Analysis of compressed speech signals in an Automatic Speaker Recognition system

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

1 Citation (Scopus)

Abstract

This paper analyzes the effects popular audio compression algorithms have on the performance of a speaker recognition system. Popular audio compression algorithms were used to compress both clean and noisy speech before being passed to a speaker recognition system. The features extracted from each speaker were 19-dimensional Mel-Frequency Cepstrum Coefficients (MFCC) and the corresponding features were modeled using a 16 mixture Gaussian Mixture Model (GMM). Our experiments show that compression will have a negative effect on recognition rates if the compressed speech is clean. However, if small amounts of white Gaussian noise are added before the speech is compressed, recognition rates can be increased by as much as 7% with certain compression algorithms.

Original languageEnglish (US)
Title of host publication2015 49th Annual Conference on Information Sciences and Systems, CISS 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781479984282
DOIs
StatePublished - Apr 15 2015
Event2015 49th Annual Conference on Information Sciences and Systems, CISS 2015 - Baltimore, United States
Duration: Mar 18 2015Mar 20 2015

Other

Other2015 49th Annual Conference on Information Sciences and Systems, CISS 2015
CountryUnited States
CityBaltimore
Period3/18/153/20/15

Fingerprint

Experiments

All Science Journal Classification (ASJC) codes

  • Information Systems

Cite this

Metzger, R. A., Doherty, J. F., & Jenkins, Jr., D. M. (2015). Analysis of compressed speech signals in an Automatic Speaker Recognition system. In 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015 [7086817] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CISS.2015.7086817
Metzger, Richard A. ; Doherty, John F. ; Jenkins, Jr., David Marion. / Analysis of compressed speech signals in an Automatic Speaker Recognition system. 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc., 2015.
@inproceedings{ecb13a281351494f9456c6d343504cb8,
title = "Analysis of compressed speech signals in an Automatic Speaker Recognition system",
abstract = "This paper analyzes the effects popular audio compression algorithms have on the performance of a speaker recognition system. Popular audio compression algorithms were used to compress both clean and noisy speech before being passed to a speaker recognition system. The features extracted from each speaker were 19-dimensional Mel-Frequency Cepstrum Coefficients (MFCC) and the corresponding features were modeled using a 16 mixture Gaussian Mixture Model (GMM). Our experiments show that compression will have a negative effect on recognition rates if the compressed speech is clean. However, if small amounts of white Gaussian noise are added before the speech is compressed, recognition rates can be increased by as much as 7{\%} with certain compression algorithms.",
author = "Metzger, {Richard A.} and Doherty, {John F.} and {Jenkins, Jr.}, {David Marion}",
year = "2015",
month = "4",
day = "15",
doi = "10.1109/CISS.2015.7086817",
language = "English (US)",
booktitle = "2015 49th Annual Conference on Information Sciences and Systems, CISS 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Metzger, RA, Doherty, JF & Jenkins, Jr., DM 2015, Analysis of compressed speech signals in an Automatic Speaker Recognition system. in 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015., 7086817, Institute of Electrical and Electronics Engineers Inc., 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015, Baltimore, United States, 3/18/15. https://doi.org/10.1109/CISS.2015.7086817

Analysis of compressed speech signals in an Automatic Speaker Recognition system. / Metzger, Richard A.; Doherty, John F.; Jenkins, Jr., David Marion.

2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7086817.

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

TY - GEN

T1 - Analysis of compressed speech signals in an Automatic Speaker Recognition system

AU - Metzger, Richard A.

AU - Doherty, John F.

AU - Jenkins, Jr., David Marion

PY - 2015/4/15

Y1 - 2015/4/15

N2 - This paper analyzes the effects popular audio compression algorithms have on the performance of a speaker recognition system. Popular audio compression algorithms were used to compress both clean and noisy speech before being passed to a speaker recognition system. The features extracted from each speaker were 19-dimensional Mel-Frequency Cepstrum Coefficients (MFCC) and the corresponding features were modeled using a 16 mixture Gaussian Mixture Model (GMM). Our experiments show that compression will have a negative effect on recognition rates if the compressed speech is clean. However, if small amounts of white Gaussian noise are added before the speech is compressed, recognition rates can be increased by as much as 7% with certain compression algorithms.

AB - This paper analyzes the effects popular audio compression algorithms have on the performance of a speaker recognition system. Popular audio compression algorithms were used to compress both clean and noisy speech before being passed to a speaker recognition system. The features extracted from each speaker were 19-dimensional Mel-Frequency Cepstrum Coefficients (MFCC) and the corresponding features were modeled using a 16 mixture Gaussian Mixture Model (GMM). Our experiments show that compression will have a negative effect on recognition rates if the compressed speech is clean. However, if small amounts of white Gaussian noise are added before the speech is compressed, recognition rates can be increased by as much as 7% with certain compression algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84929192014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929192014&partnerID=8YFLogxK

U2 - 10.1109/CISS.2015.7086817

DO - 10.1109/CISS.2015.7086817

M3 - Conference contribution

BT - 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Metzger RA, Doherty JF, Jenkins, Jr. DM. Analysis of compressed speech signals in an Automatic Speaker Recognition system. In 2015 49th Annual Conference on Information Sciences and Systems, CISS 2015. Institute of Electrical and Electronics Engineers Inc. 2015. 7086817 https://doi.org/10.1109/CISS.2015.7086817