The Evolution of Voice Authentication as a Security Method
New forms of authentication are required to secure online resources. With the rise of cloud computing and the corresponding threat of identity theft, vendors have stepped up their game in this arena: MasterCard is now using selfies for authentication and security vendors are adding new forms of multifactor tokens to their arsenal.
Another productive avenue has been the use of various biometric-based solutions for access management, such as voice authentication factors.
Biometric Authentication Takes Hold
Voice authentication and fingerprint detection both have their advantages and disadvantages when used in authentication. The good news is that you don’t have to carry anything else since you already have your voice or your fingerprints or your eyeballs. Also, using biometric factors can eliminate the need to provide personal information to verify their identity.
There are various vulnerabilities in existing multifactor techniques such as sending text one-time passcodes to a user’s smartphone. One research paper showed how “a man-in-the-browser attack can be elevated to intercept one-time passwords sent to the mobile phone and thus bypass the chain of multifactor mechanisms as used by many financial services.”
The bad news is that biometrics isn’t a binary, yes-or-no decision; voices and fingerprints require statistical analysis and sampling technologies to match with the original end user. Depending on various external elements — such as a noisy room, finger cut or poor scan — the technology could fail to recognize the user properly.
The Challenges to Voice Authentication
Having a database full of voice recordings also increases the risk for a security breach. Imagine if such information were compromised — you can’t simply assign another voice to one of your employees! But if you are trying to deploy a system across the globe, sending hardware tokens could be a hassle, so using recorded voices has some appeal.
Nevertheless, a new series of biometric authentication technologies are gaining favor because they get around some of these issues. One is TeleSign’s Behavior ID, which enables applications to analyze user behavior by tracking several actions in the background, such as examining the cadence of your keystrokes, tracking your mouse usage and other inputs. This is all collected into a score that is used to authenticate users, according to Computerworld.
The notion of biometrics can be taken to the extreme, looking inside your actual computing equipment for unique signatures. Earlier this year, Veriflow announced a way to apply mathematical verification to network authentications, Network Computing reported. It uses predictive mathematical models of your network infrastructure equipment, with the goal to “thwart breaches and outages that can happen as the result of configuration errors or poor policy management.”
Security is only just beginning to explore the potential of voice authentication. While there are drawbacks to these forms of biometric authentication, it is set to expand. Voice methods will likely be a pillar of the technology in the future.