Improving Speech Recognition Rate through Analysis Parameters

Speech signal is redundant and non-stationary by nature. Because of vocal tract inertness these variations are not very rapid and the signal can be considered as stationary in short segments. It is presumed that in short-time magnitude spectrum the most distinct information of speech is contained. T...

Full description

Saved in:
Bibliographic Details
Main Authors: Eringis Deividas, Tamulevičius Gintautas
Format: Article
Language:English
Published: Riga Technical University Press 2014-05-01
Series:Electrical, Control and Communication Engineering
Subjects:
Online Access:https://doi.org/10.2478/ecce-2014-0009
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850037850341376000
author Eringis Deividas
Tamulevičius Gintautas
author_facet Eringis Deividas
Tamulevičius Gintautas
author_sort Eringis Deividas
collection DOAJ
description Speech signal is redundant and non-stationary by nature. Because of vocal tract inertness these variations are not very rapid and the signal can be considered as stationary in short segments. It is presumed that in short-time magnitude spectrum the most distinct information of speech is contained. This is the main reason for speech signal analysis in frame-by-frame manner. The analyzed speech signal is segmented into overlapping segments (so-called frames) for this purpose. Segments of 15-25 ms with the overlap of 10-15 ms are used usually.
format Article
id doaj-art-dda5cb399cb942ad8d681b4fe96457d9
institution DOAJ
issn 2255-9159
language English
publishDate 2014-05-01
publisher Riga Technical University Press
record_format Article
series Electrical, Control and Communication Engineering
spelling doaj-art-dda5cb399cb942ad8d681b4fe96457d92025-08-20T02:56:44ZengRiga Technical University PressElectrical, Control and Communication Engineering2255-91592014-05-0151616610.2478/ecce-2014-0009ecce-2014-0009Improving Speech Recognition Rate through Analysis ParametersEringis Deividas0Tamulevičius Gintautas1PhD Student, Vilnius University Institute of Mathematics and InformaticsResearcher, Vilnius University Institute of Mathematics and InformaticsSpeech signal is redundant and non-stationary by nature. Because of vocal tract inertness these variations are not very rapid and the signal can be considered as stationary in short segments. It is presumed that in short-time magnitude spectrum the most distinct information of speech is contained. This is the main reason for speech signal analysis in frame-by-frame manner. The analyzed speech signal is segmented into overlapping segments (so-called frames) for this purpose. Segments of 15-25 ms with the overlap of 10-15 ms are used usually.https://doi.org/10.2478/ecce-2014-0009computers and information processingspeech analysisspeech recognitionspeech enhancement
spellingShingle Eringis Deividas
Tamulevičius Gintautas
Improving Speech Recognition Rate through Analysis Parameters
Electrical, Control and Communication Engineering
computers and information processing
speech analysis
speech recognition
speech enhancement
title Improving Speech Recognition Rate through Analysis Parameters
title_full Improving Speech Recognition Rate through Analysis Parameters
title_fullStr Improving Speech Recognition Rate through Analysis Parameters
title_full_unstemmed Improving Speech Recognition Rate through Analysis Parameters
title_short Improving Speech Recognition Rate through Analysis Parameters
title_sort improving speech recognition rate through analysis parameters
topic computers and information processing
speech analysis
speech recognition
speech enhancement
url https://doi.org/10.2478/ecce-2014-0009
work_keys_str_mv AT eringisdeividas improvingspeechrecognitionratethroughanalysisparameters
AT tamuleviciusgintautas improvingspeechrecognitionratethroughanalysisparameters