Speech Control Device for Translation Pause Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In communication scenarios requiring translation via signage terminals, issues arise when speakers continuously speak without pausing, leading to information overload for listeners, and attempts to interrupt or signal a pause are unnatural, disrupting the conversation flow.

Innovation Solution

A control device and method that analyzes speech content to determine when to pause the speaker, ensuring smooth translation and natural communication by identifying speech patterns and listener engagement, using a system with cameras and microphones to detect mouth position and gaze to manage voice recognition and translation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the speaker continuously speaks without pausing, then the translation process can be maintained without interruption, but the listener experiences information overload and is unable to process the information effectively

Engineering Contradiction:
Improvetranslation process continuityVSAvoidinformation processing effectiveness
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system implements periodic pauses in the speaker's speech by detecting speech continuation beyond a threshold duration and inserting pause notifications. This transforms continuous speech into periodic segments that the listener can process, resolving the contradiction between maintaining translation continuity and preventing information overload.

Inventive Principle:
Principle #19Periodic action

2Reliability

If the listener forces the speaker to stop speaking to prevent information overload, then the listener can process information, but the communication flow is disrupted and becomes unnatural

Engineering Contradiction:
Improveinformation processing effectivenessVSAvoidcommunication smoothness
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system introduces an intermediary pause notification mechanism that mediates between the speaker and listener. Instead of the listener directly forcing the speaker to stop, the system automatically generates pause notifications based on speech analysis, maintaining natural communication flow while enabling effective information processing.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback by monitoring speech duration and automatically generating pause notifications when speech continues beyond a threshold. This closed-loop control prevents information overload without requiring direct listener intervention, thus maintaining communication smoothness while ensuring information processing effectiveness.

Inventive Principle:
Principle #23Feedback

3Reliability

If the system automatically inserts pause notifications to manage speech flow, then information overload is prevented, but the naturalness of conversation may be affected

Engineering Contradiction:
Improveinformation processing effectivenessVSAvoidconversation naturalness
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system dynamically adjusts pause insertion based on real-time speech analysis. Pauses are inserted only when speech continuation exceeds a threshold duration, making the system adaptive to actual speech patterns rather than applying fixed pauses. This maintains conversation naturalness while preventing information overload.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS11188288B2Control device and control method to issue notification on occurrence of processing overflow
Publication Date: 2021.11.30 SONY GROUP CORP
  • US11188288B2 patent drawing
  • US11188288B2 patent drawing
  • US11188288B2 patent drawing

AI summary

Provided is a control device including a control unit that executes a process of causing a speaker to stop speaking on a device toward which the speaker speaks on a basis of a result obtained by analyzing speech content based on a result of the speaker speaking toward the device.