A comparison of procedures for the calculation of forensic likelihood ratios from acoustic-phonetic data: Multivariate kernel density (MVKD) versus Gaussian mixture model-universal background model (GMM-UBM)

Speech Communication

Volumn 53, Issue 2, 2011, Pages 242-256

  Morrison, Geoffrey Stewart ^a,b  

^a AUSTRALIAN NATIONAL UNIVERSITY   (Australia)

^b UNIVERSITY OF NEW SOUTH WALES   (Australia)

Author keywords

Acoustic phonetic; Forensic voice comparison; GMM UBM; Likelihood ratio; Multivariate kernel density

Indexed keywords

COEFFICIENT VALUES; CREDIBLE INTERVAL; EMPIRICAL ESTIMATE; FORENSIC VOICE COMPARISON; FORMANT TRAJECTORY; GAUSSIAN MIXTURE MODEL-UNIVERSAL BACKGROUND MODELS; GMM-UBM; KERNEL DENSITY; LIKELIHOOD RATIOS; LOG LIKELIHOOD; LOGISTIC REGRESSIONS; MALE SPEAKERS; UBM SYSTEMS;

COSINE TRANSFORMS; DISCRETE COSINE TRANSFORMS; LINGUISTICS; MODELS;

SPEECH RECOGNITION;

EID: 78649503176     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.specom.2010.09.005     Document Type: Article

References (70)

