LiveKit gets your agent talking. ShipVoice adds the money layer: per-minute Stripe billing so you charge your users for exactly what they use, plus auth, telephony, a call dashboard, and one-command deploy. Own the code, own the margin.
Launches September 2. Founding price locked for early members. No spam.
The agent you just talked to was built with ShipVoice. Open-source core: livekit-starter ↗
Built on the stack behind OpenAI, Character.ai, and Retell
Turning a working agent into something you can charge for is where the months go.
LiveKit's free starter gets you a great agent. Managed platforms rent you one per minute, forever. ShipVoice lets you own the whole thing and keep the margin.
livekit-starter (open source)
Vapi, Retell, Bland
own the code + the margin
Managed platforms stack to $0.15 to $0.36 all-in per minute once you add STT, LLM, TTS, and telephony. Above ~20,000 minutes a month, owning it wins.
Three layers: the voice core, the money layer that lets you charge, and the pieces that help you sell.
STT, LLM, and TTS wired together with turn detection and interruption handling. Provider-swappable across OpenAI, Deepgram, Cartesia, and ElevenLabs. No lock-in.
SIP and PSTN set up so your agent answers real phone calls and places them. The part that "requires deep telephony knowledge" is already done.
A polished, embeddable mic widget for the web and a mobile-ready frontend. Ship a talkable product, not a terminal demo.
The hard part of a voice product: meter every minute across STT, LLM, TTS, and LiveKit, add your margin, and charge your users automatically.
Sign-up, login, and per-user usage tracking out of the box, so you can bill each of your users for exactly what they use.
Recordings, transcripts, and per-call cost, so you and your users can see exactly what happened and what it cost.
Change the prompt, voice, and tools from a UI, so non-developers on your team can tune agents without a deploy.
Docker plus deploy configs for Fly, Render, and LiveKit Cloud. Go from clone to a running, billable agent in an afternoon.
Start from a working receptionist or outbound-caller template instead of a blank file, plus a landing-page template to sell your product.
This receptionist runs on ShipVoice and takes leads by voice. Ask it anything, or tell it what you are building.
The agent captures your name, email, and what you are building, by voice.
Clone the kit, point a phone number at it, and you have a billable AI receptionist. Then make it yours.
Set the prompt, voice, and tools from the config UI. No blank file, no deploy.
Point a SIP / PSTN number at your agent. It answers, qualifies, and books, on the first ring.
STT, LLM, TTS, and telephony usage is metered per call, with your margin added on top.
Stripe bills each user automatically for what they used. You keep the difference.
[ product walkthrough clip / dashboard screenshot goes here ]
A modern stack you already trust
The LiveKit foundation ShipVoice builds on is open source and MIT-licensed. Read it, run it, fork it. ShipVoice is that core plus the money layer (billing, dashboard, telephony, deploy) that turns it into a product you can charge for.
github.com/mahimairaja/livekit-starter ↗One-time, no subscription. Founding price locked for early members, going up at launch.
one-time · single developer · lifetime updates
Join the founding list. Launches September 2, price locked for early members.
I'm Mahimai. I build voice agents. The one you just talked to on this page is mine, running on ShipVoice.
I got tired of rebuilding the same money-and-plumbing layer for every voice product: auth, per-minute metering, billing, telephony, a dashboard. So I packaged it. The core is open source (livekit-starter); ShipVoice is that core plus everything you need to actually charge for it.
No. The free starter (our open-source livekit-starter) gets you a working agent. ShipVoice adds the business: Stripe metered billing, auth, a call dashboard, telephony, and vertical templates. The part that lets you charge your users.
ShipVoice launches September 2. Join founding access to get it first, at the lowest price.
LiveKit Agents (Python and TypeScript) for the voice backend, a Next.js dashboard and frontend, Stripe for metered billing, Postgres for data, and Docker for deploy. Model providers are swappable.
A single-developer license with unlimited projects and lifetime updates at the founding price. A team license comes later at a higher price.
Telephony is wired in, but 10DLC and A2P carrier registration is an external step (roughly two weeks at the carrier) that no code can remove. ShipVoice ships the scaffolding and a guide so you are not figuring it out alone.
You still own the code and keep your margin. That is the one thing a platform will never hand you, and it is the reason to own your stack instead of renting it.
Yes. Full refund terms are posted at checkout. If ShipVoice is not the right fit, reach out and we will make it right.
Launching September 2, 2026
Founding members get in first, at the lowest price.
Get on the founding list. Lock the lowest price and ship your voice product instead of your billing system.
Get founding access. Launches September 2.