React components
biometry-react-components provides the capture half of an integration: camera
and microphone components that handle permissions, countdowns, recording windows
and the challenge phrase, then hand you ordinary File objects.
It has no HTTP layer. Components capture media and emit files; you send those files with the Web SDK or your own backend. That separation is deliberate — your API token never has to reach the browser.
Installation
npm install biometry-react-componentsComponents
| Component | Purpose |
|---|---|
FaceCapture | Still photo of a face |
DocScan | Photo of an identity document |
FaceRecorder | Video with spoken challenge digits, for verification |
VoiceRecorder | Voice enrollment across several takes |
import { FaceRecorder } from 'biometry-react-components';
<FaceRecorder challengeDigits={challenge} // issued by your backend, e.g. "48719352" recordingSeconds={10} onConfirmRecording={({ file }) => upload(file)}/>Upload file. The voice is extracted server-side from the same video that
passed the liveness check, so there is no separate audio file to send — and a
separately supplied one is not accepted, because the voice check is bound to the
verified video.
Use a challenge from your backend
The digits the user reads should be a challenge your backend issues: we
recommend 8 distinct digits from 1–9, from a secure random source, stored against the user’s login
attempt, valid for at most 5 minutes and used once. Pass it to FaceRecorder as
challengeDigits, then have your backend send its own stored copy as
phrase. Record for about 10 seconds (recordingSeconds={10}). See
Liveness challenge for the rules, why 8 digits,
and Node and Go examples.
challengeDigits | Behaviour |
|---|---|
"48719352" or [4, 8, 7, 1, 9, 3, 5, 2] (3.0.0+ for the string form) | Shown and emitted exactly as given. |
Fewer than 7 or more than 12 digits, anything other than 0–9, spaces, an empty string or array | Throws during render. It is never swapped for other digits. Wrap the recorder in an error boundary if the value isn’t trusted. |
null or left out | Deprecated: the component generates digits in the browser and logs a one-time console.warn per component instance. Existing apps keep working, with no replay protection. |
Render the recorder once the challenge has arrived, so the user never starts with generated digits. A new value is applied only while the recorder is idle.
The payload’s phrase (the digits in words) and challenge (the digits as
shown, 3.0.0+) are for display and logging. Don’t send them as the phrase your
backend checks.
A phrase chosen or returned by the device doesn’t stop replays: someone who controls the device, or calls your backend directly, can pair a pre-recorded video of the user with whatever digits that video happens to say, and every lip-reading and voice check agrees. Only digits your backend chose, kept and compares against prove the video was recorded just now.
Each component reports twice: once when media is captured, and again once the person has confirmed it. Use the confirmed callback for anything you upload — the first fires before the user has had a chance to retake.
Hooks
useRecorder, usePermissions and useAudioPermissions back the components and
are exported for building your own capture UI.
Full component and hook reference: the
biometry-react-components README
(npm).