AI-Based Call Evaluation Control Automation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional call center evaluation technologies are time-consuming, prone to human error and bias, and can only evaluate a small fraction of the massive call volume, leading to inaccurate assessments of agent performance.

Innovation Solution

A system utilizing machine learning (ML) to automatically evaluate call interactions by parsing evaluation questions, identifying intended participant interactions in voice recordings, and setting operator controls based on ML models, incorporating feedback to enhance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional call evaluation technologies are used, then evaluators can assess calls using graphical user interfaces, but the evaluation process takes a significant amount of time and can only evaluate a small sample of calls

Engineering Contradiction:

Improveevaluation accuracyVSAvoidevaluation volume

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the mechanical manual evaluation process with an AI-based automated system. The machine learning model processes voice session recordings and evaluation questions to automatically determine interactions and set answer controls, eliminating the need for human evaluators to manually listen to and assess each call, thereby increasing evaluation volume while maintaining accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The evaluation system performs self-service by automatically processing evaluations without requiring human intervention. The machine learning model independently analyzes voice recordings, identifies interactions, and sets answer controls in the graphical user interface, allowing the system to evaluate calls autonomously at scale

Inventive Principle:

Principle #25Self-service

2Reliability

If human evaluators manually assess calls, then detailed quality scoring can be provided, but human error, bias, and fatigue lead to inaccurate reviews

Engineering Contradiction:

Improveevaluation consistencyVSAvoidquality score accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent replaces human evaluation with an AI-based system that processes voice session recordings and evaluation questions consistently without human intervention. The machine learning model applies uniform criteria to all calls, eliminating human error, bias, and variability in assessment, thereby improving both reliability and precision of quality scoring

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system incorporates feedback mechanisms where the machine learning model processes evaluation questions and voice recordings to generate answers that can be verified and refined. This feedback loop ensures consistent and accurate evaluation by continuously aligning the assessment with predefined performance standards

Inventive Principle:

Principle #23Feedback

3Ease of operation

If only a small fraction of calls are evaluated, then detailed manual analysis is possible, but the vast majority of calls remain unevaluated

Engineering Contradiction:

Improveevaluation feasibilityVSAvoidcall coverage

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent replaces manual evaluation with an automated machine learning system that can process the entire call volume efficiently. The AI model analyzes voice session recordings and sets answer controls automatically, making it feasible to evaluate all calls rather than just a small fraction, thereby dramatically increasing call coverage while maintaining operational feasibility

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250211676A1Artificial intelligence based setting of controls in a user interface

Publication Date: 2025.06.26 OPEN TEXT CORPORATION

AI summary

One embodiment comprises a system for artificial intelligence-based evaluation of an agent interaction. The system is operable to generate a transcript of the voice session recording of the call and parse the evaluation form to identify the evaluation question for an intended call participant interaction. The system is further operable to determine, using the machine learning model, that the intended call participant interaction of the evaluation question is present in the transcript. When the intended call participant interaction is present in the transcript, the system records a first answer for to the evaluation question corresponding to the voice session recording, the first answer indicating a presence of the intended call participant interaction in the voice session recording, and automatically sets the answer control in the visual display to indicate the presence of the intended call participant interaction in the voice session recording.