Action-Object Matrices for Seamless Speech App Transitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing call center systems face inefficiencies when transitioning between speech-enabled applications, as customers are often routed back to the beginning of a dialog, requiring repetitive information input, which is awkward and annoying.
Innovation Solution
The use of action-object matrices in both Automatic Call Routing (ACR) and self-service applications allows seamless integration by passing information from ACR to self-service applications, enabling a logical and coherent dialog continuation without the need for repetitive customer input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If customers are routed to self-service applications after ACR, then call center efficiency is improved, but customers are forced to repeat information already provided, increasing frustration and reducing ease of operation
Solution Approach 1:
The ACR application performs preliminary information gathering and analysis before routing the customer to the self-service application. The system pre-processes the customer's natural language input to extract relevant entities, intents, and context, storing this information in advance so that the self-service application can continue the dialogue without requiring the customer to repeat themselves.
Solution Approach 2:
The patent introduces an intermediary information transfer mechanism between ACR and self-service applications. This mediator component receives the customer's initial input, processes it through ACR, and then passes the extracted relevant information to the self-service application, enabling seamless continuation of the dialogue without customer repetition.
2Device complexity
If conventional menu systems are used for call routing, then system complexity is reduced, but customer interaction becomes less natural and more time-consuming
Solution Approach 1:
The patent replaces conventional mechanical menu-based routing systems with a speech recognition and natural language processing system. Instead of requiring customers to navigate through hierarchical menu options, the system uses speech-enabled technology to interpret customer intent directly, significantly reducing interaction time while maintaining manageable system complexity through structured action-object matrices.
3Adaptability or versatility
If speech recognition technology is implemented in multiple separate applications, then service capability is enhanced, but integration between applications becomes complex and information sharing difficult
Solution Approach 1:
The patent implements a universal action-object matrix framework that can be applied across multiple speech-enabled applications. This universal structure defines standardized actions (verbs) and objects (nouns) that create a common language for information exchange between ACR and self-service applications, enabling seamless integration while maintaining the versatility of each individual application.
Data Source
AI summary
A system, method and software are described for facilitating transitions between speech-enabled applications employing action-object matrices. In an exemplary embodiment, an automatic call router determines a service agent destination by identifying an action-object combination included in a user utterance. In addition to forwarding the user to the service agent destination, the automatic call router sends one or more aspects of the user utterance to the service agent for use therewith. The service agent may then use aspects of the forwarded information to request further information from the user or to begin performance of a user desired transaction.