Biometric Technology

Sensory’s proprietary speaker verification, face recognition, and biometric fusion algorithms leverage Sensory’s deep strength in speech processing, computer vision, and machine learning. The unique combination of face and voice recognition provides maximum security, yet remains fast, convenient and easy to use, while ensuring the highest verification rates for the user.

Biometrics aren’t just beneficial for their security—they’re also more convenient than other methods. Not all biometric solutions are created equal, and some have been known to accept false positives (a phenomenon called “spoofing”). Sensory’s novel approach utilizing passive face liveness, active voice liveness, or a combination of the two leverages a deep learning model that nearly eliminates spoofs from fraudsters using 3D masks, photos, video recordings, and more. This may be enabled without active participation (passive face liveness) from the user, so it ensures the ultimate in security, accuracy, and convenience.

Image Alt Text

Key Features

Biometrics refer to measurable physical and behavioral characteristics that can be used to verify a person’s identity. Biometric software may include face recognition technology, voice biometry, fingerprint scanning, retina scans, and other methodologies. Since such qualifiers are unique to individuals, face and voice biometrics solutions offer greater security than traditional security measures such as profiles, passwords or pass-phrases, PIN numbers, and personal information.

Authenticate users via face, voice or a fusion of face and voice.
Authenticate users via face, voice, or a fusion of face and voice
Face Liveness detection protects against spoof attacks
Face liveness detection protects against spoof attacks
FIDO UAF Certified for Android and iOS
FIDO UAF certified for Android
Automatic learning of voices over time allows apps to become personalized
Automatic learning of voices over time allows apps to become personalized

TrulySecure

TrulySecure is Sensory’s highly-flexible face and voice authentication solution, offering higher security and greater ease-of-use than traditional authentication methods. TrulySecure allows app developers or OEMs to quickly add face and voice biometrics to any mobile, desktop application, or device.

Image Alt Text)

Flexible Face and Voice Authentication Software

With combination of both face and voice biometrics in one SDK, developers can leverage both biometrics to improve recognition rates and address real-world challenges such as face masks, recordings of a voice password, very dark environments, or noisy environments. When faces are partially obstructed, the voice biometric provides the extra confidence needed to improve speaker recognition, enabling secure access. Alternatively, when used in a noisy background with poor acoustics, such as a restaurant or sporting event, the voice biometric is enhanced with a facial recognition score. Developers can leverage both face and voice biometrics in real-time to provide a seamless, contact-free experience to the end-user. Multiple enrollments let users set alternate modalities, for example, a “mask mode” where the user will be prompted for voice verification when a mask is detected. Users can even enroll a voice password with their masks on and off to ensure voice recognition works under all conditions.

Request a product brief to find out more about TrulySecure.

Image Alt Text)

How it works

 TrulySecure, Sensory’s embedded facial verification AI, attempts to validate a person based on the matching of a facial image with a locally stored, and encrypted enrolled image – by either accepting or rejecting the identity claim (one-to-one matching). This model is most commonly used to grant access to a device as seen in mobile phones, tablets, and PCs. Contrary to popular belief, embedded biometric authentication systems don’t actually store complete images of human faces in a database. Rather, they translate the image to an encrypted code and will match the user’s face to this code on the next facial recognition attempt.

Voice verification uses a similar process. Voice biometrics require you to create a voiceprint for enrollment. This will be done either by repeating pass-phrases or by providing an extended sample of your speech to the microphone. Voice biometric solutions are typically used for security purposes, and some scenarios may combine voice biometrics with face recognition during the authentication process for enhanced security in industries like law enforcement or financial services.

Image Alt Text)

Stay Connected

Get the latest speech recognition and artificial intelligence resources and industry updates.