Labeling Support Device for Contact Center Voice Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for improving service quality in contact centers through voice recognition require a large amount of training data and manual label assignment, which is time-consuming and burdensome for workers, especially when labels have a hierarchical structure without guidelines.

Innovation Solution

A label assignment support device and method that uses an existing model to estimate preliminary labels, generates a label assignment work screen for updating labels, and arranges elements and labels in a structured format to facilitate efficient label assignment by workers, reducing the need for extensive manual updating.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual label assignment is performed for creating training data, then labeling accuracy can be ensured, but the time and effort required for label assignment increases significantly

Engineering Contradiction:
Improvelabeling accuracyVSAvoidtime for label assignment
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary label assignment using an existing model before manual review. This preliminary labeling step provides a starting point that is already reasonably accurate, reducing the amount of manual work needed while maintaining high final labeling accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The existing model acts as an intermediary between complete manual labeling and no labeling. It provides preliminary labels that serve as a foundation, which workers then review and correct, rather than creating labels from scratch or relying solely on automated labels.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If workers assign labels without guidelines for hierarchical label structures, then flexibility is maintained, but the burden and complexity of label assignment increases

Engineering Contradiction:
Improveflexibility in label assignmentVSAvoidease of label assignment
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system displays the hierarchical label structure and relationships to workers during the label assignment process. This feedback mechanism helps workers understand how labels connect and constrain each other, making the process more intuitive and easier to operate while maintaining flexibility.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The label assignment process is segmented into different levels of the hierarchy. Workers can work with high-level labels first, then progressively refine to more specific subordinate labels, breaking down the complex task into manageable segments that are easier to handle.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If a large amount of training data is collected to improve model accuracy, then estimation accuracy improves, but the work required to create and process the training data increases

Engineering Contradiction:
Improveestimation accuracyVSAvoidwork efficiency in creating training data
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary label assignment using an existing model before manual review. This preliminary labeling step provides a starting point that is already reasonably accurate, reducing the amount of manual work needed while maintaining high final labeling accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The existing model serves itself by generating preliminary labels that can be reviewed and corrected by workers. This self-service capability reduces the overall workload compared to complete manual labeling, improving productivity while still achieving high accuracy through the combination of automated preliminary work and human review.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20240303265A1Labeling support device, labeling support method, and program
Publication Date: 2024.09.12 NT T INC
  • US20240303265A1 patent drawing
  • US20240303265A1 patent drawing
  • US20240303265A1 patent drawing

AI summary

A label assignment support device according to the present disclosure includes a preliminary label estimation unit that assigns preliminary labels for each of a plurality of elements, a label assignment work screen output unit that generates a label assignment work screen for each of the plurality of elements and an update operation for labels assigned to the plurality of elements by a user, the label assignment work screen indicating each of the plurality of elements and labels assigned to each of the plurality of elements in association with each other, and a label update unit that, when a label assigned to one of the elements is updated by the update operation via the label assignment work screen, assigns the label after update to the one of the elements.