Detail publikace

BUT2012 Approaches for Spoken Web Search - MediaEval 2012

Originální název

BUT2012 Approaches for Spoken Web Search - MediaEval 2012

Anglický název

BUT2012 Approaches for Spoken Web Search - MediaEval 2012

Jazyk

en

Originální abstrakt

We submitted two approaches as the required runs: Acoustic Keyword Spotting as the primary one (AKWS) and Dynamic Time Wrapping as the secondary one (DTW) for the Spoken Web Search task. We aimed at building a simple phone based language-dependent system. We experimented with universal context bottle-neck neural network classifier with 3-state phone posterior features or bottle-neck features.

Anglický abstrakt

We submitted two approaches as the required runs: Acoustic Keyword Spotting as the primary one (AKWS) and Dynamic Time Wrapping as the secondary one (DTW) for the Spoken Web Search task. We aimed at building a simple phone based language-dependent system. We experimented with universal context bottle-neck neural network classifier with 3-state phone posterior features or bottle-neck features.

BibTex


@inproceedings{BUT97056,
  author="Igor {Szőke} and Michal {Fapšo} and Karel {Veselý}",
  title="BUT2012 Approaches for Spoken Web Search - MediaEval 2012",
  annote="We submitted two approaches as the required runs: Acoustic Keyword Spotting as
the primary one (AKWS) and Dynamic Time Wrapping as the secondary one (DTW) for
the Spoken Web Search task. We aimed at building a simple phone based
language-dependent system. We experimented with universal context bottle-neck
neural network classifier with 3-state phone posterior features or bottle-neck
features.",
  address="CEUR-WS.org",
  booktitle="Working Notes Proceedings of the MediaEval 2012 Workshop",
  chapter="97056",
  edition="NEUVEDEN",
  howpublished="online",
  institution="CEUR-WS.org",
  number="927",
  year="2012",
  month="october",
  pages="1--2",
  publisher="CEUR-WS.org",
  type="conference paper"
}