Result Details

Differences between context dependent and context independent Hidden Markov Models for recognition of Czech

KARAFIÁT, M.; ČERNOCKÝ, J. Differences between context dependent and context independent Hidden Markov Models for recognition of Czech. Proc. of 8th student conference STUDENT EEICT 2002. Brno: Faculty of Electrical Engineering TUB, 2002. p. 328-332. ISBN: 80-214-2116-9.
Type
conference paper
Language
English
Authors
Karafiát Martin, Ing., Ph.D., FIT (FIT)
Černocký Jan, prof. Dr. Ing.
Abstract

Differences between context dependent and context independent Hidden Markov Models for recognition of Czech

Keywords

speech recognition, Hidden Markov models, context modeling, Czech speech recognition

Annotation

In speech recognition, context dependent (CD) models generally provide
better recognition accuracy than context independent (CI) ones.  ThIn this
paper, we describe the tests of CD models on the recognition of Czech
isolated words. The models were trained on SpeechDat-E database. The
results with different training setups are presented, and some problems
are discussed

Published
2002
Pages
328–332
Proceedings
Proc. of 8th student conference STUDENT EEICT 2002
Conference
Student EEICT 2002
ISBN
80-214-2116-9
Publisher
Faculty of Electrical Engineering TUB
Place
Brno
BibTeX 
@inproceedings{BUT10282,
  author="Martin {Karafiát} and Jan {Černocký}",
  title="Differences between context dependent and context independent Hidden Markov Models for recognition of Czech",
  booktitle="Proc. of 8th student conference STUDENT EEICT 2002",
  year="2002",
  pages="328--332",
  publisher="Faculty of Electrical Engineering TUB",
  address="Brno",
  isbn="80-214-2116-9"
}
Projects
Voice technologies for support of information society, GACR, Standardní projekty, GA102/02/0124, start: 2002-01-01, end: 2004-12-31, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Faculty of Information Technology (FIT)
Back to top