# US 12,604,062 B2 · Enhancing media consumption experience through generative AI-powered interactive companions > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,604,062 B2 - Document ID: US12604062B2 - Status: Granted - Kind: B2 - Application number: 18/625,277 - Filed: 2024-04-03 - Published: 2025-10-09 - Granted: 2026-04-14 - Earliest claimed: 2024-04-03 - Assignees: NEC Corporation Of America - Applicants: NEC Corporation Of America - Inventors: Tsvi Lev - Web page: https://patentgrep.com/patent/US12604062B2 ## CPC - H04N 21/478: Supplemental services, e.g. displaying phone caller identification, shopping application - G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) - H04N 21/44204: {Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched (monitoring of user activities for profile generation for accessing a video database G06F16/739; protecting generic digital content where the protection is independent of the precise nature of the content G06F21/10; arrangements for monitoring the use made of the broadcast services in broadcast systems H04H60/31)} - H04N 21/44218: {Detecting physical presence or behaviour of the user, e.g. using sensors to detect if the user is leaving the room or changes his face expression during a TV programme (methods or arrangements for recognising human body or animal bodies or body parts G06V40/10; methods or arrangements for acquiring or recognising human faces, facial parts, facial sketches, facial expressions G06V40/16; methods or arrangements for recognising movements or behaviour G06V40/20; arrangements for identifying users in broadcast systems H04H60/45)} - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract System and methods for enhancing the media consumption experience through generative AI-powered interactive companions. The system monitors media content in real-time, detects predefined cues, and generates contextually relevant and emotionally engaging reactions using a generative language model. These reactions are played and/or presented to the user, simulating an interactive conversation. The AI companion may adapt its responses based on user preferences, emotional states, and the context of the media content. By offering personalized interactions, emotional support, and/or contextual relevance, an immersive, engaging, and emotionally satisfying experience for users may be provided. The systems and methods may be used in entertainment, education, therapy, and social domains. ## Claims ### Claim 1 A method for enhancing media consumption experience, comprising: monitoring media content presented to a user in real-time to detect at least one predefined cue in a segment of the media content; in response to a detection of the at least one predefined cue, extracting relevant information from the segment of the media content containing the detected cue; assembling a prompt by combining a prompt template predefined for the detected cue with the extracted content from the media segment; providing the assembled prompt to a generative language model to generate a synthetic reaction to the extracted content, wherein the generative language model is trained to provide contextually appropriate and emotionally engaging responses; and presenting the generated synthetic reaction to the user in an interface facilitating a chat session between the user and at least one automated chat agent during the presentation of the media content; repeating the monitoring, the assembling, the providing, and the presenting for each of a plurality of different users during the presentation of the media content: detecting a plurality of reactions from a plurality of different users to the generated synthetic reaction; inputting the plurality of reactions to another generative language model to generate a further synthetic reaction; and presenting the further synthetic reaction to the plurality of reactions in the interface as part of the chat session, thereby promoting a multi participant session. ### Claim 2 (depends on 1) The method of claim 1, further comprising repeating the monitoring, the assembling, the providing, and the presenting for the duration of the media content, thereby providing an interactive and emotionally supportive experience for the user throughout a consumption of the media content. ### Claim 3 (depends on 1) The method of claim 1, further comprising personalizing at least one of the prompt template and responses of the generative language model based on at least one of one or more learned preferences and one or more emotional states of the user to alleviate a media consumption experience. ### Claim 4 (depends on 1) The method of claim 1, wherein the monitoring the media content comprises analyzing textual content of the media content. ### Claim 5 (depends on 1) The method of claim 1, wherein the generative language model is trained using data including at least one of reviews, comments, and forum discussions about a plurality of media content items. ### Claim 6 (depends on 1) The method of claim 1, further comprising: detecting a reaction of the user to the generated synthetic reaction; inputting the reaction to another generative language model to generate a further synthetic reaction; and presenting the further synthetic reaction to the user in the interface as part of the chat session. ### Claim 7 (depends on 1) The method of claim 1, further comprising adding the generated synthetic reaction to a summary summarizing a plurality of generated synthetic reactions created during the presentation of the media content. ### Claim 8 (depends on 1) The method of claim 1, further comprising: detecting a user engagement level with the chat session; adjusting a frequency of the presenting the generated synthetic reactions based on the detected user engagement level, wherein the frequency is increased when the user engagement level is high and decreased when the user