Audio Escalation Detection for Personnel Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Businesses and government agencies face challenges in efficiently reviewing and analyzing vast amounts of video footage to detect personnel performance issues and identify the need for de-escalation training, as manual review is burdensome and time-consuming.

Innovation Solution

A computer-implemented method and system that analyzes audio-video recordings for changes in tone and presence of negative tones to detect escalation events, generating escalation data for training and evaluation purposes, using sentiment analysis to classify interactions and identify personnel requiring additional training.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual review of video footage is used to detect personnel performance issues, then analysis accuracy can be maintained, but time consumption and burden increase significantly

Engineering Contradiction:
Improveanalysis accuracyVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical review process with an automated computer-based system that uses audio analysis, tone detection, and natural language processing to identify escalation events. This substitution maintains analysis accuracy by using sophisticated algorithms while dramatically reducing time consumption and human burden.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system introduces an intermediary automated analysis layer between the raw video footage and the final performance evaluation. This intermediary processes audio data, detects tone changes, and identifies escalation patterns, thereby maintaining precision while reducing the time required for direct human review.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If comprehensive video footage analysis is performed to identify all escalation events, then detection completeness improves, but processing complexity and resources increase

Engineering Contradiction:
Improvedetection completenessVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the analysis process into distinct modular components: audio data extraction, tone change detection, negative tone identification, escalation pattern recognition, and event classification. This segmentation maintains detection completeness by systematically analyzing all relevant aspects while reducing overall processing complexity through organized, stepwise analysis.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system focuses on detecting specific partial indicators of escalation (tone changes, negative tones, specific keywords) rather than attempting to analyze every aspect of the interaction. This partial action approach maintains reliability for identifying escalation events while reducing processing complexity by concentrating on key diagnostic features.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10573337B2Computer-based escalation detection
Publication Date: 2020.02.25 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10573337B2 patent drawing
  • US10573337B2 patent drawing
  • US10573337B2 patent drawing

AI summary

Systems and methods for computer-based escalation detection are disclosed. In embodiments, a method includes: determining an occurrence of an interaction event between a first party and a second party within a recording including audio data; analyzing the audio data; determining, based on the analyzing the audio data, an escalation during the interaction event to generate escalation data; saving the escalation data; partitioning each interaction event into a plurality of sections, wherein a first section represents a start of the interaction event, and another section represents an end of the interaction event; assigning a sentiment score for each of the plurality of sections; and calculating an overall sentiment score for the interaction event by combining the sentiment scores for each of the plurality of sections, wherein the saved escalation data includes the overall sentiment score.