TryVox

Quickstart

Make your first voice call in under 5 minutes.

Quickstart

Get up and running with TryVox in under 5 minutes. This guide walks you through making your first AI-powered voice call.

Prerequisites

  • A TryVox account (sign up at tryvox.io)
  • curl or any HTTP client

Step 1: Get Your API Key

  1. Log in to the TryVox Dashboard
  2. Navigate to Settings > API Keys
  3. Click Create API Key
  4. Copy your API key — it starts with tvx_sk_live_...

Keep this key secure. You'll use it to authenticate all API requests.

Step 2: Provision a Phone Number

Before making calls, you need a phone number. Let's purchase one from our inventory:

curl -X POST https://api.tryvox.io/v1/account/{account_id}/numbers/purchase-from-inventory \
  -H "Authorization: Bearer tvx_sk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "country": "IN",
    "type": "local"
  }'

Replace {account_id} with your account ID from the dashboard.

Response:

{
  "data": {
    "e164": "+919876543210",
    "country": "IN",
    "status": "active",
    "type": "local",
    "capabilities": ["voice", "sms"]
  }
}

Your number is now active and ready to use!

Step 3: Make an Outbound Call

Now let's make a call using your new number:

curl -X POST https://api.tryvox.io/v1/voice/accounts/{auth_id}/calls \
  -H "Authorization: Bearer tvx_sk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "from": "+919876543210",
    "to": "+14155551234",
    "answer_url": "https://your-app.com/answer"
  }'

Parameters:

  • from — Your TryVox phone number (from Step 2)
  • to — The destination phone number (E.164 format)
  • answer_url — Your webhook URL that returns VoxML instructions

Response:

{
  "call_uuid": "c7a34e6f-9d1b-4c8e-a5f2-3b9d7e8c1a4f",
  "status": "queued",
  "from": "+919876543210",
  "to": "+14155551234"
}

Step 4: Control the Call with VoxML

When the call is answered, TryVox sends a request to your answer_url. Your server should return VoxML instructions to control the call.

Here's an example response that plays a greeting and streams audio to your AI agent:

{
  "voxml_version": "1.0",
  "instructions": [
    {
      "verb": "Say",
      "text": "Hello! I am your AI assistant.",
      "engine": "google",
      "language": "en-IN"
    },
    {
      "verb": "Stream",
      "url": "wss://your-app.com/ai-agent",
      "track": "inbound_track"
    }
  ]
}

What this does:

  1. Say — Plays a text-to-speech greeting using Google's TTS engine
  2. Stream — Opens a WebSocket connection to your AI agent for real-time audio streaming

TryVox will execute these instructions in order, streaming the caller's audio to your WebSocket endpoint.

Next Steps

You've made your first call! Here's what to explore next:

Need Help?

On this page