Feature Gaussianization for Speech Recognition

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F02%3APU28124" target="_blank" >RIV/00216305:26220/02:PU28124 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Feature Gaussianization for Speech Recognition

Original language description

In Hidden Markov models, speech data are modeled by Gaussian distributions. In this paper, we propose to Gaussianize the features to better fit to this modeling. A distribution of the data is estimated and a transform function is derived. We test three methods of the transform estimation (global, speaker based, frame based) and report results on the SPINE 2000 task with Sphinx recognizer. We conclude that the proposed method is a cheap way to increase the recognition accuracy.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

JA - Electronics and optoelectronics

OECD FORD branch

—

Project

—

Continuities

Z - Vyzkumny zamer (s odkazem do CEZ)

Publication year

2002

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Conference Proceedings

ISBN

80-227-1700-2

ISSN

—

e-ISSN

—

Number of pages

4

Pages from-to

93-96

Publisher name

Slovak University of Technology in Bratislava

Place of publication

Bratislava,Slovak Republic

Event location

Bratislava

Event date

May 14, 2002

Type of event by nationality

CST - Celostátní akce

UT code for WoS article

—