Advertisement

Biometric Evidence in Forensic Automatic Speaker Recognition

  • Andrzej Drygajlo
  • Rudolf Haraksim
Chapter
Part of the Advances in Computer Vision and Pattern Recognition book series (ACVPR)

Abstract

The goal of this chapter is to provide a methodology for calculation and interpretation of biometric evidence in forensic automatic speaker recognition (FASR). It defines processing chains for observed biometric evidence of speech (univariate and multivariate) and for calculating a likelihood ratio as the strength of evidence in the Bayesian interpretation framework. The calculation of the strength of evidence depends on the speaker models and the similarity scoring used. A processing chain chosen for this purpose is in the close relation with the hypotheses defined in the Bayesian interpretation framework. Several processing chains are proposed corresponding to the scoring and direct method, which involve univariate and multivariate speech evidence, respectively. This chapter also establishes a methodology to evaluate performance of a chosen FASR method under operating conditions of casework.

Keywords

Likelihood Ratio Gaussian Mixture Model Processing Chain Speaker Recognition Universal Background Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Aitken CGG, Lucy D (2004) Evaluation of trace evidence in the form of multivariate data. Appl Stat 53:109–122MathSciNetzbMATHGoogle Scholar
  2. 2.
    Aitken CGG, Taroni F (2004) Statistics and evaluation of evidence for forensic scientists, 2nd edn. Wiley, ChichesterGoogle Scholar
  3. 3.
    Aitken CGG, Roberts P, Jackson G (2010) Fundamentals of probability and statistical evidence in criminal proceedings. Guidance for judges, lawyers, forensic scientists and expert witnesses. Practitioner guide no 1, Royal Statistical SocietyGoogle Scholar
  4. 4.
    Alexander A (2005) Forensic automatic speaker recognition using Bayesian interpretation and statistical compensation for mismatched conditions. PhD dissertation, EPFL LausanneGoogle Scholar
  5. 5.
    Alexander A, Drygajlo A (2004) Scoring and direct methods for the interpretation of evidence in forensic speaker recognition. In: Proceedings of the International Conference on Spoken Language Processing (ICSLP), Jeju, Korea, pp 2397–2400Google Scholar
  6. 6.
    Alonso Moreno V, Drygajlo A (2012) A joint factor analysis model for handling mismatched recording conditions in forensic automatic speaker recognition. In: Proceedings of the International Conference on Biometrics (ICB 2012), New Delhi, pp 484–489Google Scholar
  7. 7.
    Arcienega M, Alexander A, Zimmermann P, Drygajlo A (2005) A Bayesian network approach combining pitch and spectral envelope features to reduce channel mismatch in speaker verification and forensic speaker recognition. In: Proceedings of INTERSPEECH 2005, Lisbon, pp 2009–2012Google Scholar
  8. 8.
    Campbell JP, Shen W, Campbell WM, Schwartz R, Bonastre J-F, Matrouf D (2009) Forensic speaker recognition: a need for caution. IEEE Signal Process Mag 26:95–103CrossRefGoogle Scholar
  9. 9.
    Champod C, Meuwly D (2000) The inference of identity in forensic speaker recognition. Speech Commun 31:193–203CrossRefGoogle Scholar
  10. 10.
    Dehak N, Kenny P, Dehak R, Dumouchel P, Ouellet P (2011) Front-end factor analysis for speaker verification. IEEE Trans Audio Speech Lang Process 19:788–798CrossRefGoogle Scholar
  11. 11.
    Drygajlo A (2007) Forensic automatic speaker recognition. IEEE Signal Process Mag 24:132–135CrossRefGoogle Scholar
  12. 12.
    Drygajlo A (2009) Statistical evaluation of biometric evidence in forensic automatic speaker recognition. In: Geradts ZJ, Franke KY, Veenman CJ (eds) Computational forensics. Springer, Berlin, pp 1–12CrossRefGoogle Scholar
  13. 13.
    Drygajlo A (2009) Forensic evidence of voice. In: Li SZ (ed) Encyclopedia of biometrics. Springer, Berlin, pp 1388–1395Google Scholar
  14. 14.
    Drygajlo A (2011) Voice: Biometric analysis and interpretation of. Wiley Encyclopedia of Forensic Science. Accessed 15 Dec 2011. doi: 10.1002/9780470061589.fsa1034
  15. 15.
    Drygajlo A (2012) Automatic speaker recognition for forensic case assessment and interpretation. In: Neustein A, Patil HA (eds) Forensic speaker recognition. Law enforcement and counter-terrorism. Springer, Berlin, pp 21–39Google Scholar
  16. 16.
    Drygajlo A, (2014) From speaker recognition to forensic speaker recognition. In: Cantoni V, Dimov D, Tistarelli M (eds) Biometric authentication: first international workshop, BIOMET 2014, Sofia, Bulgaria, Revised Selected Papers. Springer, Berlin, pp 93–104Google Scholar
  17. 17.
    Drygajlo A, Meuwly D, Alexander A (2003). Statistical methods and Bayesian interpretation of evidence in forensic automatic speaker recognition. In: Proceedings of EUROSPEECH 2003, Geneva, pp 689–692Google Scholar
  18. 18.
    Drygajlo A, Ugnat L (2012) Comparative evaluation of calibrated deterministic and statistical models for forensic automatic speaker recognition systems. Presentation at the European Academy of Forensic Science Conference (EAFS 2012), The HagueGoogle Scholar
