Post-Conversation Speech Evaluation Using Comparative Metrics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The challenge of effectively evaluating and improving speech interactions in virtual communication, particularly in call centers, where agents struggle with monotony and repetitive tasks, leading to lower customer satisfaction and longer calls due to difficulty in grasping the emotional state of the other party.

Innovation Solution

A system and method for analyzing and evaluating speech interactions using an electronic device and server device, which determines conversation metrics and performance metrics to provide post-conversation feedback, enabling users to improve their speech performance and engagement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If call center agents work for many hours handling repetitive tasks, then productivity is maintained, but speech performance and customer satisfaction deteriorate

Engineering Contradiction:

Improvecall handling volumeVSAvoidspeech performance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system provides real-time feedback during conversations by analyzing audio signals and delivering performance metrics to agents through a user interface. This continuous feedback loop allows agents to monitor and adjust their speech performance during calls, maintaining high-quality interactions even during prolonged work periods.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis of conversation metrics and prepares performance evaluations before agents complete their calls. By pre-processing audio data and generating performance insights in advance, the system ensures that agents receive timely feedback without delaying call handling operations.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If agents focus on completing repetitive tasks efficiently, then productivity increases, but ability to grasp emotions and sentiments deteriorates

Engineering Contradiction:

Improvetask completion rateVSAvoidemotional understanding

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The system acts as an intermediary between agents and customers by analyzing audio signals for emotional content and delivering insights to agents. This intermediary layer enables agents to understand customer emotions without directly bearing the cognitive load of emotional analysis, allowing them to maintain both productivity and emotional awareness.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces the manual, mechanical process of emotional analysis with automated speech processing and analysis algorithms. By substituting human emotional interpretation efforts with computational analysis, the system preserves emotional understanding capabilities while maintaining high task completion rates.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If real-time feedback during conversation is provided, then speech performance improves, but system complexity increases

Engineering Contradiction:

Improvespeech performanceVSAvoidsystem architecture

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the speech processing functionality into distinct modules: audio signal acquisition, conversation metric determination, performance metric calculation, and feedback delivery. This segmentation allows each component to be optimized independently and simplifies the overall system architecture by creating modular, manageable units.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs self-service by automatically analyzing audio signals and generating performance metrics without requiring external intervention. The electronic device autonomously processes conversation data and delivers feedback, reducing the need for complex external processing systems.

Inventive Principle:

Principle #25Self-service

4Loss of information

If post-conversation evaluation is implemented, then agent performance visibility improves, but processing time increases

Engineering Contradiction:

Improveperformance feedbackVSAvoidevaluation processing time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The system performs preliminary processing of audio signals during the conversation itself, extracting and storing conversation metrics as the call progresses. By pre-processing data in real-time rather than analyzing the entire audio stream after the call ends, the system significantly reduces post-conversation evaluation time while maintaining comprehensive performance visibility.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12412584B2System with post-conversation evaluation, electronic device, and related methods

Publication Date: 2025.09.09 GN HEARING AS

AI summary

System, electronic device, and related methods, in particular a method of operating a system comprising an electronic device is disclosed, the method comprising obtaining one or more audio signals including a first audio signal of a first conversation; determining a first conversation period of the first conversation, the first conversation period having a first duration; determining first conversation metric data including a first conversation metric based on the first conversation period; determining a second conversation period of the first conversation different from the first conversation period, the second conversation period having a second duration; determining second conversation metric data including a second conversation metric based on the second conversation period; determining a first performance metric based on a change between the first conversation metric data and the second conversation metric data; outputting, via the interface of the electronic device, the first performance metric.