US-12621539-B2 - Machine learning-based custom content generation for video streaming content systems and applications

US12621539B2US 12621539 B2US12621539 B2US 12621539B2US-12621539-B2

Abstract

In various embodiments, machine learning-based custom content generation for video streaming content systems and applications is provided. One or more of the embodiments described herein, among other things, provide for a machine learning/generative artificial intelligence (GAI)-based custom video generator model that may be used to synthesize targeted video streaming content (e.g., in real-time), based at least in part on user profile data from a user profile associated with a target user, and content data comprising an indication of one or more video content objectives directed at the target user. In some embodiments, a custom video content generator as disclosed herein may produce one or more prompts based on a set of content data and the user profile data, and apply the prompts to the video generator model to generate targeted video data that may be streamed to user equipment for presentation to the target user.

Inventors

Dharmendra Adsule

Assignees

T-MOBILE INNOVATIONS LLC

Dates

Publication Date: 20260505
Application Date: 20240611

Claims (17)

1 . A system for contextual customization of video content, the system comprising: one or more processors; and one or more computer-readable media storing computer-usable instructions that, when executed by the one or more processors, cause the one or more processors to: receive user profile data representing one or more characteristics of a user; receive a set of content data comprising an indication of one or more video content objectives; execute a context prompt server comprising a context prompt generator model configured to input at least the user profile data and the set of content data to generate one or more content generation prompts for a video generator model based at least on the user profile data and the set of content data, wherein the context prompt generator model infers one or more context prompts based on the user profile data, wherein the one or more content generation prompts includes the one or more context prompts, and apply the one or more content generation prompts to the video generator model to generate targeted video content data; and transmit the targeted video content data to a user equipment client application to cause the user equipment client application to present video content based on the targeted video content data.
2 . The system of claim 1 , the one or more processors further to: stream the targeted video content data to the user equipment client application using a streaming video format.
3 . The system of claim 1 , the one or more processors further to: generate the one or more content generation prompts to include a set of one or more content prompts representing the one or more video content objectives based at least on the set of content data; and generate the one or more content generation prompts to further include a set of one or more context prompts based at least on the user profile data, wherein the set of one or more context prompts define a set of features that form a setting to place the one or more video content objectives of the set of one or more content prompts within a context.
4 . The system of claim 3 , wherein the set of one or more context prompts represents a set of characteristics that defines one or more features that form a setting for at least one of events, scenes, dialogue, action, or plot devices associated with the targeted video content data.
5 . The system of claim 3 , wherein the set of one or more content prompts are further generated based on the user profile data.
6 . The system of claim 1 , wherein the video generator model comprises at least one of: a machine learning model, a generative artificial intelligence (GAI)-based machine learning model, a deep neural network (DNN), a generative adversarial network (GAN), or a variational autoencoder (VAE).
7 . The system of claim 1 , wherein the one or more processors to receive a request for streaming video content via a network connection from user equipment (UE) comprising the user equipment client application, wherein the set of content data is based at least in part on the request for streaming video content.
8 . The system of claim 1 , wherein the set of content data at least in part comprises campaign data received via a network from an advertising data campaign platform.
9 . The system of claim 1 , the one or more processors further to receive the user profile data via a network based on data from at least one of: a consumer data platform, a customer data platform, a customer experience platform, a relationship management system, user interactions with one or more websites, and interactions within a mobile application.
10 . The system of claim 1 , the one or more processors further to query a network server for the user profile data based on a target identifier associated with the user of the user equipment client application.
11 . The system of claim 1 , the one or more processors further to query a network server for the set of content data based at least in part on the user profile data.
12 . The system of claim 1 , the one or more processors further to execute a content prompt server comprising a content prompt generator model configured to input at least one of the user profile data and the set of content data, wherein the content prompt generator model infers one or more content prompts based on the at least one of the user profile data and the set of content data, wherein the one or more content generation prompts includes the one or more content prompts.
13 . A telecommunications network, the network comprising: an operator core network; at least one edge server coupled to a core network edge of the operator core network; at least one radio access network coupled to the operator core network, wherein the at least one radio access network establishes one or more communication links between the operator core network and one or more user equipment (UE); and at least one network function executed on one or more processors configured to perform one or more operations to: generate one or more content generation prompts for input to a video generator model based on a set of content data comprising an indication of one or more video content objectives, and user profile data representing one or more characteristics of a user defining a target of the one or more video content objectives, wherein the one or more content generation prompts include a set of one or more content prompts representing the one or more video content objectives based at least on the set of content data, and include a set of one or more context prompts based at least on the user profile data, wherein the set of one or more context prompts define a set of features that form a setting to place the one or more video content objectives of the set of one or more content prompts within a context; generate targeted video content data based on applying the one or more content generation prompts to the video generator model; and stream the targeted video content data via the at least one radio access network to a user equipment client application to cause the user equipment client application to present video content based on the targeted video content data.
14 . The network of claim 13 , wherein the at least one network function comprises a custom content generator executed by the one or more processors on the at least one edge server, wherein the custom content generator comprises the video generator model.
15 . The network of claim 13 , wherein the one or more processors comprise one or more controllers of a cloud computing environment, wherein the at least one network function comprises a custom content generator executing on a worker node cluster established by the one or more controllers, wherein the custom content generator comprises the video generator model.
16 . A method comprising: generating one or more content generation prompts for input to a video generator model based at least on one or more video content objectives and user profile data, wherein the user profile data represents one or more characteristics of a user defining a target of the one or more video content objectives, the one or more content generation prompts including a set of one or more content prompts representing the one or more video content objectives based at least on the one or more video content objectives, and including a set of one or more context prompts based at least on the user profile data, wherein the set of one or more context prompts define a set of features that form a setting to place the one or more video content objectives of the set of one or more content prompts within a context; generating targeted video content data based on applying the one or more content generation prompts to the video generator model; and transmitting the targeted video content data to a user equipment client application to cause the user equipment client application to present video content based on the targeted video content data.
17 . The method of claim 16 , wherein the set of one or more context prompts represents a set of characteristics that defines one or more features that form a setting for at least one of events, scenes, dialogue, action, or plot devices associated with the targeted video content data.

