Semi-tight covariance matrices implementation in MASPER HMM training procedure

The article discuses aspects of incorporating shared linear transformations to implement semi-tight covariance matrices into MASPER HMM training procedure. The concern is on heteroscendic linear discriminative analysis (HLDA) applied to speech features. Next main implementation issues and necessary...

Full description

Saved in:
Bibliographic Details
Main Authors: Kacur, Juraj, Kozicka, Robert, Vargic, Radoslav
Format: Conference Proceeding
Language:English
Subjects:
Conferences
Covariance
Covariance matrices
Electronics
Feature extraction
Hidden Markov models
HLDA
HMM
Image processing
MASPER
semi tight matrices
Speech
Training
Transforms
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The article discuses aspects of incorporating shared linear transformations to implement semi-tight covariance matrices into MASPER HMM training procedure. The concern is on heteroscendic linear discriminative analysis (HLDA) applied to speech features. Next main implementation issues and necessary modifications to the standard MASPER training procedure are introduced. Finally an evaluation of the suggested and implemented changes is accomplished. Apart of that large vocabulary word loop test (LVWLT) has been designed and implemented in order to provide finer system evaluation. All experiments have been executed on Slovak part of MobilDat training database. Achieved results show that incorporation of semi-tight matrices is beneficial in the terms of word error rates (WER) for wide range of test scenarios and models. However, it is for the sake of the increased number of free parameters, sensitivity of complex models to the overtraining and longer training times.
ISSN:2157-8702
DOI:10.1109/IWSSIP.2016.7502777