Frame Selection for Text-independent Speaker Recognition

Abedenebi Rouigueb, Malek Nadil, Abderrahmane Tikourt

2017

Abstract

In this paper, we propose a set of criteria for the selection of the most relevant frames in order to improve text-independent speaker automatic recognition (TISAR) task. The selection is carried out on the short term Cepstral feature vectors such as PLP and MFCC and performed at the front end processing level. The proposed criteria mainly attempt to select vectors lying far from the universal background model (UBM). Experiments are conducted on the MOBIO database and show that the selection allows an improvement in complexity (time and space) and in speaker identification rate, which is appropriate for real-time TISAR systems.

Download


Paper Citation


in Harvard Style

Rouigueb A., Nadil M. and Tikourt A. (2017). Frame Selection for Text-independent Speaker Recognition . In Proceedings of the 14th International Joint Conference on e-Business and Telecommunications - Volume 1: SIGMAP, (ICETE 2017) ISBN 978-989-758-260-8, pages 51-57. DOI: 10.5220/0006392100510057


in Bibtex Style

@conference{sigmap17,
author={Abedenebi Rouigueb and Malek Nadil and Abderrahmane Tikourt},
title={Frame Selection for Text-independent Speaker Recognition},
booktitle={Proceedings of the 14th International Joint Conference on e-Business and Telecommunications - Volume 1: SIGMAP, (ICETE 2017)},
year={2017},
pages={51-57},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006392100510057},
isbn={978-989-758-260-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 14th International Joint Conference on e-Business and Telecommunications - Volume 1: SIGMAP, (ICETE 2017)
TI - Frame Selection for Text-independent Speaker Recognition
SN - 978-989-758-260-8
AU - Rouigueb A.
AU - Nadil M.
AU - Tikourt A.
PY - 2017
SP - 51
EP - 57
DO - 10.5220/0006392100510057