Communication Analysis Section Boundary Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication analysis systems struggle to accurately analyze the state of communication, particularly in group settings, as they often divide utterances at inappropriate times, leading to incomplete data and inadequate analysis.

Innovation Solution

An information processing system that includes an utterance judgement unit to determine utterance sections and an analysis section determination unit to set boundaries for actual analysis sections based on non-utterance times, ensuring that utterances are not divided mid-sentence and maintaining consistent section lengths.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If communication is divided into analysis sections at fixed time intervals, then analysis can be performed systematically, but utterances may be divided mid-sentence leading to inaccurate analysis

Engineering Contradiction:
Improveanalysis efficiencyVSAvoidutterance length accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary identification of utterance sections and their boundaries before dividing the communication into analysis sections. By knowing where utterances start and end in advance, the system can adjust analysis section boundaries to coincide with non-utterance periods, preventing mid-utterance divisions and ensuring accurate utterance length measurement while maintaining systematic analysis.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If analysis section boundaries are adjusted to avoid utterance division, then utterance accuracy improves, but section length becomes variable

Engineering Contradiction:
Improveutterance length accuracyVSAvoidanalysis section length consistency
Core Design Contradiction:
Measurement precisionVSStability of the object's composition

Solution Approach 1:

The system applies different boundary positioning strategies to different analysis sections based on local conditions. When an utterance is detected, the boundary is positioned at the nearest non-utterance time point, making each section's boundary location adaptive to local speech patterns rather than uniformly distributed, thus resolving the conflict between accuracy and consistency.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If utterance sections are identified to prevent division, then analysis accuracy improves, but processing complexity increases

Engineering Contradiction:
Improvecommunication state analysis accuracyVSAvoidsection boundary determination complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system introduces an intermediary component that identifies utterance sections and their time boundaries as a separate preprocessing step. This intermediary utterance section identification unit provides structured information about when utterances occur, which the analysis section determination unit then uses to adjust boundaries. This separation simplifies the overall complexity by dividing the task into manageable stages.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12094449B2Information processing system, information processing method, and program
Publication Date: 2024.09.17 TOYOTA JIDOSHA KK
  • US12094449B2 patent drawing
  • US12094449B2 patent drawing
  • US12094449B2 patent drawing

AI summary

An information processing system which enables more appropriate analysis on a state of communication participated by a plurality of participants is provided. The information processing apparatus includes an utterance judgement unit and an analysis section determination section. The utterance judgement unit judges an utterance section which is a time-series section of an utterance of each of a plurality of participants in the communication. The analysis section determination unit uses a reference analysis section to set boundaries of a plurality of actual analysis sections so that time points of boundaries of the plurality of actual analysis sections become times corresponding to the non-utterance times.