Detail publikace

BUT Neural Network Features for Spontaneous Vietnamese in BABEL

Originální název

BUT Neural Network Features for Spontaneous Vietnamese in BABEL

Anglický název

BUT Neural Network Features for Spontaneous Vietnamese in BABEL

Jazyk

en

Originální abstrakt

The paper deals with multiple facets of NN feature extraction training. Not surprisingly, we found that data preparation is crucial for the success of NN training. In case we dispose of data from other (well represented) languages, we should go for it as we have shown that multilingual fine-tuning outperforms unsupervised training.

Anglický abstrakt

The paper deals with multiple facets of NN feature extraction training. Not surprisingly, we found that data preparation is crucial for the success of NN training. In case we dispose of data from other (well represented) languages, we should go for it as we have shown that multilingual fine-tuning outperforms unsupervised training.

BibTex


@inproceedings{BUT111541,
  author="Martin {Karafiát} and František {Grézl} and Mirko {Hannemann} and Jan {Černocký}",
  title="BUT Neural Network Features for Spontaneous Vietnamese in BABEL",
  annote="The paper deals with multiple facets of NN feature extraction training. Not
surprisingly, we found that data preparation is crucial for the success of NN
training. In case we dispose of data from other (well represented) languages, we
should go for it as we have shown that multilingual fine-tuning outperforms
unsupervised training.",
  address="IEEE Signal Processing Society",
  booktitle="Proceedings of ICASSP 2014",
  chapter="111541",
  doi="10.1109/ICASSP.2014.6854679",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Signal Processing Society",
  year="2014",
  month="may",
  pages="5659--5663",
  publisher="IEEE Signal Processing Society",
  type="conference paper"
}