Result Details

BUT BABEL System for Spontaneous Cantonese

KARAFIÁT, M.; GRÉZL, F.; HANNEMANN, M.; VESELÝ, K.; ČERNOCKÝ, J. BUT BABEL System for Spontaneous Cantonese. Proceedings of Interspeech 2013. Proceedings of the 14th Annual Conference of the International Speech Communication Association (Interspeech 2013). Lyon: International Speech Communication Association, 2013. no. 8, p. 2589-2593. ISBN: 978-1-62993-443-3. ISSN: 2308-457X.
Type
conference paper
Language
English
Authors
Karafiát Martin, Ing., Ph.D., DCGM (FIT)
Grézl František, Ing., Ph.D., DCGM (FIT)
Hannemann Mirko, Ph.D., DCGM (FIT)
Veselý Karel, Ing., Ph.D., DCGM (FIT)
Černocký Jan, prof. Dr. Ing., DCGM (FIT)
Abstract

This article describes the novel things we have brought to our BABEL Cantonese system include 6-layer Stacked Bottle-Neck features and usingf0 at the input of this NN. We have also investigated into robustnessof SBN training (silence, normalization) and shown anefficient combination with PLP and (again!) F0 features usingRegion-Dependent transforms. Last by not least, a combinationof RDT with another popular adaptation technique (SAT) wasshown beneficial.

Keywords

speech recognition, discriminative training,bottle-neck neural networks, region-dependent transforms

URL
Annotation

This paper presents our work on speech recognition of Cantonese spontaneous telephone conversations. The key-points include feature extraction by 6-layer Stacked Bottle-Neck neural network and using fundamental frequency information at its input. We have also investigated into robustness of SBN training (silence, normalization) and shown an efficient combination with PLP using Region-Dependent transforms. A combination of RDT with another popular adaptation technique (SAT) was shown beneficial. The results are reported on BABEL Cantonese data.

Published
2013
Pages
2589–2593
Journal
Proceedings of the 14th Annual Conference of the International Speech Communication Association (Interspeech 2013)., no. 8, ISSN 2308-457X
Proceedings
Proceedings of Interspeech 2013
Conference
Interspeech Conference
ISBN
978-1-62993-443-3
Publisher
International Speech Communication Association
Place
Lyon
BibTeX 
@inproceedings{BUT103550,
  author="Martin {Karafiát} and František {Grézl} and Mirko {Hannemann} and Karel {Veselý} and Jan {Černocký}",
  title="BUT BABEL System for Spontaneous Cantonese",
  booktitle="Proceedings of Interspeech 2013",
  year="2013",
  journal="Proceedings of the 14th Annual Conference of the International Speech Communication Association (Interspeech 2013).",
  number="8",
  pages="2589--2593",
  publisher="International Speech Communication Association",
  address="Lyon",
  isbn="978-1-62993-443-3",
  issn="2308-457X",
  url="http://www.fit.vutbr.cz/research/groups/speech/publi/2013/karafiat_interspeech2013_IS131522.pdf"
}
Projects
Centrum excelence IT4Innovations, MŠMT, Operační program Výzkum a vývoj pro inovace, ED1.1.00/02.0070, start: 2011-01-01, end: 2015-12-31, completed
IARPA Building Speech Recognition for Keyword Search in a New Language in a Week with Limited Training Data (BABEL) - Babelon, BBN, start: 2012-03-05, end: 2016-11-04, completed
Security-Oriented Research in Information Technology, MŠMT, Institucionální prostředky SR ČR (např. VZ, VC), MSM0021630528, start: 2007-01-01, end: 2013-12-31, running
Speech recognition for low-resource languages, GACR, Postdoktorandské granty, GPP202/12/P604, start: 2012-01-01, end: 2014-12-31, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Department of Computer Graphics and Multimedia (DCGM)
Back to top