Real-Time Speech Analytics for Contact Center Wait Time Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing contact center systems rely on historical data to estimate expected wait times, which are often inaccurate, leading to inefficient operations and dissatisfied customers due to unpredictable call durations.
Innovation Solution
Implementing a system that uses real-time speech analytics to monitor communication sessions, estimate progress, and project the duration of remaining calls, providing a more accurate expected wait time calculation by comparing current call progress to historical statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If historical data is used to calculate expected wait time, then the calculation is simple and fast, but the accuracy of the wait time estimate deteriorates
Solution Approach 1:
The system performs preliminary actions by monitoring and analyzing call progress indicators in real-time during the call. Speech analytics and other indicators are collected and analyzed throughout the call duration to determine the actual call outcome and duration, which then feeds back to improve future EWT calculations.
Solution Approach 2:
The system implements feedback mechanisms where actual call outcomes and durations from monitored calls are fed back into the EWT calculation system. This feedback loop allows the system to continuously improve the accuracy of expected wait time estimates by comparing predicted versus actual call durations and adjusting calculations accordingly.
2Measurement precision
If real-time speech analytics is implemented to improve wait time accuracy, then the accuracy of expected wait time estimate improves, but the device complexity and computational requirements worsen
Solution Approach 1:
The system achieves multi-functionality by using a single monitoring infrastructure that simultaneously performs multiple functions: speech analytics, call progress tracking, indicator collection, and outcome determination. This unified approach reduces overall system complexity compared to having separate specialized systems for each function.
Solution Approach 2:
The system employs self-service mechanisms where the monitoring system automatically analyzes call content and progress without requiring manual intervention. Speech analytics engines automatically process audio streams, identify key indicators, and determine call outcomes, reducing the need for human analysts and simplifying operational complexity.
3Measurement precision
If real-time monitoring of call content is performed, then the accuracy of call duration projection improves, but the loss of time for processing and analyzing call data worsens
Solution Approach 1:
The system performs preliminary analysis during the call by continuously monitoring speech patterns and progress indicators in real-time. Rather than analyzing the entire call after it ends, the system identifies key milestones and progress markers throughout the call, enabling early prediction of call duration and outcome without requiring complete call analysis.
Solution Approach 2:
The system skips unnecessary processing by focusing only on relevant speech indicators and call progress markers that actually predict call duration. Rather than analyzing every word and phrase, the system identifies and processes only the critical indicators that provide predictive value, reducing processing time while maintaining accuracy.
Data Source
AI summary
System and method to calculate expected waiting time of a caller to a calling center, the method including: monitoring, by a monitor circuit, a content of a present communication session; estimating a point of progress of the monitored communication session; comparing the point of progress to a historical statistic; calculating, by a processor, a projection of a duration of a remainder of the present communication session; and providing, by a communication circuit, an expected waiting time (EWT) based upon the projection of the duration of the remainder of the present communication session. Embodiments may include a speech search process to record call progress. The speech search process may inform, diagnose or monitor a call. The speech search process may inform a supervisor of progress, to take action if necessary. The speech search process may dynamically trigger other processes and construct profiles based upon historical data.