Speech-Based Document Retrieval for Contact Center Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Contact center agents face challenges in efficiently retrieving relevant documents from a knowledge base to resolve customer inquiries, leading to increased wait times and customer frustration.

Innovation Solution

A system and method for automatic speech-based interaction resolution, where customer speech is used to query a knowledge base automatically, enabling faster retrieval and delivery of relevant documents to customers or agents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If agents manually search the knowledge base to resolve customer inquiries, then they can retrieve relevant documents, but the process consumes excessive time and increases customer wait times

Engineering Contradiction:
Improvetime to retrieve documentsVSAvoidinquiry resolution efficiency
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The system enables automatic speech-based document retrieval where the system serves itself by converting speech to text, generating search queries, and retrieving documents without human intervention. This self-service mechanism eliminates manual searching time while maintaining high productivity in inquiry resolution.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual searching process with an automated speech-based system. Speech-to-text conversion, automatic query generation, and document retrieval are performed through computational processes rather than manual agent actions, significantly reducing time loss while preserving or enhancing productivity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If the knowledge base contains comprehensive organization-specific information, then agents can find relevant documents, but the volume of information makes manual searching inefficient

Engineering Contradiction:
Improvecompleteness of knowledge baseVSAvoidcomplexity of information retrieval
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system incorporates feedback mechanisms where speech queries are converted to text, processed through search algorithms, and results are retrieved automatically. This feedback loop enables the system to handle comprehensive knowledge base information efficiently by automatically processing queries against the complete information set without increasing operational complexity for agents.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The speech-to-text conversion system and automatic query generation act as intermediaries between the comprehensive knowledge base and the retrieval process. This intermediary layer manages the complexity of searching through vast information volumes by automating the query formulation and document retrieval, maintaining information completeness while reducing retrieval complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If agents spend more time researching information to resolve inquiries, then customer satisfaction may improve through thorough answers, but contact center load increases

Engineering Contradiction:
Improvequality of customer serviceVSAvoidcontact center throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by automatically converting speech to text, generating search queries, and retrieving relevant documents before agents need to review them. This preliminary automated processing ensures thorough information retrieval maintains service quality while reducing the time agents spend on research, thereby increasing contact center throughput and productivity.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3577582B1System and method for speech-based interaction resolution
Publication Date: 2025.03.05 GENESYS CLOUD SERVICES INC
  • EP3577582B1 patent drawingFigure 1
  • EP3577582B1 patent drawingFigure 2
  • EP3577582B1 patent drawingFigure 3

AI summary

A method for automatically retrieving documents based on customer speech received at a contact center of an organization includes: receiving, by a processor, at the contact center of the organization, speech from a customer; performing, by the processor, automatic speech recognition on the received speech to generate recognized text; generating, by the processor, a search query from the recognized text; searching, by the processor, a knowledge base specific to the organization for one or more documents relevant to the search query; and returning, by the processor, the one or more documents relevant to the search query.