Training of Speaker-Clustered Acoustic Models for Use in Real-Time Recognizers

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F09%3A00501736" target="_blank" >RIV/49777513:23520/09:00501736 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Training of Speaker-Clustered Acoustic Models for Use in Real-Time Recognizers
Original language description
The paper deals with training of speaker-clustered acoustic models. Various training techniques - Maximum Likelihood, Discriminative Training and two adaptation based on the MAP and Discriminative MAP were tested in order to minimize an impact of speakerchanges to the correct function of the recognizer when a response of the automatic cluster detector is delayed or incorrect. Such situation is very frequent e.g. in on-line subtitling of TV discussions (Parliament meetings). In our experiments the bestcluster-dependent training procedure was discriminative adaptation which provided the best trade-off between recognition results with correct and non-correct cluster detector information.
Czech name
—
Czech description
—

Project
<a href="/en/project/2C06020" target="_blank" >2C06020: Elimination of the language barriers faced by the handicapped watchers of the Czech Television</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2009
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of the International Conference on Signal Processing and Multimedia Application
ISBN
978-989-674-007-8
ISSN
—
e-ISSN
—
Number of pages
5
Pages from-to
—
Publisher name
INSTICC
Place of publication
Setúbal
Event location
Miláno
Event date
Jul 10, 2009
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)