Mémoires de Fin d’Etudes
Etablissement
Université de Boumerdès - M’hamed Bougara
Affiliation
Département Electronique
Auteur
KHETTAOUI, Billal
Directeur de thèse
Dahimene Abdelhakim (Maitre de conférence)
Filière
Electronique
Diplôme
Magister
Titre
Speaker Recognition using Gaussian Mixture Model
Mots clés
Speaker Recognition ; Speech Signal ; Gaussian mixture
Résumé
Speaker recognition is a biometric operation of accepting a claimed person based on analyzing his spoken utterance. A text Independent speaker recognition system based on Gaussian Mixture Model (GMM) was developed with a specific focus on the use of a Voice Activated Detector (VAD) algorithm in the training and testing phases with a comparison between high and low quefrency coefficients provided by the Mel Frequency Cepstral Coefficients (MFCC). At the training level, a modified Estimation/Maximization (EM) algorithm is used. It is less prone to get trapped around a local maximum and so, it will have more chance to converge to the global maximum of the model. A new method of background speaker’s model selection based on the identification results is also presented. High identification rate, low False Rejection (FR) and low False Acceptance (FA) are the most important parameters of the system design
Date de soutenance
2014
Cote
621.3(043.2)/A115/KHE
Pagination
72 p.
Illusatration
ill.
Format
30 cm
Statut
Traitée