1. A. Agresti An Introduction to Categorical Data Analysis second ed. 2007 Wiley Hoboken, NJ
2. C.G.G. Aitken, and D. Lucy Evaluation of trace evidence in the form of multivariate data Appl. Statist. 53 2004 109 122 10.1046/j.0035-9254.2003.05271. x
3. C.G.G. Aitken, and D. Lucy Corrigendum: evaluation of trace evidence in the form of multivariate data Appl. Statist. 53 2004 665 666 10.1111/j.1467-9876.2004.02031.x
4. C.G.G. Aitken, D.A. Stoney, The Use of Statistics in Forensic Science 1991 Ellis Horwood Chichester, UK
5. C.G.G. Aitken, and F. Taroni Statistics and the Evaluation of Forensic Evidence for Forensic Scientist second ed. 2004 Wiley Chichester, UK
6. Alderman, T., 2004. The use of Australian-English vowel formant data sets in forensic speaker identification. In: Proc. 10th Australian Internat. Conf. on Speech Science & Technology, Australian Speech Science & Technology Association, pp. 177-182.
7. Alexander, A., 2005. Forensic automatic speaker recognition using Bayesian interpretation and statistical compensation for mismatched conditions. Ph.D. Dissertation, École polytechnique fédérale de Lausanne, Lausanne, Switzerland.
8. Alexander, A., Drygajlo, A., 2004. Scoring and direct methods for the interpretation of evidence in forensic speaker recognition. In: Proc. 8th Internat. Conf. on Spoken Language Processing (ICSLP), Jeju, Korea.
9. A. Alexander, D. Dessimoz, F. Botti, and A. Drygajlo Aural and automatic forensic speaker recognition in mismatched conditions Internat. J. Speech, Lang. Law 12 2005 214 234
10. Association of Forensic Science Providers, 2009. Standards for the formulation of evaluative forensic science expert opinion. Sci. Justice 49, 161-164. doi:10.1016/j.scijus.2009.07.004.
11. D.J. Balding Weight-of-Evidence for Forensic DNA Profiles 2005 Wiley Chichester, UK
12. Becker, T., Jessen, M., Grigoras, C. 2008. Forensic speaker verification using formant features and Gaussian mixture models. In: Proc. Interspeech 2008 Incorporating SST 2008, International Speech Communication Association, pp. 1505-1508.
13. Becker, T., Jessen, M., Grigoras, C. 2009. Speaker verification based on formants using Gaussian mixture models. In: Proc. of NAG/DAGA International Conference on Acoustics, Rotterdam.
14. Becker, T., Jessen, M., Alsbach, S., Broß, F., Meier, T., 2010. SPES - The BKA forensic automatic voice comparison system. In: Proc. Odyssey 2010: The Language and Speaker Recognition Workshop, Brno, Czech Republic, pp. 58-62.
15. Broeders, A.P.A., 1995. The role of automatic speaker recognition techniques in forensic investigations. In: Proc. Internat. Congress of Phonetic Sciences, Stockholm, Vol. 3, pp. 154-161.
16. N. Brümmer, and J. du Preez Application independent evaluation of speaker detection Comput. Speech Lang. 20 2006 230 275 10.1016/j.csl.2005.08.001
17. N. Brümmer, L. Burget, J.H. Cernocký, O. Glembek, F. Grézl, M. Karafiát, D.A. van Leeuwen, P. Matejka, P. Schwarz, and A. Strasheim Fusion of heterogenous speaker recognition systems in the STBU submission for the NIST SRE 2006 IEEE Trans. Audio Speech Lang. Process. 15 2007 2072 2084 10.1109/TASL.2007.902870
18. J. Buckleton A framework for interpreting evidence J. Buckleton, C.M. Triggs, S.J. Walsh, Forensic DNA Evidence Interpretation 2005 CRC Boca Raton, FL 27 63
19. C. Champod, and D. Meuwly The inference of identity in forensic speaker recognition Speech Commun. 31 2000 193 203 10.1016/S0167-6393(99)00078-3
20. R. Cook, I.W. Evett, G. Jackson, P.J. Jones, and J.A. Lambert A hierarchy of propositions: Deciding which level to address in casework Sci. Justice 38 1998 231 239 10.1016/S1355-0306(98)72099-4
21. J.M. Curran An introduction to Bayesian credible intervals for sampling error in DNA profiles Law, Prob. Risk 4 2005 115 126 10.1093/lpr/mgi009
22. J.M. Curran, J.S. Buckleton, C.M. Triggs, and B.S. Weir Assessing uncertainty in DNA evidence caused by sampling effects Sci. Justice 42 2002 29 37 10.1016/S1355-0306(02)71794-2
23. A. Drygajlo Forensic automatic speaker recognition IEEE Signal Process. Magazine March 2007 132 135 10.1109/MSP.2007.323278
24. R.O. Duda, P.E. Hart, and D.G. Stork Pattern Classification second ed. 2000 Wiley New York
25. Enzinger, E., 2010. Characterising formant tracks in Viennese diphthongs for forensic speaker comparison. In: Proc. 39th Audio Engineering Society Conf. on Audio Forensics: Practices and Challenges, Hillerød, Denmark.
