Vincent Hughes

Vincent Hughes
The University of York · Department of Language and Linguistic Science

Doctor of Philosophy

About

74
Publications
15,647
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
410
Citations
Citations since 2017
49 Research Items
353 Citations
20172018201920202021202220230102030405060
20172018201920202021202220230102030405060
20172018201920202021202220230102030405060
20172018201920202021202220230102030405060
Additional affiliations
February 2015 - December 2015
The University of York
Position
  • PostDoc Position

Publications

Publications (74)
Article
Full-text available
Polyfunctional words (often categorized as discourse-pragmatic features) vary in form depending on their discourse functions, prosodic contexts (such as utterance position), and usage across social groups. This study applies dynamic formant analysis to describe the detailed phonetic variation of two polyfunctional words, like and yeah, in a corpus...
Conference Paper
Full-text available
In data-driven forensic voice comparison (FVC), empirical testing of a system is an essential step to demonstrate validity and reliability. Numerous studies have focused on improving system validity, while studies of reliability are comparatively limited. In the present study, simulated scores were generated from i-vector and GMM-UBM automatic spea...
Presentation
Full-text available
A particularly important issue in forensic voice comparison (FVC) is the lack of direct correspondence in the content of different recordings. That is, recordings are unlikely to share many of the same words. Therefore, a frequently used word (or other feature) in naturally occurring speech is of value to the FVC practitioner because it permits dir...
Presentation
Full-text available
A particularly important issue in forensic voice comparison (FVC) is the lack of direct correspondence in the content of different recordings. That is, recordings are unlikely to share many of the same words. Therefore, a frequently used word (or other feature) in naturally occurring speech may be of value to the FVC practitioner. To examine the fo...
Poster
Full-text available
In data-driven forensic voice comparison (FVC), empirical testing of a system is an essential step to demonstrate validity (i.e. how well the system performs the task) and reliability (i.e. whether the system would yield the same result if the analysis were repeated). The present study focuses on system reliability, aiming to reduce the degree of u...
Poster
Full-text available
Forensic validation tends to focus on the overall performance of methods under casework conditions. This implicitly focuses the expert’s attention on discriminability (see Smith & Neal 2021), with different methods chosen, or decisions made, based low values for the validity metric used. In this paper, we argue against this view. Rather, we believe...
Poster
Full-text available
Discourse-pragmatic variables (DPVs) are complex polyfunctional linguistic items which serve to “express speaker stance; to guide utterance interpretation and to structure discourse” (Pichler, 2013 p.4). They include markers (like, yeah, just), phrases (you know, I mean), interjections (ah, oh) and longer strings. Their structural and interactional...
Article
Full-text available
Imagine you are a police officer investigating a crime in which somebody has made a threatening phone call. There may not be any physical evidence like DNA or fingerprints, but the person’s voice is captured on a recording. Experts can analyze the voice recording and compare it with the voice of a known suspect to find out how likely it is that the...
Article
The likelihood ratio (LR) framework has been widely adopted in voice (and other forensic) evidence evaluation. However, in developing any forensic comparison system, it is necessary to make subjective and pragmatic decisions, which in turn may affect the results that system produces. One such decision relates to not only the size of the samples use...
Conference Paper
In data-driven forensic voice comparison, sample size is an issue which can have substantial effects on system output. Numerous calibration methods have been developed and some have been proposed as solutions to sample size issues. In this paper, we test four calibration methods (i.e. logistic regression, regularised logistic regression, Bayesian m...
Presentation
The current study simulated skewed scores from an empirical study (Wang et al., 2019) to test the susceptibility of four calibration methods (i.e. logistic regression (Brümmer et al., 2007), regularised logistic (rlogistic) regression (Morrison & Poh, 2018), empirical lower and upper bound (ELUB, Vergeer et al., 2016) and Bayesian model (Brümmer &...
Article
Since the 1960s, there have been calls for forensic voice comparison to be empirically validated under casework conditions. Since around 2000, there have been an increasing number of researchers and practitioners who conduct forensic-voice-comparison research and casework within the likelihood-ratio framework. In recent years, this community of res...
Conference Paper
Full-text available
