Detail publikace

Dealing with Numbers in Grapheme-Based Speech Recognition

Originální název

Dealing with Numbers in Grapheme-Based Speech Recognition

Anglický název

Dealing with Numbers in Grapheme-Based Speech Recognition

Jazyk

en

Originální abstrakt

Grapheme-based speech recognition approach is suitable in situation of low resource languages, where obtaining of pronunciation dictionary is time- and cost-consuming. The paper describes the process of automatic generation of pronunciation dictionaries with emphasis on the expansion of numbers and presents results on GlobalPhone database.

Anglický abstrakt

Grapheme-based speech recognition approach is suitable in situation of low resource languages, where obtaining of pronunciation dictionary is time- and cost-consuming. The paper describes the process of automatic generation of pronunciation dictionaries with emphasis on the expansion of numbers and presents results on GlobalPhone database.

BibTex


@inproceedings{BUT97033,
  author="Miloš {Janda} and Martin {Karafiát} and Jan {Černocký}",
  title="Dealing with Numbers in Grapheme-Based Speech Recognition",
  annote="Grapheme-based speech recognition approach is suitable in situation of low
resource languages, where obtaining of pronunciation dictionary is time- and
cost-consuming. The paper describes the process of automatic generation of
pronunciation dictionaries with emphasis on the expansion of numbers and presents
results on GlobalPhone database.",
  address="Springer Verlag",
  booktitle="Proceedings of 15th International Conference on Text, Speech and Dialogue",
  chapter="97033",
  doi="10.1007/978-3-642-32790-2_53",
  edition="Lecture Notes in Computer Science, 2012, Volume 7499",
  howpublished="print",
  institution="Springer Verlag",
  number="9",
  year="2012",
  month="september",
  pages="438--445",
  publisher="Springer Verlag",
  type="conference paper"
}