Streaming Voice Signal Analysis for Contact Center Storage Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face challenges in efficiently recording and analyzing voice signals due to high storage costs and delayed issue identification, missing opportunities to address problems in real-time, leading to potential customer dissatisfaction and loyalty loss.
Innovation Solution
A method and apparatus for processing streaming voice signals to detect predetermined utterances, determining response-determinative significance, and initiating responsive actions, while deciding on short-term or long-term storage, thereby optimizing storage and reaction times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice signals are recorded for compliance and quality purposes, then compliance and quality review are improved, but storage costs increase and analysis is delayed
Solution Approach 1:
The patent extracts only the relevant portions of voice signals (segments containing predetermined utterances) for storage and analysis, rather than recording entire calls. This is achieved by continuously monitoring streaming voice signals and extracting only those segments that contain keywords or phrases of interest, thereby reducing storage requirements while maintaining compliance and quality review capabilities
Solution Approach 2:
The system performs preliminary analysis of voice signals in real-time by continuously monitoring for predetermined utterances before full recording is initiated. This preliminary detection action allows the system to identify and flag only the relevant segments for storage, avoiding the need to store entire calls and reducing storage costs while ensuring compliance
2Reliability
If voice signals are recorded for analysis, then quality improvement is achieved, but real-time response capability is lost
Solution Approach 1:
The system performs preliminary detection of predetermined utterances in real-time during the voice signal stream, enabling immediate identification of quality issues as they occur. This preliminary action allows the system to trigger real-time responses (such as agent alerts or supervisor notifications) while the call is still active, eliminating the delay inherent in post-call analysis
Solution Approach 2:
The patent implements a feedback mechanism where the real-time detection of predetermined utterances immediately triggers responses such as agent alerts, supervisor notifications, or automated interventions. This feedback loop enables quality issues to be addressed during the call rather than after it concludes, maintaining real-time response capability while improving quality
3Reliability
If all calls are recorded, then compliance requirements are met, but storage costs and analysis difficulty increase
Solution Approach 1:
The system extracts only the relevant segments containing predetermined utterances from the complete voice signal stream for storage and analysis. By using keyword detection and phrase recognition, the system identifies and extracts only those portions of calls that require compliance review, thereby meeting compliance requirements while significantly reducing storage costs and analysis complexity
Solution Approach 2:
Instead of recording all calls (excessive action), the system records only the necessary portions containing predetermined utterances (partial action). This selective recording approach ensures compliance by capturing all potentially problematic segments while avoiding the excessive storage and analysis costs associated with recording every call
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
Streaming voice signals, such as might be received at a contact center or similar operation, are analyzed to detect the occurrence of one or more unprompted, predetermined utterances. The predetermined utterances preferably constitute a vocabulary of words and/or phrases having particular meaning within the context in which they are uttered. Detection of one or more of the predetermined utterances during a call causes a determination of response-determinative significance of the detected utterance(s). Based on the response-determinative significance of the detected utterance(s), a responsive action may be further determined. Additionally, long term storage of the call corresponding to the detected utterance may also be initiated. Conversely, calls in which no predetermined utterances are detected may be deleted from short term storage. In this manner, the present invention simplifies the storage requirements for contact centers and provides the opportunity to improve caller experiences by providing shorter reaction times to potentially problematic situations.