Voice recognition is the ability to recognize the vocal print of an individual and compare it against a series of other prints to confirm the voice identity of the customer. It is not to be confused with speech recognition which transfers the spoken work into text.
Voice biometrics can be passive, where the user can say anything whilst in a call to a customer service agent and the voice is then matched during the call. Or it can be active where the caller is asked to recite a phrase or a set or randomly generated numbers or keywords.
In order to save the need for a lengthy conversation with an agent, the customer can self serve and create a new password if forgotten by using their voice as a confirmation of their identity. By simply repeating a set of randomly generated numbers, the customer will be able to reset their own password.
Agents can serve their customers quicker as there will not be any need for a lengthy investigation. The caller will have already verified themselves via voice recognition before they are put through to the agent. Alternatively using passive recognition the voice will be recognized while in a conversation with the agent and verified naturally.
The customer can use their voice on all devices, so they can log in on a PC and then use their voice to reset their password on their mobile.