Comparing spectrum estimators in speaker verification under additive noise degradation
(2012) 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) p.4769-4772- Abstract
- Different short-term spectrum estimators for speaker verification under additive noise are considered. Conventionally, mel-frequency cepstral coefficients (MFCCs) are computed from discrete Fourier transform (DFT) spectra of windowed speech frames. Recently, linear prediction (LP) and its temporally weighted variants have been substituted as the spectrum analysis method in speech and speaker recognition. In this paper, 12 different short-term spectrum estimation methods are compared for speaker verification under additive noise contamination. Experimental results conducted on NIST 2002 SRE show that the spectrum estimation method has a large effect on recognition performance and stabilized weighted LP (SWLP) and minimum variance... (More)
- Different short-term spectrum estimators for speaker verification under additive noise are considered. Conventionally, mel-frequency cepstral coefficients (MFCCs) are computed from discrete Fourier transform (DFT) spectra of windowed speech frames. Recently, linear prediction (LP) and its temporally weighted variants have been substituted as the spectrum analysis method in speech and speaker recognition. In this paper, 12 different short-term spectrum estimation methods are compared for speaker verification under additive noise contamination. Experimental results conducted on NIST 2002 SRE show that the spectrum estimation method has a large effect on recognition performance and stabilized weighted LP (SWLP) and minimum variance distortionless response (MVDR) methods yield approximately 7 % and 8 % relative improvements over the standard DFT method at -10 dB SNR level of factory and babble noises, respectively in terms of equal error rate (EER). (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/3290081
- author
- Hanilci, C. ; Kinnunen, T. ; Saeidi, R. ; Pohjalainen, J. ; Alku, P. ; Ertas, F. ; Sandberg, J. and Sandsten, Maria LU
- organization
- publishing date
- 2012
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- speaker verification, spectrum estimation
- host publication
- Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on
- pages
- 4 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- conference location
- Kyoto, Japan
- conference dates
- 2012-03-25 - 2012-03-30
- external identifiers
-
- scopus:84867590081
- ISSN
- 1520-6149
- ISBN
- 978-1-4673-0044-5 (online)
- 978-1-4673-0045-2 (print)
- DOI
- 10.1109/ICASSP.2012.6288985
- language
- English
- LU publication?
- yes
- id
- 19390f56-2fc3-4b94-8f08-7bc18dec6457 (old id 3290081)
- alternative location
- http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6288985
- date added to LUP
- 2016-04-01 14:08:02
- date last changed
- 2022-02-19 17:11:42
@inproceedings{19390f56-2fc3-4b94-8f08-7bc18dec6457, abstract = {{Different short-term spectrum estimators for speaker verification under additive noise are considered. Conventionally, mel-frequency cepstral coefficients (MFCCs) are computed from discrete Fourier transform (DFT) spectra of windowed speech frames. Recently, linear prediction (LP) and its temporally weighted variants have been substituted as the spectrum analysis method in speech and speaker recognition. In this paper, 12 different short-term spectrum estimation methods are compared for speaker verification under additive noise contamination. Experimental results conducted on NIST 2002 SRE show that the spectrum estimation method has a large effect on recognition performance and stabilized weighted LP (SWLP) and minimum variance distortionless response (MVDR) methods yield approximately 7 % and 8 % relative improvements over the standard DFT method at -10 dB SNR level of factory and babble noises, respectively in terms of equal error rate (EER).}}, author = {{Hanilci, C. and Kinnunen, T. and Saeidi, R. and Pohjalainen, J. and Alku, P. and Ertas, F. and Sandberg, J. and Sandsten, Maria}}, booktitle = {{Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on}}, isbn = {{978-1-4673-0044-5 (online)}}, issn = {{1520-6149}}, keywords = {{speaker verification; spectrum estimation}}, language = {{eng}}, pages = {{4769--4772}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Comparing spectrum estimators in speaker verification under additive noise degradation}}, url = {{http://dx.doi.org/10.1109/ICASSP.2012.6288985}}, doi = {{10.1109/ICASSP.2012.6288985}}, year = {{2012}}, }