Publication detail

Parallel training of neural networks for speech recognition

VESELÝ, K.

Original Title

Parallel training of neural networks for speech recognition

English Title

Parallel training of neural networks for speech recognition

Type

conference paper

Language

en

Original Abstract

The paper is on Parallel Training of Neural Networks for Speech Recognition. A new parallel-training tool TNet was designed and optimized for multiprocessor computers. The training acceleration rates are reported on a phoneme-state classification task.

English abstract

The paper is on Parallel Training of Neural Networks for Speech Recognition. A new parallel-training tool TNet was designed and optimized for multiprocessor computers. The training acceleration rates are reported on a phoneme-state classification task.

Keywords

neural network, phoneme classification, posterior features, backpropagation training, data parallelization

RIV year

2010

Released

29.04.2010

Publisher

Brno University of Technology

Location

Brno

ISBN

978-80-214-4078-4

Book

Proceedings of the 16th Conference STUDENT EEICT 2010

Edition

Volume 3

Edition number

NEUVEDEN

Pages from

74

Pages to

76

Pages count

3

URL

Documents

BibTex


@inproceedings{BUT35831,
  author="Karel {Veselý}",
  title="Parallel training of neural networks for speech recognition",
  annote="The paper is on Parallel Training of Neural Networks for Speech Recognition.
A new parallel-training tool TNet was designed and optimized for multiprocessor
computers. The training acceleration rates are reported on a phoneme-state
classification task.",
  address="Brno University of Technology",
  booktitle="Proceedings of the 16th Conference STUDENT EEICT 2010",
  chapter="35831",
  edition="Volume 3",
  howpublished="print",
  institution="Brno University of Technology",
  year="2010",
  month="april",
  pages="74--76",
  publisher="Brno University of Technology",
  type="conference paper"
}