Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra

Phil Rose

doi:10.1109/ICASSP.2011.5947704

Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra

Phil Rose^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

10 Citations (Scopus)

Abstract

The suitability of voiceless fricative spectra for forensic voice comparison is explored within a Likelihood Ratio-based framework. Non-contemporaneous landline telephone recordings of 99 male Japanese speakers are compared using only tokens of their voiceless alveolo-patalal fricative [6]. A subset of mean-cepstrally-subtracted LPC CCs from the fricative spectrum from dc to 5 kHz is used. GMM/UBM and multivariate likelihood ratios are extracted for the 99 target and 4851 non-target trials, and fused with logistic regression. An EER of 7.4% and log-LR cost of 0.26 is demonstrated. It is concluded that the [6] spectrum does have some individualising potential.

Original language	English
Title of host publication	2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings
Pages	5900-5903
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2011.5947704
Publication status	Published - 2011
Event	36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Prague, Czech Republic Duration: 22 May 2011 → 27 May 2011

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Conference

Conference	36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011
Country/Territory	Czech Republic
City	Prague
Period	22/05/11 → 27/05/11

Access to Document

10.1109/ICASSP.2011.5947704

Cite this

Rose, P. (2011). Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra. In 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings (pp. 5900-5903). Article 5947704 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2011.5947704

Rose, Phil. / Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra. 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings. 2011. pp. 5900-5903 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{4098fa8d6db64421b81f5a301e5572f1,

title = "Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra",

abstract = "The suitability of voiceless fricative spectra for forensic voice comparison is explored within a Likelihood Ratio-based framework. Non-contemporaneous landline telephone recordings of 99 male Japanese speakers are compared using only tokens of their voiceless alveolo-patalal fricative [6]. A subset of mean-cepstrally-subtracted LPC CCs from the fricative spectrum from dc to 5 kHz is used. GMM/UBM and multivariate likelihood ratios are extracted for the 99 target and 4851 non-target trials, and fused with logistic regression. An EER of 7.4% and log-LR cost of 0.26 is demonstrated. It is concluded that the [6] spectrum does have some individualising potential.",

keywords = "Forensic Voice Comparison, GMM/UBM, Multivariate Likelihood Ratio, cepstrum, coronal fricative spectra",

author = "Phil Rose",

year = "2011",

doi = "10.1109/ICASSP.2011.5947704",

language = "English",

isbn = "9781457705397",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "5900--5903",

booktitle = "2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings",

note = "36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 ; Conference date: 22-05-2011 Through 27-05-2011",

}

Rose, P 2011, Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra. in 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings., 5947704, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 5900-5903, 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011, Prague, Czech Republic, 22/05/11. https://doi.org/10.1109/ICASSP.2011.5947704

Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra. / Rose, Phil.
2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings. 2011. p. 5900-5903 5947704 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra

AU - Rose, Phil

PY - 2011

Y1 - 2011

N2 - The suitability of voiceless fricative spectra for forensic voice comparison is explored within a Likelihood Ratio-based framework. Non-contemporaneous landline telephone recordings of 99 male Japanese speakers are compared using only tokens of their voiceless alveolo-patalal fricative [6]. A subset of mean-cepstrally-subtracted LPC CCs from the fricative spectrum from dc to 5 kHz is used. GMM/UBM and multivariate likelihood ratios are extracted for the 99 target and 4851 non-target trials, and fused with logistic regression. An EER of 7.4% and log-LR cost of 0.26 is demonstrated. It is concluded that the [6] spectrum does have some individualising potential.

AB - The suitability of voiceless fricative spectra for forensic voice comparison is explored within a Likelihood Ratio-based framework. Non-contemporaneous landline telephone recordings of 99 male Japanese speakers are compared using only tokens of their voiceless alveolo-patalal fricative [6]. A subset of mean-cepstrally-subtracted LPC CCs from the fricative spectrum from dc to 5 kHz is used. GMM/UBM and multivariate likelihood ratios are extracted for the 99 target and 4851 non-target trials, and fused with logistic regression. An EER of 7.4% and log-LR cost of 0.26 is demonstrated. It is concluded that the [6] spectrum does have some individualising potential.

KW - Forensic Voice Comparison

KW - GMM/UBM

KW - Multivariate Likelihood Ratio

KW - cepstrum

KW - coronal fricative spectra

UR - http://www.scopus.com/inward/record.url?scp=80051620852&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2011.5947704

DO - 10.1109/ICASSP.2011.5947704

M3 - Conference contribution

SN - 9781457705397

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 5900

EP - 5903

BT - 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings

T2 - 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011

Y2 - 22 May 2011 through 27 May 2011

ER -

Rose P. Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra. In 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings. 2011. p. 5900-5903. 5947704. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2011.5947704

Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this