- Posted by mollatechadmin
- On June 14, 2017
- Security, Voice authentication, Voice biometrics
There is a big initiative or hype in BFSI (banking, financial, securities and insurance) about making authentication simpler and stronger. They have decided to go for voice authentication with consideration that it is secure, simple and cheap.
How does voice biometric authentication work?
User’s recorded voice samples (phrases where user will say his/her Date of birth, Address, Mother’s Name, Credit Limit etc) are feed to the authentication system. In future authentications, say user’s call to support center or web call, where user is asked random questions and the phrases (answers) are captured with the IVR system. Simple call to authentication system is made to match the likelihood of user’s phrases to previously stored samples, meaning DOB stored and “to be compared” sound is matched. And authentication system will return a % score or simple yes/no. Based on that, the operator/system decides that user is same of now.
Where voice authentication fails?
If you see above Voice sample matching is very much like matching your static password. It is just that you will have few static passwords instead of one password. So in essence, voice biometric is very static in nature. That is the failure in voice biometric. This problem is also with thumbprint, retina and facial biometric too.
What will make the voice authentication strong?
You need to consider below –
Add dynamic attribute into your voice biometric authentication. Add One time password that will be narrated by the user so you can do voice biometric along with OTP validation. This is super strong, simple way and two level protection.
Get in touch with us at firstname.lastname@example.org to learn more on how dynamic biometric can work for your business applications.