Text Independent Speaker Identification Using Imfcc Integrated With Ica
P. Suryakumari · IOSR Journal of Electronics and Communication Engineering · 2013
Over the years, more research work has been reported in literature regarding text independent speaker identification using MFC coefficients.MFCC is one of the best methods modeled on human auditory system.Murali et al (2011) [1] has developed a Text independent speaker identification using MFC coefficients which follows Generalized Gaussian mixer model.MFCC, because of its filter bank structure it captures the characteristics of information more effectively in lower frequency region than higher region, because of this, valuable information in high frequency region may be lost.In this paper we rectify the above problem by retrieving the information in high frequency region by inverting the Mel bank structure.The dimensionality and dependency of above features were reduced by integrating with ICA.Here Text Independent Speaker Identification system is developed by using Generalized Gaussian Mixer Model .By the experimentation, it was observed that this model outperforms the earlier existing models.