Voice Communication Analysis System for Call Authenticity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IVR systems face inefficiencies in distinguishing genuine from non-genuine incoming voice calls, leading to increased workload and reduced productivity in call centers and public service agencies.
Innovation Solution
A machine learning-based system that trains models to analyze vocal attributes of incoming calls, using a combination of frequency analysis and rate of change detection to determine call authenticity, allowing for filtering or routing of genuine calls to appropriate recipients and non-genuine calls to remedial actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If IVR systems process all incoming voice calls, then call handling coverage is maintained, but workload increases and productivity decreases due to non-genuine calls
Solution Approach 1:
The system performs preliminary analysis of voice calls using machine learning models to detect genuineness before routing to call center agents. This preliminary action filters out non-genuine calls (pranks, errors, malicious calls) before they reach the call center, allowing agents to focus only on genuine calls and improving overall productivity while maintaining coverage for legitimate inquiries
Solution Approach 2:
The machine learning-based genuineness detection system acts as an intermediary between the incoming call and the call center agent. This intermediary layer analyzes vocal attributes, frequency, and patterns to determine call authenticity, then routes only genuine calls to agents, reducing workload while maintaining service quality for legitimate calls
2Measurement precision
If manual analysis of each call is performed, then call authenticity can be determined, but time consumption increases and productivity decreases
Solution Approach 1:
The system replaces manual human analysis with an automated machine learning-based detection system. The machine learning models process vocal attributes, frequency patterns, and call characteristics automatically, providing accurate genuineness determination in seconds rather than requiring time-consuming manual review by agents
Solution Approach 2:
The system transforms the analysis approach by changing from manual qualitative assessment to automated quantitative parameter analysis. Machine learning models extract and analyze specific parameters such as frequency, vocal patterns, and temporal characteristics, enabling rapid and consistent authenticity determination across all calls
Data Source
AI summary
Techniques are disclosed for applying a trained machine learning model to incoming voice communications to determine whether the voice communications are genuine or not genuine. The trained machine learning model may identify vocal attributes within the target call and use the identified attributes, and the training, determine whether the target call is genuine or not genuine. An applied trained machine learning model may include multiple different types of trained machine learning models, where each of different types of machine learning models is trained and/or configured for a different function within the analysis.