Conversation Text Display Using Partial and Integrated Speech Recognition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional conversation support systems fail to display conversation content in real time for hearing-impaired individuals due to the need for collective voice recognition, leading to difficulties in following the conversation progress.

Innovation Solution

A conversation support system that performs real-time partial section voice recognition and integrates it with delayed utterance section recognition, displaying partial section text information sequentially and utterance section text information with distinct modes to ensure reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If collective voice recognition is executed for the entire utterance content, then voice recognition accuracy is improved, but real-time display capability deteriorates

Engineering Contradiction:

Improvevoice recognition accuracyVSAvoidreal-time display delay

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the utterance into multiple sections (first section, second section, etc.) and performs voice recognition on each section separately. The first voice recognition unit processes the first section in real-time while the second voice recognition unit processes the second section, allowing partial results to be displayed immediately without waiting for the complete utterance. This segmentation resolves the contradiction by enabling both real-time display of processed sections and accurate recognition through collective processing of all sections.

Inventive Principle:

Principle #1Segmentation

2Speed

If text is displayed sequentially in real-time, then conversation progress visibility is improved, but voice recognition accuracy deteriorates

Engineering Contradiction:

Improvetext display speedVSAvoidvoice recognition accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The patent segments the utterance into multiple sections and processes them through different voice recognition units. The first voice recognition unit provides quick results for real-time display, while the second voice recognition unit performs more thorough processing for accuracy. This allows the system to display text sequentially as sections are processed while maintaining overall recognition accuracy through the combined results of all recognition units.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different sections of the utterance are processed with different qualities and levels of processing. The first section receives rapid processing for immediate display, while subsequent sections receive more comprehensive processing. This local quality differentiation allows real-time display capability while ensuring that each section is processed with sufficient accuracy for its specific position in the sequence.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12555582B2Conversation support device, conversation support system, conversation support method, and storage medium

Publication Date: 2026.02.17 HONDA MOTOR CO LTD

AI summary

In a conversation support device, a first voice recognition unit performs voice recognition processing on the basis of a voice signal and defines partial section text information for each partial section that is a part of an utterance section, a second voice recognition unit performs voice recognition processing on the basis of the voice signal and defines utterance section text information for each utterance section, an information integration unit integrates the partial section text information into the utterance section text information to generate integration text information, and an output processing unit outputs the integration text information to the display unit after outputting the partial section text information to the display unit.