Hybrid speech enhancement with empirical mode decomposition and spectral subtraction for efficient speaker identification

Speech enhancement is a very important pre-processing step in various speech processing applications such as speech recognition, speaker identification, speech coding, and speech synthesis. In this paper, we focus on speech enhancement prior to speaker identification, because the degradations of the...

Full description

Saved in:
Bibliographic Details
Published in:International journal of speech technology 2015-12, Vol.18 (4), p.555-564
Main Authors: Abd El-Moneim, Samia, Dessouky, Moawad I., Abd El-Samie, Fathi E., Nassar, M. A., Abd El-Naby, Mohammed
Format: Article
Language:English
Subjects:
Artificial Intelligence
Engineering
Signal,Image and Speech Processing
Social Sciences
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Speech enhancement is a very important pre-processing step in various speech processing applications such as speech recognition, speaker identification, speech coding, and speech synthesis. In this paper, we focus on speech enhancement prior to speaker identification, because the degradations of the speech signals may cause difficulties in hearing, understanding, and speaker recognition. The paper presents a hybrid speech enhancement method based on empirical mode decomposition combined with spectral subtraction to improve the quality of speech signals prior to speaker identification. Simulation results show an improvement in speaker recognition rates with the proposed speech enhancement method as a pre-processing step.
ISSN:1381-2416
1572-8110
DOI:10.1007/s10772-015-9293-5