SciELO - Scientific Electronic Library Online

 
vol.12 issue3Method of robust feature extraction for a diarization system.Tool for relational database learning. author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

  • Have no cited articlesCited by SciELO

Related links

  • Have no similar articlesSimilars in SciELO

Share


Revista Cubana de Ciencias Informáticas

On-line version ISSN 2227-1899

Abstract

REYES DIAZ, Flavio J.; ROBLE GUTIERREZ, Alejandro; HERNANDEZ SIERRA, Gabriel  and  CALVO DE LARA, José Ramón. Wiener filtering to noise reduction for speaker verification.. Rev cuba cienc informat [online]. 2018, vol.12, n.3, pp.152-162. ISSN 2227-1899.

Audio signals, including voice, are in some way exposed to quality deterioration due to the environmental noise incorporation. These noises existing in the signals, provoke a quality degradation in the speaker acoustic information, bringing with it a decrease of the speaker recognition performance. In this work an analysis of the behavior of some of the main noise reduction methods is performed such as: Wiener Filter and Spectral Subtraction, to pre-procesing the noisy signals. Finally, it is proposed to apply the Wiener filtering to the stage of pre-processing the signals of a speaker recognition system. Then, our proposal are evaluated on telephone session from NIST SRE-08 for different environmental noises types, obtaining an EER improvement of 4,94% and 12,5% for both evaluation conditions.

Keywords : noise; speaker verification; Wiener filter.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License