Angular components

biometry-angular-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.

This is the library the Biometry Console itself uses for its own biometric sign-in, so it stays current with what the API expects.

Installation

npm install biometry-angular-components

The components are standalone: import the ones you need directly into a component’s imports array, with no module to register.

Components

Selector	Purpose	Emits
bio-face-capture	Still photo of a face	File
bio-doc-scan	Photo of an identity document	File
bio-face-recorder	Video with spoken challenge digits, for verification	{ file, audio, phrase }
bio-voice-recorder	Voice enrollment across several takes	{ files, phrases }

Each capture component emits twice: once when the media is captured (capture / recorded), and again once the person has confirmed it (confirmCapture / confirmRecording). Use the confirmed event for anything you upload — the first fires before the user has had a chance to retake.

Recording a verification

bio-face-recorder records video and audio together while the user reads the challenge digits aloud.

<bio-face-recorder
  [challengeDigits]="digits"
  [recordingSeconds]="10"
  (confirmRecording)="onCaptured($event)">
</bio-face-recorder>

Input	Default	Notes
challengeDigits	—	The digits to display, from your challenge endpoint
recordingSeconds	7	Widen it when the challenge is longer than the default
countdownSeconds	—	Lead-in before recording starts
rectWidth / rectHeight	720 / 1280	Capture dimensions

Match the window to the challenge

The recording window is fixed: capture stops when it expires, whether or not the person has finished speaking. A clipped final digit fails the provider’s quality check before any voice comparison happens, so the attempt is rejected without ever being scored. If your challenge is longer than the seven-second default, raise recordingSeconds to match.

Enrolling a voice

bio-voice-recorder collects several takes and emits them together.

<bio-voice-recorder (confirmRecording)="onVoiceCaptured($event)"></bio-voice-recorder>

onVoiceCaptured(payload: { files: File[]; phrases: string[] }) {
  // phrases are positional: phrases[i] was spoken in files[i]
}

It emits once, after every take is confirmed — a partial set is worse than none, so there is nothing to submit until the set is complete. takes defaults to 4, which is what the voice engine builds a print from.

Why several takes

A voiceprint built from fewer distinct recordings carries a single recording’s worth of variation. It scores well against audio captured in the same sitting and poorly on any later attempt, which surfaces as a user who enrolled successfully and then cannot sign in. Let the component collect the full set.

Services

PermissionsService and RecorderService back the components and are exported for building your own capture UI. Reach for them only if the components do not fit — they are the parts most likely to change between releases.