Detail publikace

Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System

KARAFIÁT, M., BURGET, L., ČERNOCKÝ, J.

Originální název

Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System

Anglický název

Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System

Jazyk

en

Originální abstrakt

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.

Anglický abstrakt

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.

Dokumenty

BibTex


@inproceedings{BUT18264,
  author="Martin {Karafiát} and Lukáš {Burget} and Jan {Černocký}",
  title="Using Smoothed Heteroscedastic Linear Discriminant Analysis in Large Vocabulary Continuous Speech Recognition System",
  annote="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.

", booktitle="2nd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms", chapter="18264", year="2005", month="september", pages="1", type="conference paper" }