Overview
We’ve rolled out an exciting new upgrade to Voice AI:
Knowledge base integration
, powered by smart trigger prompts and dynamic query resolution.You can now assign a
dedicated knowledge base
to each Voice-AI agent. When a user’s spoken input matches a custom trigger
, the agent will automatically search the knowledge base and return the most relevant, contextual response
ensuring intelligent, accurate, and timely interactions.This enhancement lives under
Agent Goals
(available in both Basic
and Advanced
modes), giving you more control over how and when your voice agents respond using your own knowledge resources.Why it matters
- Smarter Conversations: Agents can answer with precision by tapping into the right content at the right moment.
- Faster Setup: No need for hardcoded responses; just link a knowledge base and go.
- Scalable & Future-Proof: Updating your knowledge base automatically improves agent replies, with zero code changes.
What’s new?
- Per-Agent Knowledge Base Assignment: Attach a unique knowledge base to each voice agent to reflect its use case or audience.
- Trigger-Based Invocation: Set a trigger prompt that determines when the knowledge base should be tapped—based on the user’s intent.
- Dynamic Query Matching: Once triggered, the agent interprets the full user query and fetches the most relevant knowledge base content.
- Seamless Maintainability: Any changes to the linked knowledge base instantly reflect in live conversations—no extra logic needed.
- Create a new Knowledge Base: Click on theCreate NewButton to create a new Knowledge Base, and link it with your voice agent.
Points to Note
- You’ll find this configuration under Voice AI Agents, select your Voice AI Agent, Agent Goals > Knowledge Base, whether you're in Basic or Advanced mode.
- Ideal for FAQ handling,product info delivery, anddomain-specific voice workflows, making your agents more context aware.
- Only oneKnowledge Base per Voice AI Agentis supported.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article