodiments, the mass storage devices 1206 may be configured to store greater amounts of information than the memory 1204. The mass storage devices 1206 may further be configured for long-term storage of information. In some examples, the mass storage devices 1206 include non-volatile storage elements. Examples of such non-volatile storage elements include magnetic hard discs, optical discs, solid-state discs, flash memories, forms of electrically programmable memories (EPROM) or electrically erasable and programmable memories, and other forms of non-volatile memories known in the art.

[00128] The input devices 1208, in some examples, may be configured to receive input from a user through tactile, audio, video, or biometric channels. Examples of the input devices 1208 may include a keyboard, a keypad, a mouse, a trackball, a touchscreen, a touchpad, a microphone, one or more video cameras, image sensors, fingerprint sensors, or any other device capable of detecting an input from a user or other source, and relaying the input to the computer system 1200, or components thereof.

[00129] The output devices 1210, in some examples, may be configured to provide output to a user through visual or auditory channels. The output devices 1210 may include a video graphics adapter card, a liquid crystal display (LCD) monitor, a light emitting diode (LED) monitor, an organic LED monitor, a sound card, a speaker, a lighting device, a LED, a projector, or any other device capable of generating output that may be intelligible to a user. The output devices 1210 may also include a touchscreen, a presence-sensitive display, or other input/output capable displays known in the art.

[00130] The network interface 1212 of the computer system 1200, in some example embodiments, can be utilized to communicate with external devices via one or more data networks such as one or more wired, wireless, or optical networks including, for example, the Internet, intranet, LAN, WAN, cellular phone networks, Bluetooth radio, and an IEEE 902.11 -based radio frequency network, Wi-Fi networks®, among others. The network interface 1212 may be a network interface card, such as an Ethernet card, an optical transceiver, a radio frequency transceiver, or any other type of device that can send and receive information.

[00131] The operating system 1214 may control one or more functionalities of the computer system 1200 and/or components thereof. For example, the operating system 1214 may interact with the software applications 1216 and may facilitate one or more interactions between the software applications 1216 and components of the computer system 1200. As shown in FIG. 12, the operating system 1214 may interact with or be otherwise coupled to the software applications 1216 and components thereof. In some embodiments, the software applications 1216 may be included in the operating system 1214. In these and other examples, virtual modules, firmware, or software may be part of the software applications 1216.

[00132] Thus, systems and methods for providing interactions of an Al character model with users have been described. Although embodiments have been described with reference to specific example embodiments, it will be evident that various modifications and changes can be made to these example embodiments without departing from the broader spirit and scope of the present application. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.