Detail publikace

Query by Example Search on Speech at Mediaeval 2015

Originální název

Query by Example Search on Speech at Mediaeval 2015

Anglický název

Query by Example Search on Speech at Mediaeval 2015

Jazyk

en

Originální abstrakt

In this paper, we describe the "Query by Example Search on Speech Task" (QUESST), held as part of the MediaEval 2015 evaluation campaign. As in previous years, the proposed task requires performing language-independent audio search in a low resource scenario. This year, the task has been designed to get as close as possible to a practical use case scenario, in which a user would like to retrieve, using speech, utterances containing a given word or short sentence, including those with limited inflectional variations of words, some filler content and/or word re-orderings. We also stressed a mismatch caused by noise and reverberations.

Anglický abstrakt

In this paper, we describe the "Query by Example Search on Speech Task" (QUESST), held as part of the MediaEval 2015 evaluation campaign. As in previous years, the proposed task requires performing language-independent audio search in a low resource scenario. This year, the task has been designed to get as close as possible to a practical use case scenario, in which a user would like to retrieve, using speech, utterances containing a given word or short sentence, including those with limited inflectional variations of words, some filler content and/or word re-orderings. We also stressed a mismatch caused by noise and reverberations.

BibTex


@inproceedings{BUT163808,
  author="Igor {Szőke} and Florian {Metze} and Luis {Rodriguez-Fuentes} and Andi {Buzo} and Xavier {Anguera}",
  title="Query by Example Search on Speech at Mediaeval 2015",
  annote="In this paper, we describe the "Query by Example Search on Speech Task" (QUESST),
held as part of the MediaEval 2015 evaluation campaign. As in previous years, the
proposed task requires performing language-independent audio search in a low
resource scenario. This year, the task has been designed to get as close as
possible to a practical use case scenario, in which a user would like to
retrieve, using speech, utterances containing a given word or short sentence,
including those with limited inflectional variations of words, some filler
content and/or word re-orderings. We also stressed a mismatch caused by noise and
reverberations.",
  address="CEUR-WS.org",
  booktitle="CEUR Workshop Proceedings",
  chapter="163808",
  edition="NEUVEDEN",
  howpublished="online",
  institution="CEUR-WS.org",
  number="1436",
  year="2015",
  month="september",
  pages="1--3",
  publisher="CEUR-WS.org",
  type="conference paper"
}