Speech Call Routing via Action-Object Pair Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech recognition-enabled automatic call routing systems often misinterpret caller intent, leading to inefficient routing and significant costs due to abandoned calls, as they struggle to accurately convert spoken language into actionable instructions.
Innovation Solution
A call routing system that converts speech input into text, determines an object and action pair with confidence levels, and routes calls based on an action-object routing table, ensuring compliance with business rules and routing callers to appropriate destinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If speech recognition systems use natural language processing to enable callers to speak freely, then ease of operation is improved, but measurement precision of caller intent deteriorates
Solution Approach 1:
The patent introduces an intermediary processing layer between the acoustic model and semantic model that extracts action-object pairs from recognized speech. This intermediary structure translates free-form natural language into structured representations, enabling precise routing decisions while maintaining natural language input benefits.
Solution Approach 2:
The speech recognition system is divided into distinct modules: acoustic model for phonetic recognition, semantic model for meaning extraction, and action-object pair extraction for routing decisions. This segmentation allows each component to specialize in specific tasks, improving overall precision while maintaining ease of use.
2Productivity
If the system routes calls automatically based on speech recognition, then productivity is improved, but reliability of routing decisions deteriorates
Solution Approach 1:
The system incorporates feedback mechanisms where routing decisions are continuously refined based on extracted action-object pairs and confidence levels. The semantic model provides feedback to the routing module, allowing the system to adapt and improve routing reliability while maintaining high productivity through automated processing.
Solution Approach 2:
The patent changes the parameter representation from raw speech text to structured action-object pairs with associated confidence levels. This parameter transformation enables more reliable routing decisions by providing explicit structure and confidence metrics that the routing logic can directly utilize.
3Measurement precision
If the system uses complex speech recognition processing, then measurement precision of caller intent is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential action-object pairs from complete speech recognition processing. Rather than analyzing entire speech contexts, the system extracts key actionable elements needed for routing, reducing computational complexity while maintaining intent recognition accuracy.
Solution Approach 2:
The acoustic and semantic models perform preliminary processing to convert speech into structured action-object pairs before routing decisions are made. This preliminary structuring simplifies subsequent routing logic, reducing overall system complexity while preserving precision.
Data Source
AI summary
A method of processing a call is disclosed. The method can transform speech input from a caller of a call into text and convert the text into an object and an action. The method determines a call destination based on the object and the action. The method can route the call to a destination when a caller is not in compliance with at least one business rule. The method can further route the call to the call destination when the caller is in compliance.