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

Philip Rose

    Research output: Contribution to conferencePaper

    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 languageEnglish
    Pages5900-5903
    DOIs
    Publication statusPublished - 2011
    EventInternational Conference on Acoustics, Speech and Signal Processing (ICASSP 2011) - Prague Czech Republic
    Duration: 1 Jan 2011 → …

    Conference

    ConferenceInternational Conference on Acoustics, Speech and Signal Processing (ICASSP 2011)
    Period1/01/11 → …

    Fingerprint Dive into the research topics of 'Forensic voice comparison with secular shibboleths - A hybrid fused GMM-multivariate likelihood ratio-based approach using alveolo-palatal fricative cepstral spectra'. Together they form a unique fingerprint.

    Cite this