Skip to content

Publications

Detail of publication

Citation

Zbynek Zajic : SMOOTHING FACTOR IN DISCRIMINATIVE FEATURE ADAPTATION . Studentská Vědecká Konference, p. 57-58, 2010.

Download PDF

PDF

Abstract

While the ML criterion try to maximize the likelihood of the observation states sequence given the correct transcriptions, the DT criteria reflect the recognition error and try to minimized it. Maximum Mutual Information (MMI) criterion in Yu (2006) is one of the DT possibilities. The main problem consists in the optimization process, where mainly the weak-sense auxiliary function is used. Regrettably, it does not guarantee the convergence of the discriminative criterion. In order to adjust the stability of discriminative criteria a smoothing term is involved. Another criteria are e.g. Minimum Phone Error (MPE) or Minimum Classification Error (MCE).

Detail of publication

Title: SMOOTHING FACTOR IN DISCRIMINATIVE FEATURE ADAPTATION
Author: Zbynek Zajic
Language: English
Date of publication: 1 Jun 2010
Year: 2010
Type of publication: Conference presentations in the Czech Republic
Title of journal or book: Studentská Vědecká Konference
Page: 57 - 58
/ 2011-02-03 12:52:57 /

Keywords

Discriminative Training, Adaptation, Smoothing Factor

BibTeX

@INPROCEEDINGS{ZbynekZajic_2010_SMOOTHINGFACTORIN,
 author = {Zbynek Zajic},
 title = {SMOOTHING FACTOR IN DISCRIMINATIVE FEATURE ADAPTATION},
 year = {2010},
 journal = {Studentsk\'{a} V\v{e}deck\'{a} Konference},
 pages = {57-58},
 url = {http://www.kky.zcu.cz/en/publications/ZbynekZajic_2010_SMOOTHINGFACTORIN},
}