Voice Biometrics: Advantages and Disadvantages

Pros:
Less prone to compromise: Contrary to PIN/ Passwords storages compromised and stolen and replayed, the voice prints can not be replayed. Thus a compromised voiceprint is unusable for account access.
Anti reversing: A voiceprint is a hashed string of numbers and characters that represent how a specific individual’s voice rates on the myriad of characteristics being measured Also, it’s not possible to reverse engineer it to recover someone’s voice.
Proactive detection of known fraudsters: Each time a fraudster speaks within an IVR or to a contact center agent, the fraudster leaves his/her voiceprint in the same way that our fingers leave fingerprints when we touch an object. This enables an organization to create and store voiceprints of known fraudsters.
Non guessable: A voice is unique to the individual. It can’t be guessed unlike PINs or passwords.
Cost effective: The cost of implementation is low because there is no special hardware required. A simple telephone or microphone is all that a user needs to authenticate using her voice. Other methods of biometric authentication like fingerprinting and retinal scans require special devices.
Ease of usability: Most important to the future of voice biometrics is that it is the only biometric that allows users to authenticate remotely.
Quick enrollment: It is quick to enroll in a voice authentication system. The user is asked to speak a certain set of words or phrases, or to speak for a certain length of time.
Fast: Authentication is very fast; it can be completed in 0.5 seconds.
Less storage size: Another advantage is that the storage size of the voiceprint is small.

Cons:
Relatively low security: The biggest disadvantage is the replay attack. Hackers might attempt to gain unauthorized access to a voice authenticated system by playing back a pre-recorded voice sample from an authorized user. Need to implement proper anti-replay/ spoofing measures.

Low accurate: Person voice change, the difference in speaking instruments etc can affect the recognition. Compared to that other forms of biometrics such as retinal or fingerprint scans are more accurate and less prone to change.

Comments

Popular posts from this blog

SQL Injection in search field

File Upload through Null Byte Injection