About: Unsupervised and semi-supervised adaptation of a hybrid speech recognition system

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: Unsupervised and semi-supervised adaptation of a hybrid speech recognition system Goto Sponge NotDistinct Permalink

An Entity of Type : http://linked.opendata.cz/ontology/domain/vavai/Vysledek, within Data Space : linked.opendata.cz associated with source document(s)

Attributes	Values
rdf:type	skos:Concept http://linked.opendata.cz/ontology/domain/vavai/Vysledek
Description	This paper evaluates a recently published method for supervised and unsupervised adaptation of neural networks used in hybrid speech recognition systems. The neural networks used in the field of hybrid speech recognition have certain distinct characteristics that make the usual adaptation methods (such as retraining the neural network) unusable or uneffective. The recently published MELT (Minimum Error Linear Transform) method [1] has been developed to cope with this issue. By providing a way of establishing a link between the intermediate features and the long temporal features, the number of free variables can be reduced significantly and the resulting adaptation parameters can be estimated robustly. The experiments were performed on the WSJCAM0 speech corpus. Contrary to the original paper [1], the experiments were performed using a word recognizer instead of a phoneme recognizer. The experimental results suggest that the MELT method can be used both in an unsupervised as well as a semi-supervised manner and when applied, it leads to significant reduction of word error rate, even for a strong language model. This paper evaluates a recently published method for supervised and unsupervised adaptation of neural networks used in hybrid speech recognition systems. The neural networks used in the field of hybrid speech recognition have certain distinct characteristics that make the usual adaptation methods (such as retraining the neural network) unusable or uneffective. The recently published MELT (Minimum Error Linear Transform) method [1] has been developed to cope with this issue. By providing a way of establishing a link between the intermediate features and the long temporal features, the number of free variables can be reduced significantly and the resulting adaptation parameters can be estimated robustly. The experiments were performed on the WSJCAM0 speech corpus. Contrary to the original paper [1], the experiments were performed using a word recognizer instead of a phoneme recognizer. The experimental results suggest that the MELT method can be used both in an unsupervised as well as a semi-supervised manner and when applied, it leads to significant reduction of word error rate, even for a strong language model. (en)
Title	Unsupervised and semi-supervised adaptation of a hybrid speech recognition system Unsupervised and semi-supervised adaptation of a hybrid speech recognition system (en)
skos:prefLabel	Unsupervised and semi-supervised adaptation of a hybrid speech recognition system Unsupervised and semi-supervised adaptation of a hybrid speech recognition system (en)
skos:notation	RIV/49777513:23520/12:43916037!RIV13-MK0-23520___
http://linked.open...avai/predkladatel	Fakulta aplikovaných věd
http://linked.open...avai/riv/aktivita	P
http://linked.open...avai/riv/aktivity	P(DF12P01OVV022)
http://linked.open...vai/riv/dodaniDat	2013
http://linked.open...aciTvurceVysledku	Müller, Luděk Trmal, Jan Zelinka, Jan
http://linked.open.../riv/druhVysledku	D - Článek ve sborníku
http://linked.open...iv/duvernostUdaju	S - Úplné a pravdivé údaje nepodléhající ochraně podle zvláštních právních předpisů
http://linked.open...titaPredkladatele	Západočeská univerzita v Plzni / Fakulta aplikovaných věd
http://linked.open...dnocenehoVysledku	176074
http://linked.open...ai/riv/idVysledku	RIV/49777513:23520/12:43916037
http://linked.open...riv/jazykVysledku	eng - angličtina
http://linked.open.../riv/klicovaSlova	Neural Networks, Speech Recognition, Speaker Adaptation, MELT (en)
http://linked.open.../riv/klicoveSlovo	Speech Recognition MELT Speaker Adaptation Neural Networks
http://linked.open...ontrolniKodProRIV	[E16F1E68D0A6]
http://linked.open...v/mistoKonaniAkce	Bejing (Peking)
http://linked.open...i/riv/mistoVydani	Beijing (Peking)
http://linked.open...i/riv/nazevZdroje	Proceedings 2012 IEEE 11th International Conference on Signal Processing
http://linked.open...in/vavai/riv/obor	JD
http://linked.open...ichTvurcuVysledku	3 (xsd:int)
http://linked.open...cetTvurcuVysledku	3 (xsd:int)
http://linked.open...vavai/riv/projekt	ASR- and MT-based Access to a Large Archive of Cultural Heritage (AMALACH)
http://linked.open...UplatneniVysledku	2012
http://linked.open...iv/tvurceVysledku	Zelinka, Jan Müller, Luděk Trmal, Jan
http://linked.open...vavai/riv/typAkce	WRD - Světová
http://linked.open.../riv/zahajeniAkce	2012-10-21 (xsd:date)
number of pages	4 (xsd:int)
http://purl.org/ne...btex#hasPublisher	IEEE Press
https://schema.org/isbn	978-1-4673-2194-5
http://localhost/t...ganizacniJednotka	23520
is http://linked.open...avai/riv/vysledek of	Unsupervised and semi-supervised adaptation of a hybrid speech recognition system

Faceted Search & Find service v1.16.118 as of Jun 21 2024

Alternative Linked Data Documents: ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3240 as of Jun 21 2024, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (126 GB total memory, 58 GB memory in use)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2024 OpenLink Software