Introduction to AI Agents
An AI Agent is an advanced interactive AI designed to communicate with users like a human customer service representative, meeting communication needs across various business scenarios.
It is available in two main forms:
-
Voice Virtual Agent: Engages in natural, fluent voice conversations to perform tasks such as intelligent outbound calling. It can understand user speech, respond accurately, and execute business operations based on the dialogue.
-
Digital Virtual Agent: Provides 24/7 online text-based customer service through text channels such as websites and instant messaging tools (e.g., WhatsApp).
Both types of agents are capable of understanding user intent, giving precise responses, and executing predefined business workflows. This ensures convenient, efficient, and personalized services, greatly enhancing business interaction efficiency and customer satisfaction.
Quick Start
1. Voice Virtual Agent (Intelligent Outbound)
The operation of a Voice Virtual Agent consists of three main parts: Configuration, Testing, and Campaign Management (Voice Channel).
Configuration
You can customize how the agent “speaks,” “listens,” and “thinks” during interaction with customers:
-
Speak (TTS): Configure speech synthesis in AI Center → Voice.
-
Listen (ASR): Configure speech recognition when creating a new voice agent in AI Center → Voice Virtual Agent.
-
Act (LLM + Workflow): Design workflows and integrate large language models within the Voice Virtual Agent.
Testing
After configuration, click the Test button in the upper-right corner to verify that the AI settings are correct and the functions are working properly.
Campaign Management (Voice Channel)
-
Task Setup: In Campaign Management → Voice Channel, create a new campaign and configure outbound parameters.
-
Task Monitoring: View execution status, progress, and performance reports.
-
Detailed Records: Access detailed communication logs for analysis and troubleshooting.
2. Digital Virtual Agent (Text-based Customer Service)
The operation of a Digital Virtual Agent also consists of three main parts: Configuration, Testing, and Channel Management.
Configuration
-
Act (LLM + Workflow): Design the logic and workflows for text-based conversations. Since the interaction is purely text-based, TTS and ASR are not required.
Testing
Use the text chat window to validate the dialogue flow and logic.
Channel Management
Deploy the configured digital agent to specific business channels (e.g., WhatsApp).