Speech Recognition Engine for Call Center Activity Tracking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Call centers face challenges in tracking and predicting call center activity and operator training due to the dynamic nature of customer issues, leading to inefficient call allocation and frustrating customer experiences.

Innovation Solution

Implementing an automated speech recognition engine to monitor voice calls, detect speech patterns associated with topics, and store records in a database, generating reports and real-time assistance messages for agents to improve call handling and training.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automated speech recognition engine is implemented to monitor and detect speech patterns, then measurement precision of call center activity is improved, but device complexity increases

Engineering Contradiction:

Improvecall center activity analysisVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

An automated speech recognition engine is introduced as an intermediary component between the call center communication system and the analysis system. This engine monitors voice calls, detects speech patterns, and converts spoken language into structured data that can be stored in a database and analyzed for trends, thereby enabling precise measurement of call center activity without requiring direct complex integration between all system components.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If speech recognition engine is used to detect speech patterns in real-time, then productivity of call handling is improved, but use of energy increases

Engineering Contradiction:

Improvecall handling efficiencyVSAvoidenergy consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The speech recognition engine processes only the essential speech patterns and topics that are relevant to call center operations, rather than analyzing every aspect of the conversation in exhaustive detail. This selective processing approach maintains improved call handling efficiency by focusing on key information while reducing the overall computational energy consumption compared to complete real-time analysis of all speech content.

Inventive Principle:

Principle #16Partial or excessive action

3Reliability

If automated monitoring and detection system is implemented, then reliability of call tracking is improved, but difficulty of detecting and measuring increases

Engineering Contradiction:

Improvecall tracking accuracyVSAvoidspeech pattern analysis complexity

Core Design Contradiction:

ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent replaces manual call tracking and analysis methods with an automated speech recognition engine that uses computational algorithms to detect speech patterns. This substitution of mechanical/manual processes with automated electronic systems improves the reliability and consistency of call tracking while the specialized algorithms handle the complexity of speech pattern detection and measurement.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8130937B1Use of speech recognition engine to track and manage live call center calls

Publication Date: 2012.03.06 SPRINT SPECTRUM LLC

AI summary

A speech recognition engine monitors live call center calls between live callers and live operators and detects that certain key words are spoken. The detected key words can then be used as a basis to identify issues that are raised in the call, so as to facilitate (i) generation of statistical reports regarding call center call issues and (ii) real-time assistance of the call center operator, such as directing the call center operator to ask certain questions or take certain other actions.