26. I.W. Evett The theory of interpreting scientific transfer evidence Forensic Sci. Prog. 4 1990 141 179
27. I.W. Evett Towards a uniform framework for reporting opinions in forensic science case-work Sci. Justice 38 1998 198 202 10.1016/S1355-0306(98)72105-7
28. I.W. Evett, and J.S. Buckleton Statistical analysis of STR data A. Carraredo, B. Brinkmann, W. Bär, Advances in Forensic Haemogenetics Vol. 6 1996 Springer-Verlag Heidelberg, Germany 79 86
29. I.W. Evett, G. Jackson, J.A. Lambert, and S. McCrossan The impact of the principles of evidence interpretation on the structure and content of statements Sci. Justice 40 2000 233 239 10.1016/S1355-0306(00)71993-9
30. L. Ferrer, K. Sönmez, and E. Shriberg An anticorrelation kernel for subsystem training in multiple classifier systems J. Machine Learn. Res. 10 2009 2079 2114
31. J. González-Rodríguez, J. Fierrez-Aguilar, D. Ramos-Castro, and J. Ortega-García Bayesian analysis of fingerprint, face and signature evidences with automatic biometric systems Forensic Sci. Internat. 155 2005 126 140 10.1016/j.forsciint.2004.11.007
32. J. González-Rodríguez, A. Drygajlo, D. Ramos-Castro, M. García-Gomar, and J. Ortega-García Robust estimation, interpretation and assessment of likelihood ratios in forensic speaker recognition Comput. Speech Lang. 20 2006 331 355 10.1016/j.csl.2005.08.005
33. J. González-Rodríguez, P. Rose, D. Ramos, D. Torre, and J. Ortega-García Emulating DNA: Rigorous quantification of evidential weight in transparent and testable forensic speaker recognition IEEE Trans. Audio Speech Lang. Process. 15 2007 2104 2115 10.1109/TASL.2007.902747
34. B.J. Guillemin, and C. Watson Impact of the GSM mobile phone network on the speech signal: Some preliminary findings Internat. J. Speech Lang. Law 15 2008 193 218 10.1558/ijsll.v15i2.193
35. T. Hastie, R. Tibshirani, and J. Friedman The Elements of Statistical Learning: Data Mining, Inference, and Prediction second ed. 2009 Springer New York
36. D.W. Hosmer, and S. Lemeshow Applied Logistic Regression second ed. 2000 Wiley New York
37. Ishihara, S., Kinoshita, Y., 2008. How many do we need? Exploration of the population size effect on the performance of forensic speaker classification. In: Proc. Interspeech 2008 Incorporating SST 2008, International Speech Communication Association, pp. 1941-1944.
38. M. Jessen Forensic phonetics Lang. Ling. Compass 2 2008 671 711 10.1111/j.1749-818x.2008.00066.x
39. Kahn, J., Audibert, N., Rossato, S., Bonastre, J.-F., 2010. Intra-speaker variability effects on speaker verification performance. In: Proc. Odyssey 2010: The Language and Speaker Recognition Workshop, Brno, Czech Republic, pp. 109-116.
40. Kinoshita, Y., Osanai, T. 2006. Within speaker variation in diphthongal dynamics: What can we compare? In: Warren, P., Watson, C.I. (Eds.), Proc. 11th Australasian Internat. Conf. on Speech Science & Technology, Auckland, New Zealand. Australasian Speech Science & Technology Association, Canberra, Australia, pp. 112-117.
41. Kinoshita, Y., Ishihara, S., Rose, P., 2008. Beyond the long-term mean: Exploring the potential of F0 distribution parameters in traditional forensic speaker recognition. In: Proc. Odyssey 2008 The Speaker and Language Recognition Workshop, paper 03.
42. Y. Kinoshita, S. Ishihara, and P. Rose Exploring the discriminatory potential of F0 distribution parameters in traditional forensic speaker recognition Internat. J. Speech Lang. Law 16 2009 91 111 10.1558/ijsll.v16i1.91
43. Lewis, S.R., 1984. Philosophy of speaker identification. Police applications of speech and tape recording analysis. In: Proc. Institute of Acoustics, Vol. 6, pp. 69-77.
44. D.V. Lindley A problem in forensic science Biometrika 64 1977 207 213 10.1093/biomet/64.2.207
45. D. Lucy Introduction to Statistics for Forensic Scientists 2005 Wiley Chichester, UK
46. S. Menard Applied Logistic Regression 2002 Sage Thousand Oaks, CA
47. Meuwly, D., 2001. Reconnaissance de locuteurs en sciences forensiques: l'apport d'une approche automatique. Ph.D. Dissertation, University of Lausanne, Lausanne, Switzerland.
48. Meuwly, D., Drygajlo, A., 2001. Forensic speaker recognition based on a Bayesian framework and Gaussian mixture modelling. In: Proc. 2001: A Speaker Odyssey, The Speaker Recognition Workshop, Crete. International Speech Communication Association.
