2016
DOI: 10.1080/02564602.2016.1185976
|View full text |Cite
|
Sign up to set email alerts
|

A Review on Feature Extraction for Speaker Recognition under Degraded Conditions

Abstract: Speech is a signal that includes speaker's emotion, characteristic specification, phonemeinformation etc. Various methods have been proposed for speaker recognition by extracting specifications of a given utterance. Among them, short-term cepstral features are used excessively in speech, and speaker recognition areas because of their low complexity, and high performance in controlled environments. On the other hand, their performances decrease dramatically under degraded conditions such as channel mismatch, ad… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
12
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
6
4

Relationship

1
9

Authors

Journals

citations
Cited by 31 publications
(12 citation statements)
references
References 91 publications
(96 reference statements)
0
12
0
Order By: Relevance
“…Moreover, the evaluation procedure was done through two experiment series: the first series attempts to compare the different features extraction techniques on the overall evaluation dataset using the most appropriate modeling parameters, whereas the second series aims to assess their performances through cross-validation strategy (see the experimental protocol). During this study, our focus is put on the traditional and the widely used MFCC features, the recently introduced GFCC features, as well as, on their corresponding dynamic features [44][45][46].…”
Section: Towards An Objective Comparison Methodology Of Speaker Featuresmentioning
confidence: 99%
“…Moreover, the evaluation procedure was done through two experiment series: the first series attempts to compare the different features extraction techniques on the overall evaluation dataset using the most appropriate modeling parameters, whereas the second series aims to assess their performances through cross-validation strategy (see the experimental protocol). During this study, our focus is put on the traditional and the widely used MFCC features, the recently introduced GFCC features, as well as, on their corresponding dynamic features [44][45][46].…”
Section: Towards An Objective Comparison Methodology Of Speaker Featuresmentioning
confidence: 99%
“…Their results indicated that pure mel frequency cepstral coefficient (MFCC) approaches are the most frequently used approaches. A review of feature extraction methods for speaker recognition under degraded conditions, such as channel mismatch, additive noise, and emotional variability, was presented in [16]. Jawarkar et al [17] compared four sets of features: MFCCs, line spectral frequencies, Teager energy-based MFCCs, and the temporal energy of subband cepstral coefficients.…”
Section: Literature Reviewmentioning
confidence: 99%
“…State‐of‐the‐art ASV systems mostly use short‐term spectral features. Mel‐frequency cepstral coefficients (MFCCs), perceptual linear prediction (LP) and linear predictive CCs (LPCC) are widely used feature extraction techniques due to their considerable performance and lower‐computational complexity [1, 9, 19, 23, 2729].…”
Section: Brief Overview Of Asvmentioning
confidence: 99%