An Arabic Text To Speech Synthesis system based on Tacotron model

Please use this identifier to cite or link to this item: https://dspace.univ-ouargla.dz/jspui/handle/123456789/36703

Title:	An Arabic Text To Speech Synthesis system based on Tacotron model
Authors:	BETTAYEB, Nadjla Benettouati, Salah Eddine Boukhetta, Mottaz Bellah Bouzid, Mohamed laid
Keywords:	Tacotron TTS Deep Learning ANN Arabic language
Issue Date:	2024
Publisher:	UNIVERSITY KASDI MERBAH OUARGLA
Citation:	FACULTE DES NOUVELLES TECHNOLOGIES DE L'INFORMATIQUE ET DE LA COMMUNICATION
Abstract:	This thesis aims to build an Arabic Text To Speech (TTS) system using the deep learning model Tacotron. The first stage consists of training an Artificial Neural Network (ANN) model able to generate audio spectrograms from a given text. The second stage transforms the generated Spectrogram into speech, using the HiFi GAN audio enhancement model. The desired goals were fully reached, as the synthesized Arabic speech received a general rate of 3.66 over 5, from the system evaluators
URI:	https://dspace.univ-ouargla.dz/jspui/handle/123456789/36703
Appears in Collections:	Département d'Electronique et des Télécommunications - Master

Files in This Item:

File	Description	Size	Format
BEN TOUATI-BOUKHETTA-BOUZID.pdf		2,78 MB	Adobe PDF	View/Open

DSpace JSPUI