49. G.S. Morrison Forensic voice comparison using likelihood ratios based on polynomial curves fitted to the formant trajectories of Australian English /a/ Internat. J. Speech Lang. Law 15 2008 247 264 10.1558/ijsll.v15i2.249
50. G.S. Morrison Forensic voice comparison and the paradigm shift Sci. Justice 49 2009 298 308 10.1016/j.scijus.2009.09.002
51. G.S. Morrison Likelihood-ratio forensic voice comparison using parametric representations of the formant trajectories of diphthongs J. Acoust. Soc. Amer. 125 2009 2387 2397 10.1121/1.3081384
52. G.S. Morrison Forensic voice comparison I. Freckelton, H. Selby, Expert Evidence 2010 Thomson Reuters Sydney, Australia (Chapter 99)
53. Morrison, G.S., Kinoshita, Y., 2008. Automatic-type calibration of traditionally derived likelihood ratios: Forensic analysis of Australian English /o/ formant trajectories. In: Proc. Interspeech 2008 Incorporating SST 2008, International Speech Communication Association, pp. 1501-1504.
54. Morrison, G.S., Thiruvaran, T., Epps, J., 2010. Estimating the precision of the likelihood-ratio output of a forensic-voice-comparison system. In: Proc. Odyssey 2010: The Language and Speaker Recognition Workshop, Brno, Czech Republic, pp. 63-70.
55. National Research Council, 2009. Strengthening Forensic Science in the United States: A Path Forward. National Academies Press, Washington, DC.
56. S. Pigeon, P. Druyts, and P. Verlinde Applying logistic regression to the fusion of the NIST'99 1-speaker submissions Digit. Signal Process. 10 2000 237 248 10.1006/dspr.1999.0358
57. Ramos Castro, D., 2007. Forensic evaluation of the evidence using automatic speaker recognition systems. Ph.D. Dissertation, Universidad Autónoma de Madrid, Madrid, Spain.
58. Ramos-Castro, D., González-Rodríguez, J., Ortega-Garcia, J., 2006. Likelihood ratio calibration in a transparent and testable forensic speaker recognition framework. In: Berkling, K., Torres-Carrasquillo, P.A. (Eds.), Proc. Odyssey 2006 The Speaker and Language Recognition Workshop, IEEE. doi:10.1109/ODYSSEY.2006.248088. (Pubitemid 46933648)
59. D.A. Reynolds, T.F. Quatieri, and R.B. Dunn Speaker verification using adapted Gaussian mixture models Digit. Signal Process. 10 2000 19 41 10.1006/dspr.1999.0361
60. B. Robertson, and G.A. Vignaux Interpreting Evidence 1995 Wiley Chichester, UK
61. P. Rose Forensic Speaker Identification 2002 Taylor and Francis London, UK
62. Rose, P., 2006a. Accounting for correlation in linguistic-acoustic likelihood ratio-based forensic speaker discrimination. In: Proc. Odyssey 2006 The Speaker and Language Recognition Workshop, IEEE. doi:10.1109/ODYSSEY.2006. 248095. (Pubitemid 46938165)
63. P. Rose Technical forensic speaker recognition Comput. Speech Lang. 20 2006 159 191 10.1016/j.csl.2005.07.003
64. Rose, P., 2006c. The intrinsic forensic discriminatory power of diphthongs. In: Warren, P., Watson, C.I. (Eds.), Proc. 11th Australian Internat. Conf. on Speech Science & Technology, Auckland, pp. 64-69.
65. Rose, P., 2007. Forensic speaker discrimination with Australian English vowel acoustics. in: Trouvain, J., Barry, W.J. (Eds.), Proc. 16th International Congress on Phonetic Sciences, Saarbrücken, pp. 1817-1820.
66. P. Rose, and G.S. Morrison A response to the UK position statement on forensic speaker comparison Internat. J. Speech, Lang. Law 16 2009 139 163 10.1558/ijsll.v16i1.139
67. Rose, P., Lucy, D., Osanai, T., 2004. Linguistic-acoustic forensic speaker identification with likelihood ratios from a multivariate hierarchical random effects model: A "non-idiot's bayes" approach. In: Proc. 10th Australian Internat. Conf. on Speech Science & Technology, Sydney, pp. 492-497.
68. Rose, P., Kinoshita, Y., Alderman, T., 2006. Realistic extrinsic forensic speaker discrimination with the diphthong /a/. In: Proc. 11th Australian Internat. Conf. on Speech Science & Technology, Auckland, pp. 329-334.
69. Thiruvaran, T., Ambikairajah, E., Epps, J., 2008. FM features for automatic forensic speaker recognition. In: Proc. Interspeech 2008 Incorporating SST. International Speech Communication Association, pp. 1497-1500.
70. D.A. van Leeuwen, and N. Brümmer An introduction to application-independent evaluation of speaker recognition systems C. Müller, Speaker Classification I: Fundamentals, Features, and Methods 2007 Springer-Verlag Heidelberg, Germany 330 353 doi:10.1007/978-3-540-74200-5-19