TY - JOUR

T1 - A comparison of procedures for the calculation of forensic likelihood ratios from acoustic-phonetic data

T2 - Multivariate kernel density (MVKD) versus Gaussian mixture model-universal background model (GMM-UBM)

AU - Morrison, Geoffrey Stewart

PY - 2011/2

Y1 - 2011/2

N2 - Two procedures for the calculation of forensic likelihood ratios were tested on the same set of acoustic-phonetic data. One procedure was a multivariate kernel density procedure (MVKD) which is common in acoustic-phonetic forensic voice comparison, and the other was a Gaussian mixture model-universal background model (GMM-UBM) which is common in automatic forensic voice comparison. The data were coefficient values from discrete cosine transforms fitted to second-formant trajectories of /a/, /e/, /o/, /a/, and // tokens produced by 27 male speakers of Australian English. Scores were calculated separately for each phoneme and then fused using logistic regression. The performance of the fused GMM-UBM system was much better than that of the fused MVKD system, both in terms of accuracy (as measured using the log-likelihood-ratio cost, Cllr) and precision (as measured using an empirical estimate of the 95% credible interval for the likelihood ratios from the different-speaker comparisons).

AB - Two procedures for the calculation of forensic likelihood ratios were tested on the same set of acoustic-phonetic data. One procedure was a multivariate kernel density procedure (MVKD) which is common in acoustic-phonetic forensic voice comparison, and the other was a Gaussian mixture model-universal background model (GMM-UBM) which is common in automatic forensic voice comparison. The data were coefficient values from discrete cosine transforms fitted to second-formant trajectories of /a/, /e/, /o/, /a/, and // tokens produced by 27 male speakers of Australian English. Scores were calculated separately for each phoneme and then fused using logistic regression. The performance of the fused GMM-UBM system was much better than that of the fused MVKD system, both in terms of accuracy (as measured using the log-likelihood-ratio cost, Cllr) and precision (as measured using an empirical estimate of the 95% credible interval for the likelihood ratios from the different-speaker comparisons).

KW - Acoustic-phonetic

KW - Forensic voice comparison

KW - GMM-UBM

KW - Likelihood ratio

KW - Multivariate kernel density

UR - http://www.scopus.com/inward/record.url?scp=78649503176&partnerID=8YFLogxK

U2 - 10.1016/j.specom.2010.09.005

DO - 10.1016/j.specom.2010.09.005

M3 - Article

SN - 0167-6393

VL - 53

SP - 242

EP - 256

JO - Speech Communication

JF - Speech Communication

IS - 2

ER -