Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments

Mohd Zaizu Ilyas, Ali O Abid Noor, Khairul Anuar Ishak, Aini Hussain, Salina Abdul Samad

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

In this paper, we present a speaker verification system based on the Hidden Markov Models (HMMs) and Normalized Least Mean Square (NLMS) adaptive filtering. The aim of using NLMS adaptive filtering is to improve the HMMs performance in noisy environments. A Malay spoken digit database is used for the testing and validation modules. It is shown that, in a clean environment a Total Success Rate (TSR) of 89.97% is achieved using HMMs. For speaker verification, the true speaker rejection rate is 25.3% while the impostor acceptance rate is 9.99% and the equal error rate (EER) is 16.66%. In noisy environments without NLMS adaptive filtering TSRs of between 43.07%-51.26% are achieved for SNRs of 0-30 dBs. Meanwhile, after NLMS filtering, TSRs of between 55.18%-55.30% are achieved for SNRs 0-30 dB.

Original languageEnglish
Title of host publication2008 International Conference on Electronic Design, ICED 2008
DOIs
Publication statusPublished - 2008
Event2008 International Conference on Electronic Design, ICED 2008 - Penang
Duration: 1 Dec 20083 Dec 2008

Other

Other2008 International Conference on Electronic Design, ICED 2008
CityPenang
Period1/12/083/12/08

Fingerprint

Adaptive filtering
Hidden Markov models
Testing

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Ilyas, M. Z., Noor, A. O. A., Ishak, K. A., Hussain, A., & Abdul Samad, S. (2008). Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments. In 2008 International Conference on Electronic Design, ICED 2008 [4786632] https://doi.org/10.1109/ICED.2008.4786632

Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments. / Ilyas, Mohd Zaizu; Noor, Ali O Abid; Ishak, Khairul Anuar; Hussain, Aini; Abdul Samad, Salina.

2008 International Conference on Electronic Design, ICED 2008. 2008. 4786632.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ilyas, MZ, Noor, AOA, Ishak, KA, Hussain, A & Abdul Samad, S 2008, Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments. in 2008 International Conference on Electronic Design, ICED 2008., 4786632, 2008 International Conference on Electronic Design, ICED 2008, Penang, 1/12/08. https://doi.org/10.1109/ICED.2008.4786632
Ilyas, Mohd Zaizu ; Noor, Ali O Abid ; Ishak, Khairul Anuar ; Hussain, Aini ; Abdul Samad, Salina. / Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments. 2008 International Conference on Electronic Design, ICED 2008. 2008.
@inproceedings{862fe0962015476b877cab6545f959a8,
title = "Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments",
abstract = "In this paper, we present a speaker verification system based on the Hidden Markov Models (HMMs) and Normalized Least Mean Square (NLMS) adaptive filtering. The aim of using NLMS adaptive filtering is to improve the HMMs performance in noisy environments. A Malay spoken digit database is used for the testing and validation modules. It is shown that, in a clean environment a Total Success Rate (TSR) of 89.97{\%} is achieved using HMMs. For speaker verification, the true speaker rejection rate is 25.3{\%} while the impostor acceptance rate is 9.99{\%} and the equal error rate (EER) is 16.66{\%}. In noisy environments without NLMS adaptive filtering TSRs of between 43.07{\%}-51.26{\%} are achieved for SNRs of 0-30 dBs. Meanwhile, after NLMS filtering, TSRs of between 55.18{\%}-55.30{\%} are achieved for SNRs 0-30 dB.",
author = "Ilyas, {Mohd Zaizu} and Noor, {Ali O Abid} and Ishak, {Khairul Anuar} and Aini Hussain and {Abdul Samad}, Salina",
year = "2008",
doi = "10.1109/ICED.2008.4786632",
language = "English",
isbn = "9781424423156",
booktitle = "2008 International Conference on Electronic Design, ICED 2008",

}

TY - GEN

T1 - Normalized least mean square adaptive noise cancellation filtering for speaker verification in noisy environments

AU - Ilyas, Mohd Zaizu

AU - Noor, Ali O Abid

AU - Ishak, Khairul Anuar

AU - Hussain, Aini

AU - Abdul Samad, Salina

PY - 2008

Y1 - 2008

N2 - In this paper, we present a speaker verification system based on the Hidden Markov Models (HMMs) and Normalized Least Mean Square (NLMS) adaptive filtering. The aim of using NLMS adaptive filtering is to improve the HMMs performance in noisy environments. A Malay spoken digit database is used for the testing and validation modules. It is shown that, in a clean environment a Total Success Rate (TSR) of 89.97% is achieved using HMMs. For speaker verification, the true speaker rejection rate is 25.3% while the impostor acceptance rate is 9.99% and the equal error rate (EER) is 16.66%. In noisy environments without NLMS adaptive filtering TSRs of between 43.07%-51.26% are achieved for SNRs of 0-30 dBs. Meanwhile, after NLMS filtering, TSRs of between 55.18%-55.30% are achieved for SNRs 0-30 dB.

AB - In this paper, we present a speaker verification system based on the Hidden Markov Models (HMMs) and Normalized Least Mean Square (NLMS) adaptive filtering. The aim of using NLMS adaptive filtering is to improve the HMMs performance in noisy environments. A Malay spoken digit database is used for the testing and validation modules. It is shown that, in a clean environment a Total Success Rate (TSR) of 89.97% is achieved using HMMs. For speaker verification, the true speaker rejection rate is 25.3% while the impostor acceptance rate is 9.99% and the equal error rate (EER) is 16.66%. In noisy environments without NLMS adaptive filtering TSRs of between 43.07%-51.26% are achieved for SNRs of 0-30 dBs. Meanwhile, after NLMS filtering, TSRs of between 55.18%-55.30% are achieved for SNRs 0-30 dB.

UR - http://www.scopus.com/inward/record.url?scp=63649128107&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=63649128107&partnerID=8YFLogxK

U2 - 10.1109/ICED.2008.4786632

DO - 10.1109/ICED.2008.4786632

M3 - Conference contribution

AN - SCOPUS:63649128107

SN - 9781424423156

BT - 2008 International Conference on Electronic Design, ICED 2008

ER -