Voice-Guided Screen Summarization and Text Editing for Accessibility

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Visually impaired individuals face significant barriers in efficiently acquiring information from electronic device screens and performing text input or editing operations due to the limitations of conventional voice-based systems in providing concise summaries, flexible text editing, and comprehensive email creation through intuitive voice commands, leading to reduced independence and productivity.

Innovation Solution

A system comprising a processor that acquires and analyzes voice commands, obtains relevant screen information, and utilizes a generative artificial intelligence model to summarize or edit content, delivering the results through synthesized speech, enabling efficient text-based task management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional voice-based systems are used for information access and text operations, then visual impairment accessibility is improved, but information acquisition efficiency and task management capability deteriorate

Engineering Contradiction:

Improveaccessibility for visually impaired usersVSAvoidinformation acquisition efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent replaces conventional mechanical voice command recognition with an AI-based natural language processing system. The processor uses machine learning models to understand and interpret user intent from spoken language, enabling more efficient and accurate information retrieval and text manipulation operations compared to traditional keyword-based voice systems.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically adjusts information presentation parameters based on user needs. The processor analyzes context and user preferences to modify output format, detail level, and presentation mode, optimizing both accessibility and efficiency for visually impaired users performing various tasks.

Inventive Principle:

Principle #35Parameter changes

2Loss of information

If detailed screen information is provided to visually impaired users, then information completeness is improved, but information processing time and cognitive load increase

Engineering Contradiction:

Improveinformation completenessVSAvoidinformation processing time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The patent segments screen information into hierarchical levels of detail. The processor divides complex screen content into structured components (headings, paragraphs, key elements) and allows users to access information at different granularities, presenting only relevant portions based on current task context to reduce processing time while maintaining completeness when needed.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements partial information delivery by providing summaries and key points first, then offering detailed information on demand. This approach delivers sufficient information for most tasks without overwhelming users with complete detail, reducing processing time while maintaining the ability to access full information when required.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If comprehensive text editing capabilities are provided, then task functionality is improved, but system complexity increases

Engineering Contradiction:

Improvetext editing capabilityVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal voice-based text editing system where a single AI-powered processor handles multiple editing operations (selection, copying, pasting, deletion, formatting) through natural language commands. This multi-functional approach provides comprehensive text editing capabilities without requiring separate specialized systems for each operation, managing complexity through unified AI processing.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Ease of operation

If voice commands are made more intuitive and natural, then ease of use is improved, but recognition accuracy and processing reliability may deteriorate

Engineering Contradiction:

Improvevoice command intuitivenessVSAvoidcommand recognition accuracy

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent implements feedback mechanisms where the system confirms interpreted commands before execution and provides clarification when recognition confidence is low. The processor analyzes speech patterns and context, then validates understanding by presenting interpreted commands to users for confirmation, maintaining high reliability while accepting natural, varied speech inputs.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary speech analysis and context preparation before final command execution. The processor pre-processes spoken input by analyzing acoustic features, comparing against known command patterns, and preparing multiple interpretation options, then selects the most accurate interpretation to maintain reliability while accepting intuitive natural speech.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260065910A1System

Publication Date: 2026.03.05 SOFTBANK GROUP CORP

AI summary

A system includes a processor that is configured to acquire a voice command from a user, analyze the acquired voice command, acquire screen information based on the analyzed voice command, process the acquire screen information using a generative artificial intelligence model to summarize or edit the information, and provide the summarized or edited information to the user as synthesized speech.