Přístupnostní navigace
E-přihláška
Vyhledávání Vyhledat Zavřít
Detail publikace
MATĚJKA, P., SCHWARZ, P., ČERNOCKÝ, J., CHYTIL, P.
Originální název
Phonotactic Language Identification using High Quality Phoneme Recognition
Typ
článek ve sborníku ve WoS nebo Scopus
Jazyk
angličtina
Originální abstrakt
Phoneme Recognizers followed by Language Modeling (PRLM) haveconsistently yielded top performance in language identification(LID) task. Parallel ordering of PRLMs (PPRLM) improvesperformance even more. Since tokenizer is the most important part ofLID system the high quality phoneme recognizer is employed. Twodifferent multilingual databases for training phoneme recognizers arecompared and the amount of sufficient training data is studied.Reported results are on data from NIST2003 LID evaluation. Our four PRLM systems have Equal Error Rate(EER) of 2.4\% on 12 languages task. This result comparesfavorably to the best known result from this task.
Klíčová slova
Language identificaion, phoneme recognition
Autoři
Rok RIV
2005
Vydáno
9. 9. 2005
Nakladatel
International Speech Communication Association
Místo
Lisbon
ISSN
1018-4074
Periodikum
European Conference EUROSPEECH
Ročník
Číslo
9
Stát
Švýcarská konfederace
Strany od
2237
Strany do
2240
Strany počet
4
URL
http://www.fit.vutbr.cz/~matejkap/publi/2005/eurospeech2005.pdf
BibTex
@inproceedings{BUT21478, author="Pavel {Matějka} and Petr {Schwarz} and Jan {Černocký} and Pavel {Chytil}", title="Phonotactic Language Identification using High Quality Phoneme Recognition", booktitle="Interspeech'2005 - Eurospeech - 9th European Conference on Speech Communication and Technology", year="2005", journal="European Conference EUROSPEECH", volume="2005", number="9", pages="2237--2240", publisher="International Speech Communication Association", address="Lisbon", issn="1018-4074", url="http://www.fit.vutbr.cz/~matejkap/publi/2005/eurospeech2005.pdf" }