In this gives an overview of automatic speaker recognition technology, with an emphasis on textindependent recognition. Speaker recognition has been studied actively for several decades. We give an overview of both the classical and the state-of-the-art methods. We start with the fundamentals of automatic speaker recognition, concerning feature extraction and speaker modeling. Here, describe a Gaussian Mixture Model Universal Background Model (GMM-UBM) speaker identification system. In this

more »

... UBM system, we derive the hypothesized speaker model by adapting the parameters of UBM using the speaker's training speech and a form of Bayesian adaptation. The UBM technique is incorporated into the GMM speaker identification system to reduce the time requirement for recognition significantly.We elaborate advanced computational techniques to address robustness and session variability. In text-dependent system, the words or phrases used for verification are known beforehand and are fixed. In a text-independent system there are no constraints on the words or phrases used during verification system. The recent progress from vectors towards supervectors opens up a new area of exploration and represents a technology trend. clustering method such as K-means .This gives a reduced set of vectors known as codebook .The choice of the clustering method is not as important as optimizing the codebook size.