What are Biometrics?

Biometrics are measurements of an individual's unique physical, behavioral, and biological qualities. In practice, biometrics are used to determine a person's identity from a physical characteristic (e.g., fingerprint, handprint, face, scent, thermal image, or iris pattern), or a behavior pattern (e.g., voice or handwriting signature).

Essentially, all biometric technology works on the same principle. Each user is enrolled by the system and a copy of the enrollment data is stored in a database. When users present themselves for authentication, a new scan/sample is taken and compared with the one(s) stored in the database. If the scan matches, verification is confirmed.

Why Voice?

An article on using biometrics for security in the journal IT Professional (from the Institute of Electrical and Electronics Engineers) shows voice as the best, most acceptable biometric technology.^[1]

1 From “A Practical Guide to Biometric Security Technology,” IEEE Computer Society,
IT Professional – Security, Jan-Feb 2001, Simon Liu and Mark Silverman

People prefer voice for authentication to other biometrics because:

• Voice is intuitive – people are accustomed to recognizing other people by their voice;
• Voice is portable – your voice is always with you and no extra device is needed;
• The VSS system is not language dependent – so it can be used globally;
• Voice is non-threatening – unlike retina or iris scans;
• Voice does not have a forensic connotation – most people associate fingerprints with crime.

Voice analysis technology has been around for years. Applying it used to be tougher than rocket science. Now you can get all the benefits of advanced technology without all the complexity and overhead of managing Gigabytes of voice reference data, dealing with advanced speech technology, and worrying about all the legal issues involved.

Voice Signatures are recordings of randomly prompted phrases that speakers provide for a specific purpose like making a purchase, getting access to sensitive materials, providing authorization, etc.

The Voice Signatures are permanently retained, but only the participating parties have access to the voice signature files. Biometrics

Measuring and analyzing biologic features, or "biometrics" can identify a person. The biometric science known as Speaker Identity Verification (SIV) uses measurements of the unique features of a persons voice to verify their identity. Everyone's voice consists of a unique combination of features that make it distinct from all others. Computer analysis of the digitized voice signal can produce a very specific, quantifiable, and reliable verification of an individual's identity.

There are two distinct types of tasks that can be accomplished by voice analysis:

• Recognition of a person's identity
• Verification of a person's identity

To understand the difference is critical.

Recognition

Recognition requires the comparison of a voice sample to all of the possible matches in order to identify a specific person. This task is huge, and increasingly difficult as the size of the given population grows.

Verification

Verification, on the other hand, uses both a voice sample and an identity-claim. The task is reduced to comparing two known factors: the authenticated identity that is enrolled in the system, and the claimed identity submitted for comparison. Therefore, verification is not dependent on the size of the population, and is a much more practical, less time-consuming task.

Speaker Identity Verification

TradeHarbor's Voice Signature ServiceSM system operates on a verification model. Users are enrolled in the system to establish the voice reference data for a given identity-claim. Our Speaker Identity Verification (SIV) system employs our proprietary speech analysis technology to compare a voice sample from a claimed identity with the voice reference data from the authenticated identity.

Click on an Icon to learn more ...