Voice Menu Tree Failure Escalation to Human Advisors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice recognition systems in menu trees often fail to accurately translate user inputs, leading to customer frustration and increased call termination rates in call centers, as users may become disheartened when navigating through menu trees and encountering difficulties.
Innovation Solution
A method and system that receive voice requests via a telematics unit and wireless network, determine a voice menu tree, assess responses for failures or confirmations, convert responses to data, and record failures for provision to advisors at call centers, enabling improved navigation and customer service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If speech recognition software is used to automate customer service calls, then call center throughput increases, but accuracy of translating spoken input deteriorates
Solution Approach 1:
The patent introduces an intermediary human advisor who assists customers when the speech recognition engine fails to properly translate spoken input. The system detects recognition failures and routes them to human advisors, thereby maintaining high automated throughput while ensuring accurate translation through human intervention when needed.
Solution Approach 2:
The system implements feedback mechanisms where the speech recognition engine's performance is monitored and evaluated. When translation accuracy falls below acceptable thresholds, the system triggers escalation to human advisors, creating a feedback loop that maintains overall system accuracy while preserving automated efficiency.
2Reliability
If customers encounter difficulty navigating voice menu trees, then service accuracy improves with human assistance, but call termination rates increase due to customer frustration
Solution Approach 1:
The system performs preliminary detection of recognition failures and proactively escalates to human advisors before customers become frustrated and terminate calls. By anticipating service accuracy issues and addressing them preemptively, the system prevents call terminations while ensuring reliable service delivery.
Solution Approach 2:
The system prepares for potential recognition failures by having human advisors readily available to handle escalated cases. This cushioning mechanism ensures that when translation accuracy deteriorates, customers are smoothly transitioned to human assistance rather than terminating their calls in frustration.
3Device complexity
If customers are forced to re-navigate menu topics after speech recognition failure, then system complexity remains manageable, but customer satisfaction deteriorates
Solution Approach 1:
When speech recognition fails within the menu tree structure, a human advisor acts as an intermediary to assist customers through the menu navigation process. This prevents customers from having to re-navigate topics independently, maintaining manageable system complexity while significantly improving ease of operation and customer satisfaction.
Data Source
AI summary
A method for providing menu tree assistance includes receiving a voice request from a user via a telematics unit and a wireless network and determining a voice menu tree based on the voice request. The method further includes receiving at least one response based on the voice menu tree, determining a failure or confirmation based on the at least one response and converting the response to data based on the determined confirmation. The method further includes recording the response based on the determined failure and providing the recorded response and data to an advisor at a call center. A system and a computer readable medium including computer program code are also disclosed.