International practices in forensic speaker comparisons

Abstract

A survey relating to current practices in forensic speaker comparison testing was recently undertaken of 39 laboratories and individual practitioners across 23 countries. Questions were organised around a number of themes, including the preliminary assessment and preparation of case materials, the checking of analysts' work, frameworks used for the expression of conclusions, the use of automatic speaker recognition systems, the use of reference populations, and awareness of cognitive bias. Developmental trends in this area of forensic speech science are established by comparing responses to the present survey with those to the authors' earlier survey published in 2011.

A forensic phonetic investigation of regional variation and accommodation in West Yorkshire

“…A range of different methods of analysis can be used for FSCs, with the combined auditory and acoustic phonetic approach being the most prevalent across the world and the only approach used in the UK for evidential purposes (Gold & French, 2011;Morrison et al, 2016).…”

“…In a survey of international practices in FSC casework, Gold & French (2011) found that all respondents reported analysing vowel and consonant sounds in the course of their FSC examinations. With regards to the specific analytical approaches taken by these experts, it was reported that 97% of experts examined formants when analysing vowels and of those undertaking formant examinations, it was reported that "all measure the second resonance (F2); 87% of respondents reported measuring F1 and an equal percentage reported measuring F3" (2011, p. 300).…”

“…In line with methods commonly applied in FSC analyses in the UK (Gold & French, 2011, a combined auditory and acoustic approach was taken to analyse both segmental (vowel and consonant realisations) and supra-segmental (pitch and voice quality) parameters in the speech of participant #059 recorded during three tasks. These specific parameters were chosen as they are all routinely analysed in FSC casework.…”

The effect of word class on speaker-dependent information in the Standard Dutch vowel /aː/

“…The outcome of such research feeds into practice; in a forensic speaker comparison (FSC), one or more disputed speech recordings are compared with one or more reference recordings of a suspect in order to investigate whether the recordings might have been produced by the same or by different speakers. To make these comparisons, several methods are in use across the world, varying from auditory examination to acoustic-phonetic measurement to automatic speaker recognition (Morrison et al, 2016;Gold and French, 2019). It is theoretically important to not only compare the disputed and suspect samples to each other, and to thus assess their similarity, but to evaluate the likelihood of this similarity against background population information, to thus assess the typicality of the features under study.…”

“…Moreover, not all types of speech features, such as word use, can be included in ASR. Currently, in international surveys amongst respondents carrying out FSC the majority used an auditory/acoustic-phonetic approach (Morrison et al, 2016;Gold and French, 2019): acousticphonetic features are measured in the different speech samples, and used to assess how similar these features are between the suspect and the disputed speaker, relative to how typical they are PREPRINT VERSION of article published in JASA 14-10-2020, https://doi.org/10.1121/10.0002173 of speakers in general. Little is known, however, about how the speaker information carried by acoustic-phonetic features depends on the linguistic context from which it is sampled.…”