Contact Center Voice Data Analysis Using Psychological Behavioral Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing customer interactions in call centers lack objectivity and provide limited insights into the underlying behavioral characteristics of customers and agents, making it difficult to train agents effectively and improve customer relationships.
Innovation Solution
A method and system that separates telephonic communications into constituent voice data, applies a linguistic-based psychological behavioral model to analyze the data, and generates behavioral assessment data to provide objective insights for training and improving customer interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If post-call surveys are used to collect customer feedback, then customer perception data can be obtained, but the data collection depends on customer willingness and may not be objective
Solution Approach 1:
The patent replaces the manual/mechanical survey method with an automated acoustic analysis system. The system uses acoustic signal processing to automatically detect and analyze customer distress signals during the call, eliminating the need for customer participation in post-call surveys and providing objective, real-time data without relying on customer willingness to respond.
2Productivity
If stress analysis is performed on audio calls to determine satisfaction, then automated analysis is achieved, but the system provides limited insight into underlying behavioral characteristics
Solution Approach 1:
The patent segments the acoustic signal into multiple distinct distress signal types (anger, frustration, disappointment, confusion, etc.) and analyzes each separately. This segmentation allows the system to maintain automated efficiency while providing detailed insights into specific behavioral characteristics and emotional states, going beyond simple satisfaction/unsatisfaction binary classification.
Solution Approach 2:
The patent adds a new dimension of analysis by identifying and categorizing multiple types of distress signals simultaneously, rather than providing a single overall satisfaction metric. This multi-dimensional approach enriches the behavioral insights while maintaining automated processing efficiency.
3Loss of information
If multiple types of distress signals are identified and categorized, then detailed behavioral insights are provided, but the system complexity increases
Solution Approach 1:
The patent uses template matching and pattern recognition approaches where pre-defined acoustic patterns for different distress signal types are stored and compared against incoming signals. This copying approach allows detailed multi-type classification without requiring complex real-time analysis, as the system matches against established templates rather than generating classifications from scratch.
4Reliability
If acoustic distress signals are analyzed in real-time during calls, then objective behavioral data is captured, but the processing requirements and computational load increase
Solution Approach 1:
The patent performs preliminary processing of acoustic signals during the call by detecting and flagging distress signal occurrences in real-time. The detailed analysis and categorization can be performed subsequently on flagged segments rather than processing the entire call continuously, reducing real-time computational energy requirements while maintaining objective data capture.
Data Source
AI summary
A method for analyzing a telephonic communication between a customer and a contact center is provided. According to the method, a telephonic communication is separated into at least first constituent voice data and second constituent voice data. One of the first and second constituent voice data is analyzed by mining the voice data and applying a predetermined linguistic-based psychological behavioral model to one of the separated first and second constituent voice data. Behavioral assessment data is generated which corresponds to the analyzed voice data.