Effect of combining auditory features with acoustic parameters on the probability scales in forensic speech recognition
- 1Physics Division, Forensic Science Laboratory, Madhuban, Karnal, Haryana, India
- 2Physics Division, State Forensic Science Laboratory, Delhi, India
- 3Dept. of Applied Physics, Guru Jambeshwar University of Science and Technology, Hisar, India
Res. J. Forensic Sci., Volume 6, Issue (3), Pages 1-6, April,29 (2018)
Attempts to understand the phenomenon and mechanism of speech sounds led humans to discover visual representation of them in terms of frequency-time graphs which helps them to understand acoustic parameters, which give the voice of humans ‘uniqueness’. One of the emerging field of forensic science is using acoustic parameters and auditory features to perform speaker identification test by comparing known to unknown samples. In this paper, we consider two sets of speech samples, questioned and known specimen speech sample data base obtained from the actual crime cases. The two speech samples underwent to the method of auditory analysis and spectrographic analysis. The percentage of similarities between the unknown sample (Questioned) and the known sample were ascertained by formant frequencies, and for numerical values assigned to the auditory features. Bayes’ Theorem was used to combine objective probability obtained from the acoustic parameters and subjective probability obtained from the auditory features. These values computed to correlate with one of the nine probability scales with the help of the software programs developed by the authors. This study reveals how the resultant probability changes, if auditory features were also taken into account along with that of the acoustic parameters while calculating the final similarity percentage.
- Holmgren G.L. (1967)., Physial and Psychological Correlates of Speaker Recognition., Journal of Speech, Language, and Hearing Research, 10, 57-66. doi:10.1044/jshr.1001.57.
- Endress W., Bambach W. and Flosser G. (1971)., Voice Spectrograms as a function of Age, Voice Disguise and Voice Imitation., Journal of Acoustical Society of America, 49, 1842-1848. https://doi.org/10.1121/1.1912589.
- Tosi O., Oyer M., Lashbrock W., Pedey C., Nicol J. and Nash E. (1972)., Experiment on Voice Identification., Journal of Acoustical Society of America, 51, 2030-2043.https://doi.org/10.1121/1.1913064.
- Wolf J.J. (1972)., Efficient acoustic parameters for speaker recognition., Journal of Acoustical Society of America, 51(6), 2044-2057. https://doi.org/10.1121/1.1913065
- Hazen B. (1973)., Effects of differing phonetic contexts on spectrographic speaker identification., The Journal of the Acoustical Society of America, 54(3), 650-660. https://doi.org/10.1121/1.1913645.
- Samber M.R. (1975)., Selection of Acoustic Features for Speaker Identification., IEEE Transactions on Acoustic, Speech and Signal Processing, 23(2), 176-182. 10.1109/TASSP.1975.1162664.
- Aitken C.G.G. (2013)., Statistical Interpretation of Evidence/Bayesian Analysis., University of Edinburgh, Edinburgh, UK, 173-179. ISBN: 978-0-12-800647-4.
- Meuwly D. and Drygazlo A. (2001)., Forensic Speaker Recognition based on Bayesian Framework and Gaussian Mixture Modelling (GMM)., The Speaker Recognition Workshop Crete, Greece, 18-22, 145-150.
- Kinoshita Y. (2002)., Use of Likelihood Ratio and Bayesian Approach in Forensic Speaker Identification., School of Languages and International Education, University of Canberra. Australian Speech Science and Technology Association Inc., 297-302.
- An Introduction to Forensic Speaker Identification Procedure (2005)., Advance Interactive Training Course on Forensic Speaker Recognition., CBI Bulletin, Directorate of Forensic Science, Ministry of Home Affairs, Govt. of India, XIII(1).
- Besson O., Dobigeon N. and Tourneret J.Y. (2014)., Joint Bayesian estimation of close subspaces from noisy measurements., IEEE Signal Processing Letters, 21(2), 168-171. 10.1109/LSP.2013.2296138.
- Mathu R.S., Chaudhary S.K. and Vyas J.M. (2016)., Effect of Disguise on Fundamental Frequency of Voice., Journal of Forensic Research: Open Access, 7(3). ISSN: 2157-7145 JFR, doi:10.4172/2157-7145.1000327.
- Bhall B., Singh C.P., Dhar R. and Soni R. (2016)., Auditory and Acoustic Features from Clue-Words Sets for Forensic Speaker Identification and its Correlation with Probability Scales., Journal of Forensic Research: Open Access, 7. ISSN: 2157-7145 JFR, doi:10.4172/2157-7145.1000338.