  19. 19.
    Drygajlo A, Jessen M, Gfroerer S, Wagner I, Vermeulen J, Niemi T (2015) Methodological guidelines for best practice in forensic semiautomatic and automatic speaker recognition including guidance on the conduct of proficiency testing and collaborative exercises. ENFSI, Verlag für Polizeiwissenschaft, FrankfurtGoogle Scholar
  20. 20.
    Evett IW, Buckleton JS (1996) Statistical analysis of STR data. In: Carracedo A, Brinkmann B, Bär W (eds) Advances in forensic haemogenetics, vol 6. Springer, Berlin, pp 79–86Google Scholar
  21. 21.
    Evett IW (1998) Towards a uniform framework for reporting opinions in forensic science casework. Sci Justice 38:198–202CrossRefGoogle Scholar
  22. 22.
    Gonzalez-Rodriguez J, Drygajlo A, Ramos-Castro D, Garcia-Gomar M, Ortega-Garcia J (2006) Robust estimation, interpretation and assessment of likelihood ratios in forensic speaker recognition. Comput Speech Lang 20:331–355CrossRefGoogle Scholar
  23. 23.
    Gonzalez-Rodriguez J, Rose P, Ramos D, Toledano DT, Ortega-Garcia J (2007) Emulating DNA: rigorous quantification of evidential weight in transparent and testable forensic speaker recognition. IEEE Trans Audio Speech Lang Process 15:2104–2115CrossRefGoogle Scholar
  24. 24.
    Hansen JHL, Taufiq H (2015) Speaker recognition by machines and humans. IEEE Signal Process Mag 32:74–99CrossRefGoogle Scholar
  25. 25.
    Haraksim R (2014) Validation of likelihood ratio methods used in forensic evidence evaluation: Application in forensic fingerprints. PhD dissertation, University of Twente, EnschedeGoogle Scholar
  26. 26.
    Haraksim R, Ramos D, Meuwly D, Berger CEH (2015) Measuring coherence of computer-assisted likelihood ratio methods. Forensic Sci Int 249:123–132CrossRefGoogle Scholar
  27. 27.
    Jackson G, Jones S, Booth G, Champod C, Evett I (2006) The nature of forensic science opinion—a possible framework to guide thinking and practice in investigations and in court proceedings. Sci Justice 46:33–44CrossRefGoogle Scholar
  28. 28.
    Jackson G, Aitken C, Roberts P (2015) Case assessment and interpretation of expert evidence. Guidance for judges, lawyers, forensic scientists and expert witnesses. Practitioner guide no 4Google Scholar
  29. 29.
    Kinnunen T, Li H (2010) An overview of text-independent speaker recognition: from features to supervectors. Speech Commun 52:12–40CrossRefGoogle Scholar
  30. 30.
    Li SZ, Jain A (eds) (2015) Encyclopedia of biometrics, 2nd edn. Springer, USGoogle Scholar
  31. 31.
    Meuwly D (2001) Reconnaissance automatique de locuteurs en sciences forensiques: l’apport d’une approche automatique. PhD dissertation, University of LausanneGoogle Scholar
  32. 32.
    Meuwly D, El-Maliki M, Drygajlo A (1998) Forensic speaker recognition using Gaussian Mixture Models and a Bayesian framework. In: COST-250 workshop on speaker recognition by man and by machine: directions for forensic applications, Ankara, pp. 52–55Google Scholar
  33. 33.
    Meuwly D, Drygajlo A (2001) Forensic speaker recognition based on a Bayesian framework and Gaussian Mixture Modelling (GMM). In: Proceedings of ODYSSEY 2001, Crete, pp 145–150Google Scholar
  34. 34.
    Meuwly D, Haraksim R, Ramos D (2016) A guideline for the validation of likelihood ratio methods used for forensic evidence evaluation. To appear in Forensic Science InternationalGoogle Scholar
  35. 35.
    Morrison GS (2009) Forensic voice comparison and the paradigm shift. Sci Justice 49:298–308CrossRefGoogle Scholar
  36. 36.
    Morrison GS (2010) Forensic voice comparison. In: Freckelton I, Selby H (eds) Expert evidence (Chapter 99). Thomson Reuters, SydneyGoogle Scholar
  37. 37.
    Ramos-Castro D (2007) Forensic evaluation of the evidence using automatic speaker recognition systems. PhD dissertation, Universidad Autónoma de MadridGoogle Scholar
  38. 38.
    Ramos D, Gonzalez-Rodriguez J (2013) Reliable support: Measuring calibration of likelihood ratios. Forensic Sci Int 230:156–169CrossRefGoogle Scholar
  39. 39.
    Robertson B, Vignaux GA (1995) Interpreting evidence. Evaluating forensic science in the courtroom. Wiley, Chichester etc.Google Scholar
  40. 40.
    Rose P (2002) Forensic speaker identification. Taylor & Francis, LondonCrossRefGoogle Scholar
  41. 41.
    Rose P (2006) Technical forensic speaker recognition: evaluation, types and testing of evidence. Comput Speech Lang 20:159–191CrossRefGoogle Scholar
  42. 42.
    Van Leeuwen DA, Brümmer N (2007) An introduction to application-independent evaluation of speaker recognition systems. In: Müller C (ed) Speaker classification I: fundamentals, features, and methods. Springer, Berlin, pp 330–353CrossRefGoogle Scholar
  43. 43.
    Van Leeuwen D, Brümmer N (2013) The distribution of calibrated likelihood-ratios in speaker recognition. In: Proceedings of INTERSPEECH 2013, Lyon, pp 1619–1623Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Swiss Federal Institute of Technology Lausanne (EPFL)LausanneSwitzerland

Personalised recommendations