Text-to-Content Model Selection System for Automated Design Element Suggestion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users creating content, such as slide presentations, face inefficiencies in incorporating design elements like images and emojis, as they need to search, capture, and insert these elements manually, which is time-consuming and bandwidth-intensive.

Innovation Solution

A system that uses machine-learned text-to-content models to analyze textual inputs and suggest design elements, allowing users to interact with and improve model performance, with a selection system that combines outputs from multiple models to provide complete content suggestions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual searching, capturing, and inserting of design elements is used, then users have full control over content selection, but the process becomes time-consuming and bandwidth-intensive

Engineering Contradiction:

Improveease of incorporating design elementsVSAvoidtime required for manual content insertion

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system automatically generates content suggestions by analyzing the user's document and autonomously inserting relevant design elements, images, or other content without requiring manual user actions for each insertion step

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical operations (searching, capturing, inserting) with an automated machine learning-based system that processes document content and generates suggestions through algorithmic analysis

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If manual searching, capturing, and inserting of design elements is used, then users have full control over content selection, but network bandwidth usage increases

Engineering Contradiction:

Improveease of incorporating design elementsVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

Ease of operationVSLoss of energy

Solution Approach 1:

The system performs automated content analysis and suggestion generation locally or through efficient server processing, reducing the need for continuous network operations and manual user interactions that consume bandwidth

Inventive Principle:

Principle #25Self-service

3Device complexity

If a single text-to-content model is used, then the system complexity is reduced, but the completeness of content suggestions decreases

Engineering Contradiction:

Improvenumber of text-to-content modelsVSAvoidcompleteness of content suggestions

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent combines outputs from multiple text-to-content models through a selection system that aggregates suggestions from different models to provide more complete and comprehensive content recommendations

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The selection system serves as a universal component that processes and integrates outputs from various specialized text-to-content models, enabling the system to handle diverse content types without requiring each model to be universally competent

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Productivity

If multiple text-to-content models are run and their outputs are mixed, then the completeness of content suggestions improves, but the system complexity increases

Engineering Contradiction:

Improvecompleteness of content suggestionsVSAvoidcomplexity of selection and mixing system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The selection system pre-establishes mixing constraints and rules for combining model outputs, organizing the complexity in advance rather than making complex decisions in real-time during content generation

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11270059B2Machine learning model-based content processing framework

Publication Date: 2022.03.08 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A textual user input is received and a plurality of different text-to-content models are run on the textual user input. A selection system attempts to identify a suggested content item, based upon the outputs of the text-to-content models. The selection system first attempts to generate a completed suggestion based on outputs from a single text-to-content model. It then attempts to mix the outputs of the text-to-content models to obtain a completed content suggestion.