Voice Based Authentication using Deep Learning

Abstract

Recognizing a person's individuality with the aid of his/her voice is referred to as Automatic Speaker Recognition (ASR). Speaker popularity falls into the class of biometric safety structures. Biometric is related to human traits or individuality. Biometric verification or realistic authentication is used to apprehend an character via his/her voice character characteristic. Voice biometric consists of behavioral or physiological measurements of the person. Behavioral biometric is accomplished by Voice, Signature, Keystrokes, and Typing and so forth. Whereas physiological biometric consists of iris, face, retina, finger-prints, ear, DNA and so forth. Now a days voice biometric is rising studies region. The market is already in a transition from conventional password authentication procedure to password much less authentication technique. Because of its precise function to become aware of each man or woman with exceptional developments, voice reputation is gaining velocity as authentication method; however, it comes with a few inherent obstacles. To overcome obstacles up to some extent, by using the usage of Deep Learning Approach in present Voice primarily based authentication will help in lowering the quandary of voice based authentication.