Canonical AI's Voice AI Evaluation platform offers comprehensive monitoring and analytics for Voice AI agents, enabling businesses to map call journeys and gain real-time insights into call performance. By identifying critical issues and anomalies, the platform helps improve the efficiency and effectiveness of voice-based customer interactions.
Key Features and Functionality:
- Intelligent Insights: The platform monitors Voice AI agent call journeys, providing real-time alerts on a per-agent basis for critical issues.
- Successful Path Anomalies: Detects deviations in call paths with historically high success rates and notifies users of anomalies.
- Duration Percentile Analysis: Analyzes the 75th percentile of call durations, alerting users to significant deviations when failures occur.
- Time Bucket Monitoring: Creates time-based buckets for call durations, identifying and notifying users of unexpected failures within these periods.
- Visualizations and Dashboards: Provides enterprise-grade analytics to visualize Voice AI agent call journeys, including:
- Call Path Analysis: Maps the entire call journey from greeting to outcome, highlighting successful and unsuccessful paths.
- Agent Response Time: Measures the time taken for agents to respond, aiding in performance assessment.
- Signal-to-Noise Ratio: Evaluates audio clarity, categorizing it as great, good, fair, or poor.
- Sad Path Analysis: Identifies major failure categories within call interactions.
- Call Duration Distribution: Explores time domains associated with successful and failed calls.
Primary Value and Problem Solved:
The Voice AI Evaluation platform addresses the challenge of maintaining high-quality, efficient Voice AI interactions by providing detailed analytics and real-time monitoring. It enables businesses to quickly identify and rectify issues within their Voice AI systems, ensuring optimal performance and enhanced customer satisfaction. By offering insights into call paths, response times, and audio quality, the platform empowers organizations to make data-driven decisions to improve their voice-based services.