Implementation of Cepstral Voice Activity Detector

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00315575" target="_blank" >RIV/68407700:21230/17:00315575 - isvavai.cz</a>

Výsledek na webu

<a href="http://radio.feld.cvut.cz/conf/poster/proceedings/Poster_2017/Section_IC/IC_070_Kosek.pdf" target="_blank" >http://radio.feld.cvut.cz/conf/poster/proceedings/Poster_2017/Section_IC/IC_070_Kosek.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Implementation of Cepstral Voice Activity Detector

Popis výsledku v původním jazyce

Analysis of Cepstral Voice Activity Detector (VAD) behavior in different acoustic conditions is presented in this paper. The analyzed VADs use computation of shorttime signal energy and cepstral distance as a criterion value (characteristic) for detecting speech activity and also several different methods of heuristic thresholding are used for a decision about speech and non-speech activity in current shorttime frame. The implementation of described VADs was written in C++ programming language and integrated into CtuCopy speech processing tool as an internal module and it also allows the application of VAD in speech feature extraction. Implemented VADs were tested on the QUT-NOISETIMIT corpus and their behavior under different acoustic conditions is studied and discussed in the experimental part.

Název v anglickém jazyce

Implementation of Cepstral Voice Activity Detector

Popis výsledku anglicky

Analysis of Cepstral Voice Activity Detector (VAD) behavior in different acoustic conditions is presented in this paper. The analyzed VADs use computation of shorttime signal energy and cepstral distance as a criterion value (characteristic) for detecting speech activity and also several different methods of heuristic thresholding are used for a decision about speech and non-speech activity in current shorttime frame. The implementation of described VADs was written in C++ programming language and integrated into CtuCopy speech processing tool as an internal module and it also allows the application of VAD in speech feature extraction. Implemented VADs were tested on the QUT-NOISETIMIT corpus and their behavior under different acoustic conditions is studied and discussed in the experimental part.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20201 - Electrical and electronic engineering

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2017

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the International Student Scientific Conference Poster – 21/2017

ISBN

978-80-01-06153-4

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

Praha

Datum konání akce

23. 5. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—