Science Alert
 
   
Information Technology Journal
  Year: 2010 | Volume: 9 | Issue: 1 | Page No.: 107-115
DOI: 10.3923/itj.2010.107.115
Improving Speaker Verification in Noisy Environments using Adaptive Filtering and Hybrid Classification Technique
M.Z. Ilyas, S.A. Samad, A. Hussain and K.A. Ishak

Abstract:
This study describes two approaches of improving speaker verification in noisy environments. The first approach is implementation of a speaker verification classification technique base on hybrid Vector Quantization (VQ) and Hidden Markov Models (HMMs) in clean and noisy environments. The second approach is implementation of Adaptive Noise Cancelation (ANC) as pre-processing for noise removal. The motivation to implement hybrid classification technique is to improve the HMMs performance. It is shown that, by using the hybrid technique, an Equal Error Rate (EER) of 11.72% is achieved compared to HMM alone, which achieved 16.66% in clean environments. However, both techniques show degradation in noisy environments. In order to address these problems, an Adaptive Noise Cancellation (ANC) technique using adaptive filtering is implemented in the pre-processing stage due to its ability to separate overlapping speech frequency bands. Investigations using Least-Mean-Square (LMS), Normalized Least-Mean-Square (NLMS) and Recursive Least-Squares (RLS) adaptive filtering are conducted to find the best solution for the speaker verification system.
 [Abstract]   [Fulltext PDF]   [Fulltext HTML]   [References]   [View Citation]  [Report Citation]

COMMENTS
13 September, 2012
najet:
I am very interested by your paper entitled : Improving Speaker Verification in Noisy Environments using Adaptive Filtering and Hybrid Classification Technique" and I would like to ask you if youcan send me the database you have used in your reserach. With warmest thank to your collaboration I note your collaboration as an author in my reserach papers. Thank you an other time for your help
 
REPLY TO ABOVE COMMENTS
.
 
 
 
 

Who are you?