Covariance updates for discriminative training by constrained line search
نویسندگان
چکیده
We investigate the recent Constrained Line Search algorithm for discriminative training of HMMs and propose an alternative formula for variance update. We compare the method to standard techniques on a phone recognition task.
منابع مشابه
An ellipsoid constrained quadratic programming perspective to discriminative training of HMMs
In this paper, we reformulate the optimization in discriminative training (DT) of HMMs as an ellipsoid constrained quadratic programming (ECQP) problem, where a second order of the nonlinear space is adopted. We show that the unique optimal solution of ECQP can be obtained by an efficient line search and no relaxation is needed as in a general quadratically constrained quadratic programming (QC...
متن کاملFull Covariance Modelling for Speech Recognition
HMM-based systems for Automatic Speech Recognition typically model the acoustic features using mixtures of multivariate Gaussians. In this thesis, we consider the problem of learning a suitable covariance matrix for each Gaussian. A variety of schemes have been proposed for controlling the number of covariance parameters per Gaussian, and studies have shown that in general, the greater the numb...
متن کاملPhishing website detection using weighted feature line embedding
The aim of phishing is tracing the users' s private information without their permission by designing a new website which mimics the trusted website. The specialists of information technology do not agree on a unique definition for the discriminative features that characterizes the phishing websites. Therefore, the number of reliable training samples in phishing detection problems is limited. M...
متن کاملDiscriminatively trained sparse inverse covariance matrices for low resource acoustic modeling
We propose a method to discriminatively train acoustic models with sparse inverse covariance (precision) matrices in order to improve the model robustness when training data is insufficient. Acoustic models with sparse inverse covariance matrices were previously proposed to address the problem of over-fitting when training data is inadequate. Since many of the entries of the inverse covariance ...
متن کاملImproving Accuracy of DGPS Correction Prediction in Position Domain using Radial Basis Function Neural Network Trained by PSO Algorithm
Differential Global Positioning System (DGPS) provides differential corrections for a GPS receiver in order to improve the navigation solution accuracy. DGPS position signals are accurate, but very slow updates. Improving DGPS corrections prediction accuracy has received considerable attention in past decades. In this research work, the Neural Network (NN) based on the Gaussian Radial Basis Fun...
متن کامل