Conversational AI Training Agents via Synthetic Calls
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current agent training methods in contact centers are time-consuming and qualitative, relying on human trainers to monitor live calls, which can lead to inconsistent training and potentially poor customer experiences due to the need for human trainers to create interaction traffic.
Innovation Solution
A system utilizing conversational artificial intelligence (AI) to train contact center agents through virtual calls, where a chatbot interacts with agents to assess their responses based on predefined elements, analyzing duration, accuracy, and efficiency, and evaluating agent fatigue over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If human trainers monitor live calls to train agents, then training can be conducted with human guidance, but the process becomes time-consuming and cumbersome
Solution Approach 1:
The patent uses synthetic customer voices that are copies or simulations of real customer interactions. These synthetic voices allow agents to practice with realistic scenarios without requiring actual customers or human trainers to create training traffic, thereby reducing time consumption while maintaining training effectiveness
Solution Approach 2:
The system enables agents to train themselves by interacting with synthetic customer voices. The automated evaluation system provides immediate feedback on agent responses, allowing agents to independently practice and improve without requiring continuous human trainer involvement, thus reducing both time and operational complexity
2Quantity of substance
If human trainers create traffic for new agents, then training interactions can be generated, but the process becomes cumbersome and difficult to normalize
Solution Approach 1:
The synthetic customer voice system serves multiple functions: it generates unlimited training interactions, provides standardized evaluation criteria, and can be deployed across multiple agents simultaneously. This universal system replaces the need for individual human trainers to create traffic for each agent, normalizing the training process while increasing interaction volume
Solution Approach 2:
The system changes the parameter of training traffic generation from human-dependent to AI-generated. By using synthetic voices with controllable parameters (scenarios, topics, difficulty levels), the system can generate consistent, normalized training interactions without the complexity of coordinating human trainers
3Reliability
If live client interactions are used for agent training, then real training scenarios are provided, but actual clients receive poor experiences
Solution Approach 1:
The patent introduces synthetic customer voices as an intermediary between the training system and the agent. These synthetic voices provide realistic training scenarios without involving actual customers, thereby eliminating the harmful effect of poor customer experiences while maintaining training effectiveness through lifelike interactions
Solution Approach 2:
The system creates copies of real customer interactions through synthetic voices. These copies preserve the essential characteristics and challenges of actual customer service scenarios, providing reliable training effectiveness without exposing real customers to potential poor service during the training process
Data Source
AI summary
A system for utilizing conversational artificial intelligence (AI) to train contact center agents according to an embodiment includes at least one processor and at least one memory comprising a plurality of instructions stored therein that, in response to execution by the at least one processor, causes the system to place a virtual call from an automated training system to an agent device of an agent, connect the virtual call to a chatbot in response to establishing a communication connection with the agent device, transmit one or more statements from the chatbot, receive, from the agent device, one or more agent responses of the agent corresponding to the one or more statements, and analyze the one or more agent responses to determine one or more training characteristics associated with AI-based contact center training of the agent.