This HTML5 document contains 46 embedded RDF statements represented using HTML+Microdata notation.

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

Namespace Prefixes

PrefixIRI
dctermshttp://purl.org/dc/terms/
n20http://localhost/temp/predkladatel/
n19http://linked.opendata.cz/resource/domain/vavai/projekt/
n10http://linked.opendata.cz/resource/domain/vavai/riv/tvurce/
n13http://linked.opendata.cz/resource/domain/vavai/subjekt/
n11http://linked.opendata.cz/resource/domain/vavai/vysledek/RIV%2F49777513%3A23520%2F12%3A43915990%21RIV13-MK0-23520___/
n9http://linked.opendata.cz/ontology/domain/vavai/
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/
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/
n12http://linked.opendata.cz/ontology/domain/vavai/riv/duvernostUdaju/
xsdhhttp://www.w3.org/2001/XMLSchema#
n18http://linked.opendata.cz/ontology/domain/vavai/riv/aktivita/
n7http://linked.opendata.cz/ontology/domain/vavai/riv/jazykVysledku/
n17http://linked.opendata.cz/ontology/domain/vavai/riv/obor/
n15http://linked.opendata.cz/ontology/domain/vavai/riv/druhVysledku/
n5http://reference.data.gov.uk/id/gregorian-year/

Statements

Subject Item
n2:RIV%2F49777513%3A23520%2F12%3A43915990%21RIV13-MK0-23520___
rdf:type
n9:Vysledek skos:Concept
rdfs:seeAlso
http://download.springer.com/static/pdf/237/chp%253A10.1007%252F978-3-642-32790-2_58.pdf?auth66=1352100308_6ca3feab5ab1f038650cd3b971c1f53b&ext=.pdf
dcterms:description
The worst problem the adaptation is dealing with is the lack of adaptation data. This work focuses on the feature Maximum Likelihood Linear Regression (fMLLR) adaptation where the number of free parameters to be estimated significantly decreases in comparison with other adaptation methods. However, the number of free parameters of fMLLR transform is still too high to be estimated properly when dealing with extremely small data sets. We described and compared various methods used to avoid this problem, namely the initialization of the fMLLR transform and a linear combination of basis matrices varying in the choice of the basis estimation (eigen decomposition, factor analysis, independent component analysis and maximum likelihood estimation). Initialization methods compensate the absence of the test speaker's data utilizing other suitable data. Methods using linear combination of basis matrices reduce the number of estimated fMLLR parameters to a smaller number of weights to be estimated. Experiments are aimed to compare results of proposed basis and initialization methods. The worst problem the adaptation is dealing with is the lack of adaptation data. This work focuses on the feature Maximum Likelihood Linear Regression (fMLLR) adaptation where the number of free parameters to be estimated significantly decreases in comparison with other adaptation methods. However, the number of free parameters of fMLLR transform is still too high to be estimated properly when dealing with extremely small data sets. We described and compared various methods used to avoid this problem, namely the initialization of the fMLLR transform and a linear combination of basis matrices varying in the choice of the basis estimation (eigen decomposition, factor analysis, independent component analysis and maximum likelihood estimation). Initialization methods compensate the absence of the test speaker's data utilizing other suitable data. Methods using linear combination of basis matrices reduce the number of estimated fMLLR parameters to a smaller number of weights to be estimated. Experiments are aimed to compare results of proposed basis and initialization methods.
dcterms:title
Robust Adaptation Techniques Dealing with Small Amount of Data Robust Adaptation Techniques Dealing with Small Amount of Data
skos:prefLabel
Robust Adaptation Techniques Dealing with Small Amount of Data Robust Adaptation Techniques Dealing with Small Amount of Data
skos:notation
RIV/49777513:23520/12:43915990!RIV13-MK0-23520___
n9:predkladatel
n13:orjk%3A23520
n3:aktivita
n18:P
n3:aktivity
P(DF12P01OVV022)
n3:cisloPeriodika
neuveden
n3:dodaniDat
n5:2013
n3:domaciTvurceVysledku
n10:8612889 n10:3020614 n10:6895972
n3:druhVysledku
n15:J
n3:duvernostUdaju
n12:S
n3:entitaPredkladatele
n11:predkladatel
n3:idSjednocenehoVysledku
165837
n3:idVysledku
RIV/49777513:23520/12:43915990
n3:jazykVysledku
n7:eng
n3:klicovaSlova
ASR, adaptation, fMLLR, robustness, initialization, basis
n3:klicoveSlovo
n8:ASR n8:initialization n8:fMLLR n8:robustness n8:basis n8:adaptation
n3:kodStatuVydavatele
DE - Spolková republika Německo
n3:kontrolniKodProRIV
[7F5B05249D7E]
n3:nazevZdroje
Lecture Notes in Computer Science
n3:obor
n17:JD
n3:pocetDomacichTvurcuVysledku
3
n3:pocetTvurcuVysledku
3
n3:projekt
n19:DF12P01OVV022
n3:rokUplatneniVysledku
n5:2012
n3:svazekPeriodika
7499
n3:tvurceVysledku
Machlica, Lukáš Zajíc, Zbyněk Müller, Luděk
s:issn
0302-9743
s:numberOfPages
8
n20:organizacniJednotka
23520