PurpleCallio

For developers

Build real-time communication into your application.

Add audio, video, screen sharing, real-time speech translation and AI voice agents using PurpleCallio's hosted UI, React components, headless SDK or REST APIs.

Fastest integration

Hosted UI

A ready-made meeting interface.

Custom UI

React

Composable components for React products.

Full control

Headless SDK

The engine beneath your own interface.

Backend integration

REST API

Create and manage calls from any trusted server environment.

Early access · Live Translate

Let two people talk, each in their own language.

Add real-time voice translation and captions to any 1:1 call with one flag. PurpleCallio runs speech recognition, translation and synthesized voice on the server for 60 languages, any to any, with misunderstanding flags, budgets and metrics built in. Billed per minute for each person translating, from the same credit balance.

  1. 01Turn it on per callAdd callerLanguage, receiverLanguage and a translation block to POST /calls.Read this step
  2. 02Show captions and play voiceTranslationClient or the React UI Kit streams DualView captions and translated audio.Read this step
  3. 03Catch misunderstandingsMeaningGuard flags changed amounts, dates and names; listeners can ask to repeat.Read this step
  4. 04Keep costs in limitsPer-call minutes, project budgets and a public estimate endpoint, enforced on the server.Read this step
// Server: create a translated call with your API key
await fetch(`${API_URL}/calls`, {
  method: "POST",
  headers: { "x-api-key": API_KEY, "Content-Type": "application/json" },
  body: JSON.stringify({
    callerId: "agent_7", receiverId: "cust_42",
    callerLanguage: "en", receiverLanguage: "es",
    translation: { mode: "voice", maxMinutes: 30, onLimit: "captions_then_off" },
  }),
});

// Browser: each participant enables translation for themselves
const translation = new TranslationClient({ serverUrl, token, speak: "es", hear: "es", mode: "voice" });
await translation.enable();
translation.on("caption", (c) => render(c.original, c.translated));
translation.on("flagged", (c) => showCheck(c.phraseId));

New · AI Voice Agents

Build AI voice agents with the same API keys and SDK.

Real-time, interruptible voice conversations that look things up and take actions through your own APIs. PurpleCallio runs the audio relay, the model connection, tool calls, transcripts and per-minute credit billing. Unlocks after your first verified credit purchase.

Read the AI agents docs
  1. 01Configure an agentInstructions, greeting, voice, language and tools, set in the dashboard.Read this step
  2. 02Create a session on your serverclient.ai.createSession() returns a short-lived token for the browser.Read this step
  3. 03Talk to it in the browserUse the headless PurpleCallioAgentCall or the drop-in React component.Read this step
  4. 04Let it call your APIsSigned tool requests to your endpoints, with optional user confirmation.Read this step
// Server: create a session with your API key
const { token } = await client.ai.createSession({ agentId, participant: { id: user.id } });

// Browser: start the conversation from a click
const call = new PurpleCallioAgentCall({ token, apiUrl });
await call.start();

Official SDKs for every platform

Choose the integration that fits your product

Start with the amount of frontend control you need. Every path keeps call creation and authorization in your backend.

Hosted UI

Ship a call experience with minimal frontend work.

Best for
Your backend creates the call and decides who can join.
What you get
Meeting UI, media controls, signaling, WebRTC setup, and TURN relay.
Installation
No frontend package required.
Read Hosted UI documentation →
integration.ts
import { PurpleCallioClient } from "@purplecallio/sdk";

const client = new PurpleCallioClient({ apiKey: process.env.PURPLECALLIO_API_KEY! });
const call = await client.createCall({ callerId, receiverId });
redirect(call.callerUrl);

How the pieces connect

Select a node to see its role in a production communication flow. PurpleCallio works around browser WebRTC; it does not replace the browser media layer.

Your application
PurpleCallio

REST API: Create and manage calls from trusted backend code.

What do I actually have to build?

Hosted UI

You build: You build authentication, business logic, and call creation.

PurpleCallio handles: PurpleCallio handles the meeting UI, media controls, signaling, WebRTC setup, and TURN relay.

React

You build: You build your product layout, business logic, and branded UI.

PurpleCallio handles: PurpleCallio handles participant state, media streams, controls, and the connection lifecycle.

Headless SDK

You build: You control the full UI, interactions, and application state.

PurpleCallio handles: PurpleCallio provides the communication engine, signaling, WebRTC media lifecycle, and relay path.

Developer toolbox

Start with a server-controlled call

Keep API keys on the server, issue participant-specific access only after your application authorizes a user, and then choose the meeting surface that fits the product.

Build with PurpleCallio

Explore the documentation to choose hosted UI, React components, or the headless SDK for your integration.