Dialogue History Editing Interface for Call Center Transcript Summarization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Creating a dialogue history in call centers is time-consuming due to the large amount of textualized utterances from speech recognition, and automatic generation may not produce a sufficiently accurate summary, especially when errors occur in extracting regard and regard confirmation utterances.

Innovation Solution

A display device and editing support device that display and allow modification of dialogue scenes, utterance types, and focus point information, enabling efficient creation of dialogue histories by predicting dialogue scenes, utterance types, and extracting focus points, with an interface for adding, deleting, or modifying these elements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If service persons manually review all textualized utterances to create dialogue history, then the accuracy of dialogue history is improved, but the time required for creation increases significantly

Engineering Contradiction:

Improveaccuracy of dialogue historyVSAvoidtime required for creation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system extracts only the necessary components (regard utterances and regard confirmation utterances) from the complete dialogue transcript, presenting a filtered subset to service persons for review. This extraction approach maintains accuracy by focusing on critical elements while reducing the overall volume of content requiring manual verification, thereby addressing the time-consuming nature of manual review.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary automatic extraction and organization of regard utterances and regard confirmation utterances before presenting them to service persons. This preliminary processing reduces the manual workload by pre-identifying relevant segments, allowing service persons to focus their review efforts on verifying and refining the automatically generated content rather than reviewing the entire dialogue from scratch.

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If automatic generation is used to create dialogue history from regard utterances, then the time required is reduced, but the accuracy and appropriateness of the summary deteriorates

Engineering Contradiction:

Improvetime required for creationVSAvoidaccuracy of dialogue history

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system implements a feedback mechanism where service persons can review, correct, and refine the automatically generated dialogue history. The interface allows service persons to provide feedback on the accuracy of extracted regard utterances and their corresponding confirmations, enabling the system to learn from and improve upon its automatic generation based on human verification and correction input.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system segments the dialogue into distinct regard utterances and regard confirmation utterances, presenting them as separate identifiable units to service persons. This segmentation allows for more precise manual verification and correction of specific elements rather than requiring review of the entire dialogue, thereby maintaining accuracy while reducing the overall time investment required.

Inventive Principle:

Principle #1Segmentation

3Quantity of substance

If the system presents only extracted regard utterances and regard confirmation utterances, then the amount of information to review is reduced, but the completeness of the summary may be insufficient when extraction errors occur

Engineering Contradiction:

Improveamount of information to reviewVSAvoidcompleteness of summary

Core Design Contradiction:

Quantity of substanceVSLoss of information

Solution Approach 1:

The system provides a dynamic interface where service persons can interactively adjust the level of detail displayed. Based on feedback and correction actions, the system can dynamically modify the presentation of dialogue content, expanding or contracting the scope of displayed utterances to ensure completeness while maintaining efficient review processes. This dynamic adaptability allows the system to respond to extraction errors by adjusting the information presentation rather than being fixed in its approach.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11749258B2Device and method for supporting creation of reception history, non-transitory computer readable recording medium

Publication Date: 2023.09.05 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

A display device for displaying an utterance and information extracted from the utterance, includes an input/output interface configured to display display blocks for a series of dialogue scenes in chronological order of acquisition of utterances, each of the series of dialogue scenes being indicated by dialogue scene data stored in correspondence with utterance data indicating a corresponding one of the utterances, display dialogue scene information within each of the display blocks for the series of dialogue scenes, the dialogue scene information including the corresponding one of the utterances, an utterance type indicating a type of the corresponding one of the utterances, or utterance focus point information of the corresponding one of the utterances, and switch, based on an operation input, between displaying the dialogue scene information and not displaying the dialogue scene information within each of the display blocks for the series of dialogue scenes.