| |
Document Type
|
:
|
Latin Dissertation
|
Language of Document
|
:
|
English
|
Record Number
|
:
|
148587
|
Doc. No
|
:
|
ET20379
|
Main Entry
|
:
|
Narendranat h Malayat h
|
Title Proper
|
:
|
Data-Driven Methods for Extracting Features from Speech
|
Note
|
:
|
This document is digital این مدرک بصورت الکترونیکی می باشد
|
Abstract
|
:
|
Feature extraction plays a major role in any form of pattern recognition- Currentfeature extraction methods used for automatic speech recognition (ASR) and speakerverification rely mainly on properties of speech production (modeled by d-pole filters)and perception (critical band integration simulated by Mel/Bark filter bank). We proposestochastic methods t o design feature extraction methods which are trained t o alleviate theunwanted variability present in speech signal, In this dissertation we show that such data-driven met hods provide significant advantages over the conventional met hods for featureextraction..
|
Subject
|
:
|
Electericl tess
|
|
:
|
برق
|
electronic file name
|
:
|
TL43506.pdf
|
Title and statement of responsibility and
|
:
|
Data-Driven Methods for Extracting Features from Speech [Thesis]
|
| | | | | |
| | |  |
http://lib.khuisf.ac.ir/site/catalogue/148587
| | |
| | |  |
|