Detail publikace

Phonotactic Language Identification using High Quality Phoneme Recognition

Pavel Matejka, Petr Schwarz, Jan Cernocky, Pavel Chytil

Originální název

Phonotactic Language Identification using High Quality Phoneme Recognition

Anglický název

Phonotactic Language Identification using High Quality Phoneme Recognition

Jazyk

en

Originální abstrakt

Phoneme Recognizers followed by Language Modeling (PRLM) have consistently yielded top performance in language identification (LID) task. Parallel ordering of PRLMs (PPRLM) improves performance even more. Since tokenizer is the most important part of LID system the high quality phoneme recognizer is employed. Two different multilingual databases for training phoneme recognizers are compared and the amount of sufficient training data is studied. Reported results are on data from NIST 2003 LID evaluation. Our four PRLM systems have Equal Error Rate (EER) of 2.4% on 12 languages task. This result compares favorably to the best known result from this task.

Anglický abstrakt

Phoneme Recognizers followed by Language Modeling (PRLM) have consistently yielded top performance in language identification (LID) task. Parallel ordering of PRLMs (PPRLM) improves performance even more. Since tokenizer is the most important part of LID system the high quality phoneme recognizer is employed. Two different multilingual databases for training phoneme recognizers are compared and the amount of sufficient training data is studied. Reported results are on data from NIST 2003 LID evaluation. Our four PRLM systems have Equal Error Rate (EER) of 2.4% on 12 languages task. This result compares favorably to the best known result from this task.

Dokumenty

BibTex


@inproceedings{BUT17756,
  author="Pavel {Matějka} and Petr {Schwarz} and Jan {Černocký} and Pavel {Chytil}",
  title="Phonotactic Language Identification using High Quality Phoneme Recognition",
  annote="Phoneme Recognizers followed by Language Modeling (PRLM) have
consistently yielded top performance in language identification
(LID) task. Parallel ordering of PRLMs (PPRLM) improves
performance even more. 
Since tokenizer is the most important part of
LID system the high quality phoneme recognizer is employed. Two
different multilingual databases for training phoneme recognizers are
compared and the amount of sufficient training data is studied.
Reported results are on data from NIST
2003 LID evaluation. Our four PRLM systems have Equal Error Rate
(EER) of 2.4% on 12 languages task. This result compares
favorably to the best known result from this task.
",
  booktitle="submitted to Eurospeech 2005",
  chapter="17756",
  year="2005",
  month="april",
  pages="1",
  type="conference paper"
}