Publication detail

Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data

KARAFIÁT, M. SZŐKE, I. ČERNOCKÝ, J.

Original Title

Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data

English Title

Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data

Type

conference paper

Language

en

Original Abstract

This paper is on using the gradient descent optimization for acoustics training from heterogeneous data. We study the use of heterogeneous data for training of acoustic models.

English abstract

This paper is on using the gradient descent optimization for acoustics training from heterogeneous data. We study the use of heterogeneous data for training of acoustic models.

Keywords

speech, acoustic models, heterogeneous data, HLDA system, gradient descent training, robustness

RIV year

2010

Released

06.09.2010

Publisher

Springer Verlag

Location

Brno

ISBN

978-3-642-15759-2

Book

Proc. Text, Speech and Dialog 2010

Edition

LNAI 6231

Edition number

NEUVEDEN

Pages from

322

Pages to

329

Pages count

8

URL

Documents

BibTex


@inproceedings{BUT34926,
  author="Martin {Karafiát} and Igor {Szőke} and Jan {Černocký}",
  title="Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data",
  annote="This paper is on using the gradient descent optimization for acoustics training
from heterogeneous data. We study the use of heterogeneous data for training of
acoustic models.",
  address="Springer Verlag",
  booktitle="Proc. Text, Speech and Dialog 2010",
  chapter="34926",
  edition="LNAI 6231",
  howpublished="print",
  institution="Springer Verlag",
  journal="Lecture Notes in Computer Science (IF 0,513)",
  number="9",
  year="2010",
  month="september",
  pages="322--329",
  publisher="Springer Verlag",
  type="conference paper"
}