Description

BACKGROUND Targeted internet advertising, also known as online behavioral advertising, is a method used by advertisers to show advertisements that are more likely to be relevant to a user that views the advertisements. This is based on information about a user's behavior online, such as the websites visited by a user, the applications used by a user, and the searches and/or purchases performed by a user. Targeted advertising offers several benefits, both for the advertisers and the consumers. By focusing on a specific audience, advertisers can optimize their use of computing and network resources by reaching the people who are most likely to be interested in their products or services. SUMMARY This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used in isolation as an aid in determining the scope of the claimed subject matter. In contrast with presently available technologies, one or more of the embodiments described herein, among other things, provide for a machine learning/generative artificial intelligence (GAI)-based custom video generator model that may be used to synthesize targeted video streaming content (e.g., in real-time), based at least in part on user profile data from a user profile associated with a target user and content data comprising an indication of one or more video content objectives directed at the target user. In some embodiments, a custom video content generator as disclosed herein may produce one or more prompts based on a set of content data, and the user profile data, and apply the prompts to the video generator model to generate targeted video data that may be streamed to user equipment for presentation to the target user. BRIEF DESCRIPTION OF THE DRAWINGS Aspects of the present disclosure are described in detail herein with reference to the attached Figures, which are intended to be exemplary and non-limiting, wherein: FIG. 1 is a diagram illustrating an example machine learning-based custom content generator system, in accordance with some embodiments described herein; FIG. 2 is a diagram illustrating an example machine learning-based custom content generator system, in accordance with some embodiments described herein; FIG. 3 is a diagram illustrating an example network configurations for an example machine learning-based custom content generator system, in accordance with some embodiments described herein; FIG. 4 is a diagram illustrating an example telecommunications network environment comprising a network function for providing machine learning-based custom content generation as a network service, in accordance with some embodiments described herein; FIG. 5 is a flow chart illustrating an example method for machine learning-based custom content generation, in accordance with some embodiments described herein; FIG. 6 is an example computing device, in accordance with some embodiments described herein; and FIG. 7 is an example cloud computing platform, in accordance with some embodiments described herein. DETAILED DESCRIPTION In the following detailed description, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of specific illustrative embodiments in which the embodiments may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the embodiments, and it is to be understood that other embodiments may be utilized and that logical, mechanical, and electrical changes may be made without departing from the scope of the present disclosure. The following detailed description is, therefore, not to be taken in a limiting sense. Advertisements delivered based on online behavioral advertising are more likely to be relevant to a consumer's interests, which can lead to higher engagement rates. Consumers see advertisements that are more relevant to their interests, which can make their online experience more personalized and less cluttered with irrelevant advertisements. Targeted ads can help consumers discover new products or services that they might be interested in but weren't actively searching for. If consumers are in the market for a particular product or service, targeted ads can save them time by bringing relevant offers to their attention. When a consumer engages in online activities (e.g., browses the Internet, interacts with social media, purchases items and/or browses catalogs of online retailers, views newsfeeds, watches and/or listens to streaming content such as movies and/or podcasts, etc.) data about those online activities is collected through various means such as cookies, pixels, and device identifiers (device IDs). This data may then be collected and aggregated, for example, by a consumer data platform (CDP), to build a user profile as