Voice AI Made Simple | Beginners Guide to n8n + ElevenLabs (FREE n8n Template)

Key Concepts

• Voice AI Agent: An automated system that interacts with users through voice, simulating a human assistant.
• 11 Labs: A platform for creating realistic and customizable AI voices and voice agents.
• N8N: A no-code workflow automation platform used to build the backend logic for the voice agent.
• Webhooks: A method for applications to communicate with each other in real-time, used to connect 11 Labs and N8N.
• System Prompt: Instructions given to the AI agent to define its persona, goals, and behavior.
• Custom Tools: Functions that the AI agent can use to interact with external services, such as checking availability or booking meetings.
• Knowledge Base: A repository of information that the AI agent can access to answer user questions.
• JSON: A data format used for transmitting data between a server and a web application.
• LLM (Large Language Model): The AI model used to process and generate text, such as GPT-4.

Creating a Personal Consultation Voice AI Agent

Overview

The video demonstrates how to create a personal consultation voice AI agent using 11 Labs and N8N. The agent can check availability and book consultation meetings, providing a hands-free solution for scheduling. The creator provides a free template and resources to help viewers build their own agent.

Setting up N8N Workflow

1. Sign up for N8N: Create a free account on N8N.
2. Create a Workflow: Start a new workflow in N8N.
3. Import JSON File: Download the provided JSON file from the community resources and import it into the N8N workflow. This file contains the pre-built backend logic for the agent.
   • The JSON file includes the necessary nodes for handling webhooks, interacting with the AI agent, and connecting to Google Calendar.

Configuring 11 Labs Agent

1. Sign up for 11 Labs: Create a free account on 11 Labs. Consider upgrading to a paid plan for more tokens.
2. Navigate to Agents Platform: Go to the "Creative Platform" and then select "Agents" to access the voice agent creation interface.
3. Create a New Agent: Click on "New Agent" and choose "Blank Agent." Give the agent a name.
4. Configure Agent Settings:
   • Language: Set the agent language to English.
   • First Message: Customize the initial greeting message (e.g., "Hi, I'm Jarvis. How can I help you today?").
   • System Prompt: Copy the provided system prompt from the community resources and paste it into the system prompt field.
     • The system prompt defines the agent's persona (e.g., Zubar's personal assistant), goals (e.g., checking availability and booking meetings), and behavior (e.g., responding professionally and efficiently).
     • The prompt instructs the agent to use specific tools for checking availability and booking meetings.
   • LLM: Select a Large Language Model (LLM) such as GPT-4 or GPT-5.
5. Add Custom Tools:
   • Get Available Slots Tool:
     • Create a new tool of type "Webhook Tool."
     • Set the "Post URL" to the webhook URL from the N8N workflow.
     • Add a description for the tool (e.g., "Use this tool to check availability of the requested date and time for appointments").
     • Add a query parameter named "request" to send the user's request to the N8N workflow.
   • Book Meeting Tool:
     • Create another webhook tool.
     • Set the "Post URL" to the same webhook URL from the N8N workflow.
     • Add a description (e.g., "Use this tool to book a consultation meeting").
     • Add query parameters for "date," "email," "time," "phone number," and "name" to collect the necessary information from the user.
6. Add Knowledge Base (Optional):
   • Upload a PDF document containing information about consultation fees or other relevant details.
   • The agent can access this information to answer user questions.
7. End Call Tool (Optional):
   • Enable the "End Call" tool to allow the agent to hang up after the conversation is over.

Configuring N8N Workflow Details

1. Webhook Node:
   • Set the "Path" to a unique identifier (e.g., "test").
   • Ensure the webhook is set to "Test" mode for development.
2. AI Agent Node:
   • Map the incoming data from the webhook to the AI agent's input fields.
   • Map the user's name, email, phone number, date, and time to the appropriate fields.
   • Include the user's query request to provide context to the agent.
   • The system prompt in N8N is simpler, focusing on the agent's role as a helpful assistant and providing guidelines for using the available tools.
3. Google Calendar Nodes:
   • Get Events Node:
     • Connect to your Google Calendar using your Google credentials.
     • Set the "Resource" to "Event."
     • Configure the node to retrieve events from your calendar within a specified time range.
   • Create Event Node:
     • Connect to your Google Calendar.
     • Set the "Resource" to "Event" and the "Operation" to "Create."
     • Map the data from the AI agent to the event fields, such as start time, end time, and attendee email.
     • Add the attendee's email to the "Attendees" field to send a meeting invitation.
4. SER API Node (Optional):
   • Integrate with the SER API to allow the agent to answer questions about current events.
5. Respond to Webhook Node:
   • Configure the node to send a response back to 11 Labs.
   • Stringify the JSON output to ensure it is valid.

Testing and Deployment

1. Test the Agent: Use the "Test Agent" button in 11 Labs to test the agent's functionality.
2. Embed the Agent on Your Website:
   • Copy the provided JavaScript code snippet from 11 Labs.
   • Embed the code into your website's JavaScript to add the voice agent as a widget.

Monetization

The creator mentions that they have a course on how to monetize these agents, including pricing strategies and how to sell them to clients.

Community Resources

The creator provides access to a free community where users can find resources, ask questions, and get support. The community also offers courses on building and monetizing AI agents.

Notable Quotes

• "Hi, I'm Jarvis. How can I help you today?" - Example of a customizable first message for the AI agent.
• "You're Z's personal assistant. You're professional, efficient, blah, blah, blah." - Example of a system prompt defining the agent's persona.
• "Use this tool to check availability of the requested date and time for appointments." - Example of a description for the "Get Available Slots" tool.

Technical Terms

• Webhook: A method for applications to communicate with each other in real-time.
• System Prompt: Instructions given to the AI agent to define its persona, goals, and behavior.
• Custom Tools: Functions that the AI agent can use to interact with external services.
• Knowledge Base: A repository of information that the AI agent can access to answer user questions.
• JSON: A data format used for transmitting data between a server and a web application.
• LLM (Large Language Model): The AI model used to process and generate text.
• Query Parameters: Values passed in the URL of a webhook request.
• Body Parameters: Values passed in the body of a webhook request.

Logical Connections

• The 11 Labs agent uses webhooks to communicate with the N8N workflow.
• The N8N workflow uses Google Calendar nodes to check availability and book meetings.
• The system prompt in 11 Labs instructs the agent to use specific tools in the N8N workflow.
• The knowledge base in 11 Labs provides the agent with information to answer user questions.

Synthesis/Conclusion

The video provides a comprehensive guide to creating a personal consultation voice AI agent using 11 Labs and N8N. By leveraging the provided template and resources, users can build a functional agent that can automate scheduling tasks. The video also touches on the potential for monetizing these agents by selling them to businesses. The key takeaways are the importance of a well-defined system prompt, the use of custom tools to extend the agent's capabilities, and the potential for integrating the agent into a website as a widget.