Voicetta
    Voicetta.

    Voicetta API

    Voicetta API

    Build your own product on top of Voicetta. Use voice, SMS and WhatsApp through one communication backend, with webhooks and a REST API for everything that happens.

    Build on Voicetta

    Voicetta handles customer communication across voice, SMS, and WhatsApp.

    You build the product around it.

    Use the Voicetta API to power your own dashboard, CRM, white-label product, workflow, or customer experience.

    You can receive events through webhooks, or read the same data through the REST API.

    What Voicetta handles

    Voicetta takes care of the communication layer.

    That includes:

    • voice calls
    • SMS
    • WhatsApp
    • transcripts
    • conversation analysis
    • outcomes and intents
    • evaluations
    • recordings
    • follow-ups
    • conversation history

    Your product decides what happens with that information.

    Store it in your database. Show it in your own UI. Add it to a CRM. Trigger a workflow. Build reports. Or build an entirely new product around it.

    What you can build

    A white-label product

    Voicetta runs the communication and AI agents.

    Your customers use your product, your UI, and your branding.

    They do not need to know that Voicetta is behind it.

    Your own communication dashboard

    Build your own inbox, conversation timeline, activity feed, reporting, or customer view across voice, SMS, and WhatsApp.

    CRM or PMS integrations

    Push calls, messages, transcripts, outcomes, and customer information into the systems your team already uses.

    Workflow automation

    Trigger actions based on what happened in a conversation.

    For example:

    • create a lead after an inbound call
    • notify a sales rep
    • update a booking
    • escalate an unhappy customer
    • send a follow-up
    • mark a conversation as resolved

    Your own AI product

    Use Voicetta as the communication layer behind a product you are building.

    You do not need to build the telephony, messaging, conversation storage, transcripts, recordings, or conversation evaluation yourself.

    The API

    There are three main resources.

    ResourceEventAPI
    Voice callcall.completed/v1/calls
    SMS or WhatsApp messagemessage.received, message.sent/v1/messages
    Completed text conversationthread.completed/v1/threads

    Messages and completed threads are linked through thread_id.

    Messages tell you what is happening.

    Threads tell you how the conversation ended and how it was evaluated.

    Webhooks or REST API

    You can use webhooks, the REST API, or both.

    WebhookREST API
    Who starts itVoicetta sends the eventYou make the request
    Best forKeeping your product up to dateHistory, backfills, recovery
    AuthenticationSigned webhookAPI key
    EndpointYour HTTPS URLhttps://api.voicetta.com/v1

    For most integrations, start with webhooks.

    Use the REST API when you need historical data or need to recover something your backend missed.

    The same underlying objects are available through both, so you can use one data model.

    Getting access

    1. Open Settings → Developers in Voicetta.
    2. Create an API key, a webhook endpoint, or both.
    3. Choose the events and data you want.
    4. Start building.

    API keys and webhook secrets are shown once when created. If you lose one, rotate it and create a new one.

    When events arrive

    Events are sent when the relevant data is ready.

    EventFiresTypical delay
    call.completedPost-call processing is finishedA few seconds after hang-up
    message.receivedA customer message is received15–30 seconds
    message.sentVoicetta sends a message15–30 seconds
    thread.completedA text conversation is closed and gradedAbout 30 minutes after the last message

    A text conversation is different from a message.

    The message event tells you what is happening now.

    The thread event tells you how the conversation went after it is finished.

    Version 1

    All API paths use /v1.

    Within version 1, we only add fields. We do not rename, remove, retype, or change the meaning of existing fields.

    Breaking changes get a new API version.

    Design your integration to:

    • ignore fields it does not recognize
    • treat missing fields as normal
    • switch on event
    • ignore events it does not recognize instead of failing

    New events and fields should not break an existing integration.

    For AI coding tools

    Every developer page has a Markdown version.

    • This page: /developers/overview.md
    • All developer docs: /developers.md
    • OpenAPI 3.1 spec: /open-api

    The OpenAPI spec is generated from the running API.

    You can point a code generator at it or give /developers.md to your coding agent.