Voice Biometrics Migration

To continue using Voice Biometrics VDK technologies, set up your models based on your use case:

• Create a Text-Dependent Voice Biometrics Model to authenticate users (e.g., log in with a passphrase).

• Create a Text-Independent Voice Biometrics Model to identify users (e.g., “Who is speaking?”).

What’s New?

• VDK Studio Web now includes live enrollment via VDK Service.

• Export voice prints to test voice biometrics directly in your product.

Key elements

Overview

1. Model Name

The identifier of your biometric model.

2. Model Mode

A biometric model is either text-dependent or text-independent.

3. Voice Prints Enrollment (optional)

You can enroll voice prints either through live enrollment or by using existing audio recordings.

Where to find them?

We’ve included several screenshots with annotations corresponding to the key elements.

VDK Studio Desktop

Desktop version

VDK Studio Online

Online version

You can then use the widget to create Voice Prints as examples in your product. This helps you test and replicate the Voice Biometrics workflow in your product during development.

To enable Voice Biometrics Authentication or Identification, you must integrate a similar enrollment workflow into your product. This allows your users to register their own Voice Prints and dynamically configure the feature.