engagement level is low. ### Claim 9 (depends on 1) The method of claim 1, wherein the at least one predefined cue is determined based on a script, transcript, or similar description of the contents of the media content. ### Claim 10 (depends on 1) The method of claim 1, wherein the generative language model is further trained using reviews, comments, or forum discussions related to the media content. ### Claim 11 (depends on 1) The method of claim 1, wherein the chat session includes interactions with other users consuming the same media content, and the generated synthetic reactions consider the discussions with these other users. ### Claim 12 (depends on 1) The method of claim 1, wherein the presenting of the generated synthetic reaction is performed using a voice that emulates a character from the media content. ### Claim 13 (depends on 1) The method of claim 1, further comprising: determining a physical or mental state of the user; and adjusting the timing or content of the generated synthetic reactions based on the determined physical or mental state. ### Claim 14 (depends on 1) The method of claim 1, wherein the monitoring the media content further comprises analyzing visual cues in the media content, including at least one of facial expressions, scene changes and transitions, visual objects and symbols, character movements and actions, or visual aesthetics and style. ### Claim 15 (depends on 1) The method of claim 1, further comprising: determining an objective for the chat session, wherein the objective is related to at least one of reducing user loneliness, assessing a mental state of the user, or educating the user; and generating the synthetic reactions in accordance with the determined objective. ### Claim 16 A system for enhancing media consumption experience and alleviating user loneliness, comprising: one or more storage units for storing a code and a generative language model; one or more processors adapted to execute the code to: monitor media content presented in real-time and detect at least one predefined cue in a segment of the media content; in response to a detection of the at least one predefined cue, extract relevant information from the segment of the media content containing the detected cue; assemble a prompt by combining a prompt template predefined for the detected cue with the extracted content from the media segment; provide the assembled prompt to the generative language model to generate a synthetic reaction to the extracted content, wherein the generative language model is trained to provide contextually appropriate and emotionally engaging responses; and instruct a presentation of the generated synthetic reaction to the user in an interface facilitating a chat session between the user and at least one automated chat agent during the presentation of the media content; detecting a confusion or a misunderstanding of the user based on the user's reaction to the generated synthetic reaction; providing clarification or additional information to address the confusion or the misunderstanding in the chat session. ### Claim 17 (depends on 16) The system of claim 16, wherein the one or more processors are further adapted to execute the code to: analyze audio content of the media content to detect the at least one predefined cue, wherein the at least one predefined cue includes at least one of a tone, a pitch, an intensity, or an emotion in the audio content. ### Claim 18 (depends on 16) The system of claim 16, wherein the one or more processors are further adapted to execute the code to: generate a summary of the chat session, the summary including key points, interesting reactions, or decisions made during the chat session; store the summary for future reference or sharing with others. ### Claim 19 (depends on 16) The system of claim 16, wherein the one or more processors are further adapted to execute the code to: identify a product, a service, or a topic relevant to the media content; generate a recommendation or an advertisement related to the identified product, service, or topic; present the recommendation or the advertisement to the user during the chat session at an appropriate time based on the context of the media content and the chat session. ### Claim 20 (depends on 16) The system of claim 16, wherein the one or more processors are further adapted to execute the code to: determine a point in the media content associated with a specific event or scene; and select the prompt template based on the determined point in the media content. ### Claim 21 (depends on 16) The system of claim 16, wherein the one or more processors are further adapted to execute the code to: receive a user response to the generated synthetic reaction; provide the user response as input to the generative language model to generate a follow-up synthetic reaction; and present the follow-up synthetic reaction to the user in the chat session. ### Claim 22 A method for enhancing media consumption experience, comprising: identifying media content currently presented to a user; in response to a detection of at least one predefined cue in content of a segment of the media content, selecting one of a plurality of prompt templates predefined for the at least one predefined cue; providing a prompt assembled from the selected prompt template and content extracted from the segment of the media content to a generative language model to induce a synthetic reaction to the content extracted from the segment; and presenting the synthetic reaction in a chat session held between the user and at least one automated chat agent about the media content during a presentation of the media content; detecting a confusion or a misunderstanding of the user based on the user's reaction to the generated synthetic reaction; providing clarification or additional information to address the confusion or the misunderstanding in the chat session.