Computation of TRAP Speech Features Using MATLAB and QuickNet Toolkit

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F12%3A00195401" target="_blank" >RIV/68407700:21230/12:00195401 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Computation of TRAP Speech Features Using MATLAB and QuickNet Toolkit

Popis výsledku v původním jazyce

This paper describes the computation of TRAP speech features in MATLAB environment. Two basic approaches of TRAP features are discussed, commonly with its implementation as MATLAB functions. As TRAP features are based on computation of aposterior probabilities of particular phones using artifficial nerual networks, the possible usage of available MATLAB Neural Network Toolbox is presented. The classification phase of ANN can be impemented without serious problems, however, the training of used ANN is out of computational abilities of MATLAB Neural Network Toolbox. To overcome this problem, the usage of external QuickNet toolkit (realizing the implementation of ANNs) is introduced commonly with possible combinations with MATLAB functions. Finally, basicexperiments with TRAP features computation working as simple phone classifier are presented.

Název v anglickém jazyce

Computation of TRAP Speech Features Using MATLAB and QuickNet Toolkit

Popis výsledku anglicky

This paper describes the computation of TRAP speech features in MATLAB environment. Two basic approaches of TRAP features are discussed, commonly with its implementation as MATLAB functions. As TRAP features are based on computation of aposterior probabilities of particular phones using artifficial nerual networks, the possible usage of available MATLAB Neural Network Toolbox is presented. The classification phase of ANN can be impemented without serious problems, however, the training of used ANN is out of computational abilities of MATLAB Neural Network Toolbox. To overcome this problem, the usage of external QuickNet toolkit (realizing the implementation of ANNs) is introduced commonly with possible combinations with MATLAB functions. Finally, basicexperiments with TRAP features computation working as simple phone classifier are presented.

Druh

D - Stať ve sborníku

CEP obor

JA - Elektronika a optoelektronika, elektrotechnika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2012

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

POSTER 2012 - 16th International Student Conference on Electrical Engineering

ISBN

978-80-01-05043-9

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

1-4

Název nakladatele

Czech Technical University in Prague

Místo vydání

Praha

Místo konání akce

Prague

Datum konání akce

17. 5. 2012

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

—