Automated Multimedia Call Center Agent Using Speech Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Call center agents face challenges in efficiently resolving customer issues, particularly with complex problems like home wireless router setup, as existing systems lack the ability to provide comprehensive multimedia guidance and often require human intervention.

Innovation Solution

An automated multimedia call center agent that uses speech recognition and rich multimedia content, including documentation, FAQs, and video tutorials, to identify and address customer issues, and transfers sessions to live agents when necessary, allowing for multimedia communication to facilitate equipment connection and configuration support.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional voice call systems are used for customer support, then human agent intervention is required to resolve issues, but this increases operational costs and reduces efficiency

Engineering Contradiction:
Improveissue resolution efficiencyVSAvoidautomation level
Core Design Contradiction:
ProductivityVSExtent of automation

Solution Approach 1:

The system enables customers to resolve issues independently through automated speech recognition and multimedia content delivery. The automated agent guides customers through troubleshooting steps using voice interactions and provides multimedia resources (videos, documentation, FAQs) that customers can access on their devices, eliminating the need for human agent intervention in routine issues.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical human agent system with an automated electronic system comprising speech recognition software, multimedia content management, and automated communication protocols. The automated agent processes customer inquiries, retrieves appropriate multimedia content, and delivers guidance through various channels, substituting human mechanical intervention with automated electronic processes.

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

2Reliability

If comprehensive multimedia content is provided to customers, then issue resolution effectiveness improves, but system complexity increases

Engineering Contradiction:
Improveproblem resolution effectivenessVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The automated agent system performs multiple functions within a single platform: speech recognition for customer input, natural language processing for intent identification, multimedia content retrieval from various sources (videos, documentation, FAQs), and multi-channel content delivery (email, SMS, in-app). This multi-functional approach provides comprehensive support while consolidating complexity into an integrated system.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system introduces an automated agent as an intermediary between the customer and the extensive multimedia content library. The agent acts as a mediator that understands customer queries through speech recognition, translates them into appropriate content requests, and delivers relevant multimedia resources, thereby managing the complexity of content delivery while ensuring effective problem resolution.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Extent of automation

If automated speech recognition is implemented, then human intervention is reduced, but accuracy in understanding complex customer issues may decrease

Engineering Contradiction:
Improveautomation levelVSAvoidspeech recognition accuracy
Core Design Contradiction:
Extent of automationVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where the automated agent analyzes customer responses, evaluates the effectiveness of provided multimedia content, and adjusts subsequent interactions accordingly. The system can request clarification when speech recognition uncertainty is detected and validates whether the provided content resolved the issue, continuously improving accuracy through feedback loops.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary speech recognition processing to identify customer intent and issue type before full interaction begins. By pre-processing and categorizing customer inputs, the system can route complex queries to more sophisticated handling mechanisms and prepare appropriate multimedia content in advance, improving overall recognition accuracy for complex issues.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8848897B2Automated multimedia call center agent
Publication Date: 2014.09.30 VERIZON PATENT & LICENSING INC
  • US8848897B2 patent drawing
  • US8848897B2 patent drawing
  • US8848897B2 patent drawing

AI summary

An automated multimedia call center device may receive a verbal request for information from a user device during a multimedia session between the automated multimedia call center device and the user device. The automated multimedia call center device may further obtain a group of recognition results for the verbal request using speech recognition, cause at least two recognition results of the group of recognition results to be visually displayed on the user device, receive selection of one recognition result of the at least two recognition results, perform a search using the selected one recognition result to obtain multimedia content, and provide the multimedia content to the user device.