​

Voice Workflow Quickstart

​

🧩 Step 1: Create a Workflow

1. Visit the Workflow List.
2. Click Create Workflow to initiate a new workflow.

​

📝 Step 2: Add Introduction Node

• An Introduction Node will be auto-created.
• Add a prompt to guide the initial conversation. For example:

Acknowledge the user's voice agent use case, then ask for information about the user. Check if the user is available to talk.

• Click Save to confirm your node.

​

🔀 Step 3: Create Condition & Conversation Nodes

1. Add a Condition Edge from the Introduction Node:
   • Example condition: User is free
2. Connect it to a Conversation Node:
   • Example prompt: If user is interested in attending workshop, proceed with relevant details.
3. Click Save to store the configuration.

​

⚙️ Step 4: Global Configuration

• LLM Model — select the model to be used throughout the workflow.
• Transcriber — choose the speech-to-text provider.
• Text to Speech — configure the voice for agent responses.

⚠️ Note: These three configurations apply globally across all nodes in the workflow. They remain consistent for the entire conversation flow.