This HTML5 document contains 48 embedded RDF statements represented using HTML+Microdata notation.

The embedded RDF content will be recognized by any processor of HTML5 Microdata.

Namespace Prefixes

PrefixIRI
n14http://linked.opendata.cz/ontology/domain/vavai/riv/typAkce/
dctermshttp://purl.org/dc/terms/
n16http://purl.org/net/nknouf/ns/bibtex#
n9http://localhost/temp/predkladatel/
n5http://linked.opendata.cz/resource/domain/vavai/vysledek/RIV%2F49777513%3A23520%2F09%3A00501738%21RIV10-MSM-23520___/
n12http://linked.opendata.cz/resource/domain/vavai/projekt/
n4http://linked.opendata.cz/resource/domain/vavai/riv/tvurce/
n18http://linked.opendata.cz/ontology/domain/vavai/
n21https://schema.org/
shttp://schema.org/
skoshttp://www.w3.org/2004/02/skos/core#
n3http://linked.opendata.cz/ontology/domain/vavai/riv/
n2http://linked.opendata.cz/resource/domain/vavai/vysledek/
rdfhttp://www.w3.org/1999/02/22-rdf-syntax-ns#
n8http://linked.opendata.cz/ontology/domain/vavai/riv/klicoveSlovo/
n10http://linked.opendata.cz/ontology/domain/vavai/riv/duvernostUdaju/
xsdhhttp://www.w3.org/2001/XMLSchema#
n19http://linked.opendata.cz/ontology/domain/vavai/riv/aktivita/
n15http://linked.opendata.cz/ontology/domain/vavai/riv/jazykVysledku/
n17http://linked.opendata.cz/ontology/domain/vavai/riv/druhVysledku/
n6http://linked.opendata.cz/ontology/domain/vavai/riv/obor/
n11http://reference.data.gov.uk/id/gregorian-year/

Statements

Subject Item
n2:RIV%2F49777513%3A23520%2F09%3A00501738%21RIV10-MSM-23520___
rdf:type
skos:Concept n18:Vysledek
dcterms:description
This paper deals with speaker adaptation techniques well suited for the task of online subtitling. Two methods are briefly discussed, namely MAP adaptation and fMLLR. The main emphasis is laid on the description of improvements involved in the process of adaptation subject to the time requirements. Since the adaptation data are gathered continuously, simple modifications of the accumulated statistics have to be carried out in order to make the adaptation more accurate. Another proposed improvement efficiently employs the combination of fMLLR and MAP. In the case of online adaptation no prior transcriptions of the data are available. They are handled by a recognition system, thus it is suitable to assign a well-applied confidence measure to each of the transcriptions. We have performed experiments focused on the trade-off between the adaptation speed and the amount of adaptation data. We were able to gain a relative reduction of WER 16.2 %. This paper deals with speaker adaptation techniques well suited for the task of online subtitling. Two methods are briefly discussed, namely MAP adaptation and fMLLR. The main emphasis is laid on the description of improvements involved in the process of adaptation subject to the time requirements. Since the adaptation data are gathered continuously, simple modifications of the accumulated statistics have to be carried out in order to make the adaptation more accurate. Another proposed improvement efficiently employs the combination of fMLLR and MAP. In the case of online adaptation no prior transcriptions of the data are available. They are handled by a recognition system, thus it is suitable to assign a well-applied confidence measure to each of the transcriptions. We have performed experiments focused on the trade-off between the adaptation speed and the amount of adaptation data. We were able to gain a relative reduction of WER 16.2 %.
dcterms:title
Fast speaker adaptation in automatic online subtitling Fast speaker adaptation in automatic online subtitling
skos:prefLabel
Fast speaker adaptation in automatic online subtitling Fast speaker adaptation in automatic online subtitling
skos:notation
RIV/49777513:23520/09:00501738!RIV10-MSM-23520___
n3:aktivita
n19:P
n3:aktivity
P(2C06020), P(LC536)
n3:dodaniDat
n11:2010
n3:domaciTvurceVysledku
n4:6579760 n4:2152517 n4:8612889 n4:3020614
n3:druhVysledku
n17:D
n3:duvernostUdaju
n10:S
n3:entitaPredkladatele
n5:predkladatel
n3:idSjednocenehoVysledku
314704
n3:idVysledku
RIV/49777513:23520/09:00501738
n3:jazykVysledku
n15:eng
n3:klicovaSlova
ASR; online subtitling; speaker adaptation; fMLLR; MAP
n3:klicoveSlovo
n8:MAP n8:ASR n8:online%20subtitling n8:fMLLR n8:speaker%20adaptation
n3:kontrolniKodProRIV
[AA59B81CDEDC]
n3:mistoKonaniAkce
Miláno
n3:mistoVydani
Setúbal
n3:nazevZdroje
Proceedings of the International Conference on Signal Processing and Multimedia Applications
n3:obor
n6:JD
n3:pocetDomacichTvurcuVysledku
4
n3:pocetTvurcuVysledku
4
n3:projekt
n12:2C06020 n12:LC536
n3:rokUplatneniVysledku
n11:2009
n3:tvurceVysledku
Psutka, Josef Pražák, Aleš Zajíc, Zbyněk Machlica, Lukáš
n3:typAkce
n14:WRD
n3:zahajeniAkce
2009-07-10+02:00
s:numberOfPages
5
n16:hasPublisher
INSTICC
n21:isbn
978-989-674-007-8
n9:organizacniJednotka
23520