This HTML5 document contains 47 embedded RDF statements represented using HTML+Microdata notation.

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

Namespace Prefixes

PrefixIRI
n23http://linked.opendata.cz/ontology/domain/vavai/riv/typAkce/
dctermshttp://purl.org/dc/terms/
n19http://purl.org/net/nknouf/ns/bibtex#
n7http://localhost/temp/predkladatel/
n22http://linked.opendata.cz/resource/domain/vavai/projekt/
n9http://linked.opendata.cz/resource/domain/vavai/riv/tvurce/
n21http://linked.opendata.cz/resource/domain/vavai/subjekt/
n12http://linked.opendata.cz/ontology/domain/vavai/
n10https://schema.org/
shttp://schema.org/
rdfshttp://www.w3.org/2000/01/rdf-schema#
skoshttp://www.w3.org/2004/02/skos/core#
n3http://linked.opendata.cz/ontology/domain/vavai/riv/
n24http://bibframe.org/vocab/
n16http://linked.opendata.cz/resource/domain/vavai/vysledek/RIV%2F49777513%3A23520%2F13%3A43920629%21RIV14-TA0-23520___/
n2http://linked.opendata.cz/resource/domain/vavai/vysledek/
rdfhttp://www.w3.org/1999/02/22-rdf-syntax-ns#
n11http://linked.opendata.cz/ontology/domain/vavai/riv/klicoveSlovo/
n20http://linked.opendata.cz/ontology/domain/vavai/riv/duvernostUdaju/
xsdhhttp://www.w3.org/2001/XMLSchema#
n8http://linked.opendata.cz/ontology/domain/vavai/riv/jazykVysledku/
n5http://linked.opendata.cz/ontology/domain/vavai/riv/aktivita/
n14http://linked.opendata.cz/ontology/domain/vavai/riv/druhVysledku/
n13http://linked.opendata.cz/ontology/domain/vavai/riv/obor/
n18http://reference.data.gov.uk/id/gregorian-year/

Statements

Subject Item
n2:RIV%2F49777513%3A23520%2F13%3A43920629%21RIV14-TA0-23520___
rdf:type
n12:Vysledek skos:Concept
rdfs:seeAlso
http://link.springer.com/chapter/10.1007%2F978-3-642-40585-3_8
dcterms:description
Adaptation techniques are necessary in automatic speech recognizers to improve a recognition accuracy. Linear Transformation methods (MLLR or fMLLR) are the most favorite in the case of limited available data. The fMLLR is the feature-space transformation. This is the advantage with contrast to MLLR that transforms the entire acoustic model. The classical fMLLR estimation involves maximization of the likelihood criterion based on individual Gaussian components statistic.We proposed an approach which takes into account the overall likelihood of a HMM state. It estimates the transformation to optimize the ML criterion of HMM directly using gradient descent algorithm. Adaptation techniques are necessary in automatic speech recognizers to improve a recognition accuracy. Linear Transformation methods (MLLR or fMLLR) are the most favorite in the case of limited available data. The fMLLR is the feature-space transformation. This is the advantage with contrast to MLLR that transforms the entire acoustic model. The classical fMLLR estimation involves maximization of the likelihood criterion based on individual Gaussian components statistic.We proposed an approach which takes into account the overall likelihood of a HMM state. It estimates the transformation to optimize the ML criterion of HMM directly using gradient descent algorithm.
dcterms:title
A Direct Criterion Minimization based fMLLR via Gradient Descend A Direct Criterion Minimization based fMLLR via Gradient Descend
skos:prefLabel
A Direct Criterion Minimization based fMLLR via Gradient Descend A Direct Criterion Minimization based fMLLR via Gradient Descend
skos:notation
RIV/49777513:23520/13:43920629!RIV14-TA0-23520___
n12:predkladatel
n21:orjk%3A23520
n3:aktivita
n5:P
n3:aktivity
P(TA01030476)
n3:dodaniDat
n18:2014
n3:domaciTvurceVysledku
n9:2963671 n9:3020614
n3:druhVysledku
n14:D
n3:duvernostUdaju
n20:S
n3:entitaPredkladatele
n16:predkladatel
n3:idSjednocenehoVysledku
58586
n3:idVysledku
RIV/49777513:23520/13:43920629
n3:jazykVysledku
n8:eng
n3:klicovaSlova
ASR, adaptation, fMLLR, gradient descend, Hessian matrix.
n3:klicoveSlovo
n11:Hessian%20matrix. n11:ASR n11:adaptation n11:fMLLR n11:gradient%20descend
n3:kontrolniKodProRIV
[FA98483EBE93]
n3:mistoKonaniAkce
Pilsen, Czech Republic
n3:mistoVydani
Heidelberg
n3:nazevZdroje
Text, Speech and Dialogue
n3:obor
n13:JD
n3:pocetDomacichTvurcuVysledku
2
n3:pocetTvurcuVysledku
2
n3:projekt
n22:TA01030476
n3:rokUplatneniVysledku
n18:2013
n3:tvurceVysledku
Vaněk, Jan Zajíc, Zbyněk
n3:typAkce
n23:WRD
n3:zahajeniAkce
2013-09-01+02:00
s:issn
0302-9743
s:numberOfPages
8
n24:doi
10.1007/978-3-642-40585-3_8
n19:hasPublisher
Springer-Verlag
n10:isbn
978-3-642-40584-6
n7:organizacniJednotka
23520