Doc. Ing. Lukáš Burget, Ph.D.

KARAFIÁT Martin, BURGET Lukáš and ČERNOCKÝ Jan. Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System. In: 2nd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms. Edinbourgh, Scotland: University of Edinburgh, 2005, p. 8.
Publication language:english
Original title:Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System
Title (cs):Užití SHLDA pro rozpoznávání řeči s velkým slovníkem
Pages:8
Proceedings:2nd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms
Conference:2nd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms
Series:tento článek nebyl zařazen mezi Revised Selected Papers, nevyšel v LNCS 3869
Place:Edinbourgh, Scotland, GB
Year:2005
Publisher:University of Edinburgh
URL:https://www.fit.vutbr.cz/~karafiat/publi/2005/karafiat_mlmi2005.pdf [PDF]
Keywords
speech recognition, LVCSR, HLDA, feature transform, dimensionality reduction
Annotation
In this work, we verify that SHLDA can be advantageously used also for Large Vocabulary
Continuous Speech Recognition.
Abstract
In the state-of-the-art speech recognition systems, Heteroscedastic Linear Discriminant Analysis (HLDA)
is becoming popular technique allowing for feature decorrelation and dimensionality reduction.
However, HLDA relies on statistics, which may not be reliably estimated when only limited amount of
training data is available. Recently, Smoothed HLDA (SHLDA) was proposed
as a robust modification of
HLDA. Previously, SHLDA was successfully used for feature combination in
small vocabulary recognition
experiments. In this work, we verify that SHLDA can be advantageously used also for Large Vocabulary
Continuous Speech Recognition.

BibTeX:
@INPROCEEDINGS{
   author = {Martin Karafi{\'{a}}t and Luk{\'{a}}{\v{s}} Burget and Jan
	{\v{C}}ernock{\'{y}}},
   title = {Using Smoothed Heteroscedastic Linear Discriminant Analysis
	in Large  Vocabulary Continuous Speech Recognition System},
   pages = {8},
   booktitle = {2nd Joint Workshop on Multimodal Interaction and Related
	Machine Learning Algorithms},
   series = {tento {\v{c}}l{\'{a}}nek nebyl za{\v{r}}azen mezi Revised
	Selected Papers, nevy{\v{s}}el v LNCS 3869},
   year = {2005},
   location = {Edinbourgh, Scotland, GB},
   publisher = {University of Edinburgh},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=7929}
}

Your IPv4 address: 54.81.105.205
Switch to IPv6 connection

DNSSEC [dnssec]