"A Survey of Speaker Recognition and Techniques"
Joel Kimble 2003
Abstract
Over the past several decades, one of the most popular areas of study for computer science researchers has been biometrics which are automated methods for recognizing a person based on a physiological characteristic. One of the most natural ways to recognize a person is by the sound of their voice. The goal of voice and speaker recognition is to develop a system that can be used to uniquely identify individuals solely on the sound of their voice. The goal of this I.S. is to implement and compare multiple text-independent speaker recognition systems. To do this it is necessary to employ a process of feature extraction and transformation as well as multiple classification methods and evaluate their effectiveness.