AL-Lisaniyyat
Volume 30, Numéro 2, Pages 183-195
2024-12-30
Formants And Prosodic Features' Effects On Arabic Speaker Identification Accuracy In Noisy Environements.
Authors : Boubakeur Khadidja Nesrine . Debyeche :mohamed .
Abstract
This study investigates the use of formants and prosodic features, specifically pitch and intensity, for speaker identification in real conditions. To enhance the robustness of the acoustic models against speech signal variations in noisy environments, Mel-Frequency Cepstral Coefficient (MFCC) are added to these features. A Speaker Identification system based on Hidden Markov Models (HMM) is implemented in the independent text mode. The combination of formants and prosodic features with cepstral features improves the identification accuracy, particularly in high-noise environments, up to 10%, in comparison to an MFCC based system. The results show that the use of multivariate feature vectors significantly improves the performance of an identification system in the presence of noise compared to an MFCC-based system.
Keywords
ASI ; MFCC ; Prosodic features ; Pitch ; Energy ; Formants ; HMM ; Noise
Les articles similaires
Akak Malika
.
Sayoud Halim
.
pages 43-51.
Friha Souad
.
Mansouri Nora
.
Taleb Ahmed Abdelmalik
.
pages 101-107.
Lounnas Khaled
.
Gamgani Abderrahmane
.
Aliane Imad Feth-ennour
.
pages 196-215.
Djebali-haouche Tassadit
.
pages 167-173.
