Discriminative Feature Extraction based on PCA Gaussian Mixture Models
نویسنده
چکیده
A discriminative feature extraction based on Principal component analysis (PCA) and Gaussian mixture models is presented to increase the discrimative capability of modified two dimentional root cepstrum analysis (MTDRC). The exprimental results show that F-ratio tests indicate better separability of phonemes by using discriminative feature extraction than MTDRC. Key–Words: Feature extraction,PCA , Gaussian mixture model,MTDRC
منابع مشابه
In search of target class definition in tandem feature extraction
In the tandem feature extraction scheme a Multi-Layer Perceptron (MLP) with softmax output layer is discriminatively trained to estimate context independent phoneme posterior probabilities on a labeled database. The outputs of the MLP after nonlinear transformation and Principal ComponentAnalysis (PCA) are used as features in a Gaussian Mixture Model (GMM) based recognizer. The baseline tandem ...
متن کاملLog-Linear Mixture Models for Patch-Based Object Recognition
In this work, we present a new log-linear model for object recognition in images using local features. Object recognition is a building block for applications in many fields of image understanding. To recognise objects in images, commonly local features and statistical models are used in a Bayesian framework. Based on an approach that uses Gaussian mixtures to model object appearance we present...
متن کاملImproved Algorithm for Fully-automated Neural Spike Sorting based on Projection Pursuit and Gaussian Mixture Model
For the analysis of multiunit extracellular neural signals as multiple spike trains, neural spike sorting is essential. Existing algorithms for the spike sorting have been unsatisfactory when the signal-to-noise ratio (SNR) is low, especially for implementation of fully-automated systems. We present a novel method that shows satisfactory performance even under low SNR, and compare its performan...
متن کاملConnectionist Feature Extraction for Conventional Hmm Systems
Hidden Markov model speech recognition systems typically use Gaussian mixture models to estimate the distributions of decorrelated acoustic feature vectors that correspond to individual subword units. By contrast, hybrid connectionist-HMM systems use discriminatively-trained neural networks to estimate the probability distribution among subword units given the acoustic observations. In this wor...
متن کاملTandem connectionist feature extraction for conventional HMM systems
Hidden Markov model speech recognition systems typically use Gaussian mixture models to estimate the distributions of decorrelated acoustic feature vectors that correspond to individual subword units. By contrast, hybrid connectionist-HMM systems use discriminatively-trained neural networks to estimate the probability distribution among subword units given the acoustic observations. In this wor...
متن کامل