Result Details

Robust Speech Recognition in Unknown Reverberant and Noisy Conditions

HSIAO, R.; MA, J.; HARTMANN, W.; KARAFIÁT, M.; GRÉZL, F.; BURGET, L.; SZŐKE, I.; ČERNOCKÝ, J.; WATANABE, S.; CHEN, Z.; MALLIDI, S.; HEŘMANSKÝ, H.; TSAKALIDIS, S.; SCHWARTZ, R. Robust Speech Recognition in Unknown Reverberant and Noisy Conditions. In Proceedings of 2015 IEEE Automatic Speech Recognition and Understanding Workshop. Scottsdale, Arizona: IEEE Signal Processing Society, 2015. p. 533-538. ISBN: 978-1-4799-7291-3.
Type
conference paper
Language
English
Authors
Hsiao Roger, FIT (FIT)
Ma Jeff
Hartmann William
Karafiát Martin, Ing., Ph.D., DCGM (FIT)
Grézl František, Ing., Ph.D., DCGM (FIT)
Burget Lukáš, doc. Ing., Ph.D., DCGM (FIT)
Szőke Igor, Ing., Ph.D., DCGM (FIT)
Černocký Jan, prof. Dr. Ing., DCGM (FIT)
Watanabe Shinji
Chen Zhuo
Mallidi Sri Harish, FIT (FIT)
Heřmanský Hynek, prof. Ing., Dr. Eng., DCGM (FIT)
Tsakalidis Stavros, FIT (FIT)
Schwartz Richard, FIT (FIT)
Abstract

In this paper, we describe our work on the ASpIRE (AutomaticSpeech recognition In Reverberant Environments)challenge, which aims to assess the robustness of automaticspeech recognition (ASR) systems. The main characteristic ofthe challenge is developing a high-performance system withoutaccess to matched training and development data. Whilethe evaluation data are recorded with far-field microphones innoisy and reverberant rooms, the training data are telephonespeech and close talking. Our approach to this challengeincludes speech enhancement, neural network methods andacoustic model adaptation, We show that these techniquescan successfully alleviate the performance degradation due tonoisy audio and data mismatch.

Keywords

ASpIRE challenge, robust speech recognition

URL
Annotation

In this paper, we describe our work in the ASpIRE challenge. We experiment and evaluate different approaches to tackling the performance degradation due to noise and data mismatch. Our approaches include audio enhancement, data augmentation, unsupervised DNN adaptation, and system combination.

Published
2015
Pages
533–538
Proceedings
Proceedings of 2015 IEEE Automatic Speech Recognition and Understanding Workshop
Conference
The 2015 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2015)
ISBN
978-1-4799-7291-3
Publisher
IEEE Signal Processing Society
Place
Scottsdale, Arizona
DOI
10.1109/ASRU.2015.7404841
UT WoS
000380604800076
EID Scopus
2-s2.0-84964470918
BibTeX 
@inproceedings{BUT120392,
  author="Roger {Hsiao} and Jeff {Ma} and William {Hartmann} and Martin {Karafiát} and František {Grézl} and Lukáš {Burget} and Igor {Szőke} and Jan {Černocký} and Shinji {Watanabe} and Zhuo {Chen} and Sri Harish {Mallidi} and Hynek {Heřmanský} and Stavros {Tsakalidis} and Richard {Schwartz}",
  title="Robust Speech Recognition in Unknown Reverberant and Noisy Conditions",
  booktitle="Proceedings of 2015 IEEE Automatic Speech Recognition and Understanding Workshop",
  year="2015",
  pages="533--538",
  publisher="IEEE Signal Processing Society",
  address="Scottsdale, Arizona",
  doi="10.1109/ASRU.2015.7404841",
  isbn="978-1-4799-7291-3",
  url="https://www.fit.vut.cz/research/publication/11067/"
}
Files
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
Information mining in speech acquired by distant microphones, MV, Bezpečnostní výzkum České republiky 2015-2020, VI20152020025, start: 2015-10-01, end: 2020-09-30, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Department of Computer Graphics and Multimedia (DCGM)
Back to top