Conversational Writing Assistant for Voice-Guided Outline Generation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing writing tools are inadequate for untrained writers who struggle to structure their ideas and convey them effectively, especially those who are not familiar with computers, as they require sequential typing and structured organization.

Innovation Solution

A writing assistant system that engages in a turn-based verbal conversation to collect and structure ideas, generating an outline or summary without requiring users to type or write, using AI and deep learning to guide the conversation and ensure relevance and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If traditional writing tools are used, then structured organization can be achieved, but the operation becomes difficult for untrained writers

Engineering Contradiction:
Improveease of useVSAvoidcomplexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical typing system with a voice-based conversational system. Instead of requiring users to physically type on keyboards or touchpads, the system captures spoken words and converts them into structured text automatically, eliminating the need for manual input operations while maintaining structured output organization.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an AI language model as an intermediary between the user's spoken ideas and the final structured document. The model acts as a mediator that takes unstructured speech input, processes it through conversational turns, and outputs properly organized text, thereby shielding users from the complexity of document structure while preserving ease of use.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If sequential writing process is used, then structured organization is maintained, but productivity decreases for untrained writers

Engineering Contradiction:
Improvewriting speedVSAvoidtime to structure ideas
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent performs preliminary structuring actions automatically through the AI model's conversational processing. Instead of requiring users to spend time planning and organizing their ideas sequentially, the system proactively guides the conversation to extract and organize key information in real-time, completing the structuring work before the user even finishes speaking.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent enables continuous information capture throughout the conversation process. Rather than requiring users to pause and organize their thoughts between writing segments, the system continuously processes spoken input and maintains structured organization throughout the entire interaction, eliminating idle time and keeping the useful action of idea capture uninterrupted.

Inventive Principle:
Principle #20Continuity of useful action

3Ease of operation

If keyboard input is required, then text accuracy can be achieved, but ease of operation deteriorates for non-computer users

Engineering Contradiction:
ImproveaccessibilityVSAvoididea transfer accuracy
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent substitutes the keyboard input mechanism with voice capture technology. Users speak their ideas naturally without needing to learn keyboard typing or navigate computer interfaces, dramatically improving accessibility for non-computer users while speech-to-text conversion maintains accurate capture of their intended messages.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4031962B1Freeform conversation writing assistant
Publication Date: 2026.03.04 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP4031962B1 patent drawingFigure 1
  • EP4031962B1 patent drawingFigure 2A~2C
  • EP4031962B1 patent drawingFigure 3A~3C

AI summary

A device performs functions of: receiving a user input to initiate a conversation session for generating an outline for a writing; generating a voice output asking a question regarding the writing; receiving a voice input from the user responding to the voice output; identifying, based on the received voice input, content of the voice input responding to the voice output; repeating, until a predetermined condition is met, the steps of generating the voice output, receiving the voice input, and identifying the content of the voice input, wherein the question asked via each voice output is generated in response to the content of the voice input responding to the preceding voice output; and generating, based on the content of the voice inputs, the outline for the writing.