Build with Call2Me

The shortest path from sign-up to a working voice agent on a real phone number. Five steps, five minutes, zero credit card.

How to authenticate with Call2Me — Bearer API keys for server-to-server, JWT for dashboard sessions, and rotation patterns that don't break production.

Reserve a temporary phone number for 15 minutes — perfect for prospect demos, hackathons, and 'try before signup' flows. The shortest path from idea to a ringing phone.

The dashboard chat that operates the whole platform for you — create agents, check calls, buy numbers, read your balance — by calling the same API you use, with a confirmation step before anything that writes, deletes or pays.

How to configure a Call2Me voice agent — system prompts, LLM model, voice, knowledge base, post-call extraction, transfer targets, recording, and webhooks.

How to upload PDFs, web pages, and documents so your voice agent answers from your actual content instead of hallucinating.

How Call2Me's voice stack works — Deepgram Nova-3 STT, ElevenLabs Flash v2.5 TTS or GPT Realtime speech-to-speech, sub-500ms target latency. Picking the right voice per language.

How calls work on Call2Me — buying numbers, inbound routing, outbound campaigns, transfers, recordings, and what shows up in the Calls tab.

Run outbound voice campaigns end-to-end — CSV upload, concurrency, retries, real-time monitoring, post-call data extraction. From setup to first live call in 30 minutes.

Book a voice AI call to fire at a specific date and time — once, or on a recurring schedule. Time zones, recurrence rules, and how to wire schedules to post-call actions.

Define structured fields to extract from every call, then trigger webhooks or auto-schedule follow-ups based on the extracted values.

Send a message automatically when a call or chat ends — to the caller, to you or to your team — over WhatsApp, SMS, email or your own URL, with conditions on the extracted call data.

The 26 voices on Call2Me — 16 ElevenLabs Flash v2.5 voices for standard mode and 10 GPT Realtime voices — with languages, gender and how to audition them before going live.

Run text-based conversations with the same agent that answers your phone — embedded widget, REST API, or your own UI. Same prompt, same knowledge base, different channel.

Drop a single script tag and your voice agent becomes a chat widget on every page — same prompt, same knowledge base, same brand.

Buy a phone number from Call2Me, bring your own via SIP, route inbound calls to agents — the lifecycle of a number on the platform.

Connect your existing SIP carrier (NetGSM, Twilio, Telnyx, Asterisk) to Call2Me — keep your numbers and contracts, get the voice AI on top.

Run a voice agent inside any web app — no phone number needed. Get a LiveKit token, embed the SDK, talk to the agent from a browser tab in under 200ms latency.

Simultaneous two-way interpretation between 31 languages, over the phone or in a browser. Set up a language pair, share a link or dial out, and both sides hear each other in their own language.

Open an AI voice session with a Call2Me agent from your own frontend using an API key — get a LiveKit token and connect. No widget required.

Send SMS from a Call2Me number you own and read inbound replies through the API. US and Canada destinations, $0.02 per segment, with rate and length limits that keep a leaked key from running up a bill.

The Call2Me Chrome extension translates the audio of any tab — a video, a webinar, a meeting — live into 39 languages. Free daily minutes without an account, more when you link your Call2Me account.

Call2Me REST API — base URL, authentication (JWT + API keys), endpoint groups (agents, calls, knowledge base, widgets, payments, webhooks), and the full OpenAPI spec.

Let a voice agent invoke your APIs mid-call — book a meeting, look up an order, transfer to a human. The function-calling protocol, parameter schemas, and the patterns that keep latency low.

Post events to Call2Me from your own frontend, and read live audio over the voice WebSocket. There is no event-streaming WebSocket — this page documents what actually exists.

HTTP status codes Call2Me returns, rate limit headers, retry-after semantics, and the patterns that make integrations survive carrier hiccups.

Receive call_started, call_ended, call_analyzed and transfer events from an agent, plus signed account-level events for sessions and balance. This page documents the two delivery paths as they actually behave.

Export any completed call as an IETF vCon container — parties, dialog (transcript + recording reference) and analysis (summary, sentiment, extraction) in one portable JSON file. Opt-in per agent.

Mint short-lived, scoped Call2Me tokens with your API key so a mobile client can chat and talk to your agents without ever holding your sk_ key. Usage is billed to your tenant wallet.

Connect Call2Me to any MCP client and place calls, read transcripts, manage interpreters and buy numbers in plain language. 37 tools behind OAuth 2.1 with PKCE — your API key never reaches the model.

Per-minute voice base ($0.10), telephony (+$0.01/min per leg), outbound (+$0.03/min per leg), recording (+$0.005/min), chat widget messages ($0.01/msg). $5 free credit on signup. No subscriptions.

How Call2Me bills — prepaid wallet model, top-ups via Paddle, auto-recharge thresholds, and where to find every charge in the dashboard.

Where Call2Me stores data, what's encrypted, what's logged, and how to handle PII responsibly when your agent might hear it.

Custom domain, branding, tenant routing, and per-tenant billing — everything you need to resell voice AI under your own brand.

Invite customers to your white-label brand, grant them credit from your balance, set your margin, and track what each one earns you.

Plain-language definitions for every term that shows up in voice AI — STT, TTS, VAD, RAG, SIP, BYOC, function calling, and the rest.