Skip to main content

Quickstart

Get your first AI phone agent up and running in minutes.

Prerequisites​

  • A VOCALS account at dashboard.usevocals.com
  • API keys for at least one provider in each category (STT, LLM, TTS)
  • A Twilio account with a phone number (for making/receiving calls)

Step 1: Configure providers​

Navigate to Settings > Providers in the dashboard and add your AI provider credentials:

StageRecommended ProviderWhat it does
STTDeepgramConverts caller speech to text
LLMOpenAI (GPT-4o)Generates conversational responses
TTSElevenLabsConverts response text to speech

For each provider, enter your API key and select a model.

Step 2: Create an agent​

Go to Agents and click Create Agent:

  1. Name — Give your agent a descriptive name (e.g., "Customer Support")
  2. System Prompt — Define how the agent should behave: 
    You are a helpful customer support agent for Acme Corp.
    Be friendly, concise, and helpful. If you can't answer a
    question, offer to transfer the caller to a human agent.
  3. Providers — Select the STT, LLM, and TTS providers you configured
  4. Welcome Message — Optional greeting played when a call connects (e.g., "Hi, thanks for calling Acme Corp. How can I help you today?")
  5. Language — Set the agent's language (default: English)

Click Save to create the agent.

Step 3: Connect a phone number​

  1. Go to Phone Numbers and click Add Number
  2. Enter your Twilio phone number
  3. Assign it to the agent you created
  4. VOCALS will automatically configure the Twilio SIP trunk to route calls to your agent

Step 4: Make a test call​

Call the phone number you configured. You should hear the welcome message, and then be able to have a conversation with your AI agent.

Step 5: Monitor in the dashboard​

After the call, check the Dashboard to see:

  • Call duration and status
  • Conversation transcript
  • Latency metrics (STT, LLM, TTS)
  • Cost breakdown

Next steps​