Pricing

Free for everyone. Always.

Signchat is open source and runs entirely in your browser. Use it for a video call with a friend, a class group project, or a company all-hands — same code, same price.

Two ways to use it

Free forever, with an enterprise option when you need one.

Most people only ever need the free app. Companies that want SSO, a private deployment, and a humans-on-call relationship can layer Signchat for Teams on top — same product, just governed.

Free / Open source

Most people

$0forever, no account

Every Signchat feature, in your browser, with no install. Bring your own API keys for any usage-billed providers.

Real-time sign-to-voice with the custom ASL classifier
Live captions of the hearing peer's side
Auto and proofread modes — you have the final word
Bridge desktop app for FaceTime, Zoom, Meet, Teams, Discord
Full source on GitHub — fork, audit, self-host

Start a call View source

Enterprise

For organizations

Let's talk

The same product, with SSO, a private deployment, audit trails, and a contractual SLA with a named contact.

Everything in Free
SAML SSO and SCIM provisioning
On-prem or your VPC deployment
Audit logs and admin console
Dedicated support and accessibility-team reviews

Talk to us See enterprise

What "free" actually means

No fine print, no $0-then-paywall.

Signchat doesn't run a relay or a paid tier of the product itself. The only costs that exist are billed by the model vendors — and they're billed to your keys, not to us.

What stays local

Your camera and the ASL classifier

MediaPipe and our ONNX classifier run inside your browser tab. No frames leave the device on the per-turn data path. See the architecture doc for the wire diagram.

What goes to providers

Tokens to Gemini, audio from ElevenLabs

Recognized sign tokens go directly from your browser to Gemini for sentence stitching, and to ElevenLabs for voice. Both are billed by those vendors directly to your own keys — Signchat is not in the middle.

FAQ

Things people usually ask before their first call.

Why is it free?
Because the people who need real-time ASL-to-voice the most are the least likely to have a budget for it. The whole product runs in your browser with no relay — there's no server to pay for and no per-call cost to pass on. The OSS version on GitHub is the same code we ship to enterprise customers.
Do I need an account?
No. Open signchat.org/start, allow camera and mic, and you're in a call. We don't track you, we don't mint a user identifier, and we don't store transcripts.
Where does my video go?
Nowhere we can see. Camera frames are processed locally by MediaPipe and our custom ASL classifier. Only the resulting tokens are sent — directly from your browser — to Gemini for sentence stitching and to ElevenLabs for voice synthesis. Signchat operates no relay.
Can I self-host or fork it?
Yes. The web app, the Bridge desktop companion, the ASL classifier model, and the sign pipeline are all in the monorepo. Bring your own OpenRouter and ElevenLabs keys (or swap them for your own inference endpoint) and you're done.
What about schools and non-profits?
The free app already covers you, but if you want a written letter, sample lesson plans, or rollout help, see Signchat for schools. It's zero-cost.
What does an enterprise contract add?
SSO/SAML, SCIM provisioning, a private deployment in your VPC, audit logs, and a contractual SLA with a named technical contact. The product itself is identical — see Signchat for Teams.

Start a call. It's free.

Open the app, allow camera and mic, share the room link. No account, no install — just talk.

Start a call