Publication Details

MGB-3 but system: Low-resource ASR on Egyptian YouTube data

VESELÝ, K.; BASKAR, M.; DIEZ SÁNCHEZ, M.; BENEŠ, K. MGB-3 but system: Low-resource ASR on Egyptian YouTube data. In Proceedings of ASRU 2017. Okinawa: IEEE Signal Processing Society, 2017. p. 368-373. ISBN: 978-1-5090-4788-8.
Czech title
MGB-3 BUT Systém: egyptské rozpoznávání řeči s omezenými zdroji
Type
conference paper
Language
English
Authors
Veselý Karel, Ing., Ph.D. (DCGM)
Baskar Murali Karthick, Ing., Ph.D.
Diez Sánchez Mireia, M.Sc., Ph.D. (DCGM)
Beneš Karel, Ing. (DCGM)
URL
Keywords

MGB-3, ASR adaptation, low-resource ASR, Egyptian Arabic, diarization

Abstract

This paper presents a series of experiments we performed duringour work on the MGB-3 evaluations. We both describethe submitted system, as well as the post-evaluation analysis.Our initial BLSTM-HMM system was trained on 250 hoursof MGB-2 data (Al-Jazeera), it was adapted with 5 hours ofEgyptian data (YouTube). We included such techniques asdiarization, n-gram language model adaptation, speed perturbationof the adaptation data, and the use of all 4 correctreferences. The 4 references were either used for supervisionwith a confusion network, or we included each sentence 4xwith the transcripts from all the annotators. Then, it was alsohelpful to blend the augmented MGB-3 adaptation data with15 hours of MGB-2 data. Although we did not rank with oursingle system among the best teams in the evaluations, we believethat our analysis will be highly interesting not only forthe other MGB-3 challenge participants.

Annotation

This paper presents a series of experiments we performed during our work on the MGB-3 evaluations. We both describe the submitted system, as well as the post-evaluation analysis. Our initial BLSTM-HMM system was trained on 250 hours of MGB-2 data (Al-Jazeera), it was adapted with 5 hours of Egyptian data (YouTube). We included such techniques as diarization, n-gram language model adaptation, speed perturbation of the adaptation data, and the use of all 4 correct references. The 4 references were either used for supervision with a confusion network, or we included each sentence 4x with the transcripts from all the annotators. Then, it was also helpful to blend the augmented MGB-3 adaptation data with 15 hours of MGB-2 data. Although we did not rank with our single system among the best teams in the evaluations, we believe that our analysis will be highly interesting not only for the other MGB-3 challenge participants.

Published
2017
Pages
368–373
Proceedings
Proceedings of ASRU 2017
Conference
2017 IEEE AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING WORKSHOP (ASRU), Okinawa, JP
ISBN
978-1-5090-4788-8
Publisher
IEEE Signal Processing Society
Place
Okinawa
DOI
UT WoS
000426066100051
EID Scopus
BibTeX
@inproceedings{BUT144502,
  author="Karel {Veselý} and Murali Karthick {Baskar} and Mireia {Diez Sánchez} and Karel {Beneš}",
  title="MGB-3 but system: Low-resource ASR on Egyptian YouTube data",
  booktitle="Proceedings of ASRU 2017",
  year="2017",
  pages="368--373",
  publisher="IEEE Signal Processing Society",
  address="Okinawa",
  doi="10.1109/ASRU.2017.8268959",
  isbn="978-1-5090-4788-8",
  url="https://www.fit.vut.cz/research/publication/11595/"
}
Files
Back to top