Please use this identifier to cite or link to this item:

Title: Optimized European Portuguese Speech-To-Text using Deep Learning
Authors: Medeiros, Eduardo
Corado, Leonel
Rato, Luis
Quaresma, Paulo
Salgueiro, Pedro
Keywords: speech
deep learning
Issue Date: Oct-2022
Publisher: APRP
Citation: Medeiros, E., Corado,L., Rato, L., Quaresma, P., Salgueiro, P., Optimized European Portuguese Speech-To-Text using Deep Learning, RECPAD2022, 28th Portuguese Conference on Pattern Recognition, School of Technology and Management – Politécnico de Leiria, 2022.
Abstract: We have developed an ASR system for European Portuguese implement ing the QuartzNet [3] architecture with the NeMo [4] framework. Two approaches were used in this work: from scratch and using transfer learning. The experiments were data-driven focused instead of algorithm finetuning. Experiments confirm that models developed using transfer learning have shown better results (WER=0.0513) than developing models from scratch (WER=0.1945).
Type: article
Appears in Collections:INF - Artigos em Livros de Actas/Proceedings

Files in This Item:

File Description SizeFormat
RECPAD22_Speech2Text.pdf119.25 kBAdobe PDFView/Open
FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpaceOrkut
Formato BibTex mendeley Endnote Logotipo do DeGóis 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Dspace Dspace
DSpace Software, version 1.6.2 Copyright © 2002-2008 MIT and Hewlett-Packard - Feedback
UEvora B-On Curriculum DeGois