Build a Voice AI Chatbot with ElevenLabs and InfraNodus Knowledge Experts
VerifiedSets up an AI agent using OpenAI, Gemini models and memory.
What this workflow does
This automation configures an AI Agent node connected to OpenAI and Gemini chat models with Simple Memory for stateful dialogues.
It targets n8n users building multi-model AI agents that require persistent context without external vector stores.
Who is this for?
Product and support teams building voice-enabled assistants; solo developers or small teams wanting quick GraphRAG chatbots without heavy vector-store setup.
What problem it solves
Creating a voice chatbot that can query multiple knowledge bases at once usually requires complex RAG pipelines; this workflow gives the agent instant access to InfraNodus graphs via simple tool calls.
Live workflow preview
Interactive canvas of every node and connection — scroll and click to explore. Powered by n8n's preview.
Open the template on n8n to import and run it. View source template →
What it automates
Customer support voice bot
Embed ElevenLabs agent on a help site so callers receive answers drawn from several product knowledge graphs without manual retrieval setup.
Internal knowledge assistant
Let team members ask domain questions by voice and receive answers routed to the correct expert graph (finance, HR, engineering) via the AI agent.
Research briefing tool
Researchers speak queries and the agent selects the right InfraNodus graph to synthesize cross-document insights in one response.
How the workflow works
The 4 nodes in this automation, in order.
- 1AI Agent@n8n/n8n-nodes-langchain.agent
- 2OpenAI Chat Model@n8n/n8n-nodes-langchain.lmChatOpenAi
- 3Simple Memory@n8n/n8n-nodes-langchain.memoryBufferWindow
- 4Google Gemini Chat Model@n8n/n8n-nodes-langchain.lmChatGoogleGemini
Apps & integrations used
How to set up Build a Voice AI Chatbot with ElevenLabs and InfraNodus Knowledge Experts
- 1Import the workflow into n8n and activate the Webhook node.
- 2Add your ElevenLabs Conversational AI agent and point its tool endpoint to the n8n webhook URL.
- 3Connect the AI Agent node and attach either OpenAI Chat Model or Google Gemini Chat Model.
- 4Link the Simple Memory node using the sessionID passed from ElevenLabs.
- 5Create one tool per InfraNodus knowledge graph inside the AI Agent and paste the auto-generated descriptions.
- 6Test by sending a voice prompt through ElevenLabs and verify the agent selects the correct tool.
How to customize this workflow
- →Swap OpenAI Chat Model for Google Gemini Chat Model or vice versa.
- →Add more InfraNodus graphs as additional tools to the AI Agent.
- →Replace the ElevenLabs webhook trigger with another voice or chat platform.
- →Insert an extra formatting node before the response is returned to ElevenLabs.
Build a Voice AI Chatbot with ElevenLabs and InfraNodus Knowledge Experts: pros & cons
Pros
- +No vector-store import pipelines required
- +GraphRAG gives relational answers across documents
- +Easy to reuse the same graphs in other n8n workflows
- +Voice interface ready via ElevenLabs embed or API
Cons
- –Depends on paid ElevenLabs and InfraNodus accounts
- –Tool descriptions must be maintained manually if graphs change
- –Session memory is basic Simple Memory only
Frequently asked questions
It connects an ElevenLabs voice agent to an n8n AI Agent that routes questions to one or more InfraNodus GraphRAG knowledge bases.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar workflows
Other ai & llm automations worth a look.