Security System in Speech Recognition?

Speaker recognition is one of the effectively used biometric authentication system that actually identify the speaker on the basis of vocal characteristics. The speaker identification depends on different voice features such as the intensity analysis, voice pitch analysis, voice feature extraction etc. This recognition process is also affected from different factors such as the background noise, instrumentation noise etc. In this paper, noise effective approach is suggested to define an effective speaker recognition process. The robustness of the recognition system is improved with the definition of an integrated layered model.