Detail publikačního výsledku

TRAP based features for LVCSR of meeting data

KARAFIÁT, M.; GRÉZL, F.; ČERNOCKÝ, J.

Originální název

TRAP based features for LVCSR of meeting data

Anglický název

TRAP based features for LVCSR of meeting data

Druh

Stať ve sborníku mimo WoS a Scopus

Originální abstrakt

This paper describes using temporal patterns (TRAPs) feature extraction in large vocabulary continuous speech recognition (LVCSR) of meeting data. Frequency differentiation and local operators are applied to critical-band speech spectrum. Tests are performed with HMM recognizer on ICSI meetings database. We show that TRAP features in combination with standard ones lead to improvement of word-error rate (WER).

Anglický abstrakt

This paper describes using temporal patterns (TRAPs) feature extraction in large vocabulary continuous speech recognition (LVCSR) of meeting data. Frequency differentiation and local operators are applied to critical-band speech spectrum. Tests are performed with HMM recognizer on ICSI meetings database. We show that TRAP features in combination with standard ones lead to improvement of word-error rate (WER).

Klíčová slova

Speech recognition, TRAP, LVCSR

Klíčová slova v angličtině

Speech recognition, TRAP, LVCSR

Autoři

KARAFIÁT, M.; GRÉZL, F.; ČERNOCKÝ, J.

Vydáno

07.06.2004

Nakladatel

Sunjin Printing Co,

Místo

Jeju Island

Kniha

Proc. 8th International Conference on Spoken Language Processing

ISSN

1225-4111

Svazek

2004

Číslo

10

Strany od

437

Strany do

440

Strany počet

4

URL

http://www.fit.vutbr.cz/~karafiat/publi/2004/karafiat_icslp2004.pdf

BibTex

@inproceedings{BUT17131,
  author="Martin {Karafiát} and František {Grézl} and Jan {Černocký}",
  title="TRAP based features for LVCSR of meeting data",
  booktitle="Proc. 8th International Conference on Spoken Language Processing",
  year="2004",
  volume="2004",
  number="10",
  pages="437--440",
  publisher="Sunjin Printing Co,",
  address="Jeju Island",
  url="http://www.fit.vutbr.cz/~karafiat/publi/2004/karafiat_icslp2004.pdf"
}