A significant question for forensic voice comparison, and for speaker recognition more generally, is the extent to which different input features capture complementary speaker-specific information. Understanding complementarity allows us to make predictions about how combining methods using different features may produce better overall performance....
Conference Paper
Full-text available
A significant question for forensic voice comparison, and for speaker recognition more generally, is the extent to which different input features capture complementary speaker-specific information. Understanding complementarity allows us to make predictions about how combining methods using different features may produce better overall performance....
Article
Automatic Speaker Recognition (ASR) systems are designed to provide the user with statistics relating to the similarity of two or more speech samples and to the typicality of those shared features in the wider population. When an ASR system is used as part of a forensic investigation, the user must decide what counts as the appropriate ‘wider popul...
Article
Full-text available
Among phoneticians, the Vocal Profile Analysis (VPA) is one of the most widely used methods for the componential assessment of voice quality. Whether the ultimate goal of the VPA evaluation is the comparative description of languages or the characterization of an individual speaker, the VPA protocol shows great potential for different research area...
Article
Within the field of forensic voice comparison (FVC), there is growing pressure for experts to demonstrate the validity and reliability of the conclusions they reach in casework. One benefit of a fully data-driven approach that utilises databases of speakers to compute numerical likelihood ratios (LRs) is that it is possible to estimate validity and...
Article
Forensic speech science is the application of speech analysis methods to forensic recordings; in many jurisdictions this is predominantly the application of sociophonetics. Sociophonetics and forensic speech science have developed as independent research areas with their own aims, methodologies and identities, and the gap between the fields has arg...
Poster
Full-text available
The likelihood-ratio (LR) framework has been employed in many forensic voice comparison studies to test the speaker-specificity of individual vowels and phonetic sequences (e.g. Morrison, 2009; Zhang, Morrison & Thiruvaran, 2011; Rose & Wang, 2016). Other studies have looked at the effects of sample size on LR outputs (e.g. Hughes, 2014, Hughes & F...
Article
Full-text available
This paper contributes to the ongoing discussion about the distinction between observations and propositions in forensic inference, with a specific focus on forensic voice comparison casework conducted in the UK. We outline both linguistic and legal issues which make the evaluation of voice evidence and the refinement of propositions problematic in...
Article
Full-text available
This study considers the role of different cognitive units in sound change: phonemes, contextual variants and words. We examine /u/‐fronting and /j/‐dropping in data from three generations of Derby English speakers. We analyze dynamic formant data and auditory judgments, using mixed effects regression methods, including generalized additive mixed m...
Article
Full-text available
Semi-automatic systems based on traditional linguistic-phonetic features are increasingly being used for forensic voice comparison (FVC) casework. In this paper, we examine the stability of the output of a semi-automatic system, based on the long-term formant distributions (LTFDs) of F1, F2, and F3, as the channel quality of the input recordings de...
Article
There is now abundant evidence that phonetic forms are shaped by probabilistic effects reflecting predictability or informativity . We outline a number of challenges for such work, where theoretical claims are often based on small differences in acoustic measurements, or interpretations of small statistical effect sizes. We outline caveats about th...
Article
Full-text available
Cluster analysis is a way of classifying individual cases into groups on the basis of their similarity in respect of a defined set of variables. In this investigation, our cases are 99 male speakers of Southern Standard British English, matched for age and occupation. The classifying variables are the perceptual ratings given by three phoneticians...
Poster
Full-text available
Cluster analysis is a way of classifying individual cases into groups on the basis of their similarity in respect of a defined set of variables. In this investigation, our cases are 99 male speakers of Southern Standard British English, matched for age (18-25) and occupation (students). The classifying variables are the perceptual ratings given by...
Article
Full-text available
The likelihood ratio (LR) is now widely accepted as the appropriate framework for evaluating expert evidence. However, an empirical issue in forensic voice comparison is the number of speakers required to generate robust LR output and adequately test system performance. In this study, Monte Carlo simulations were used to synthesise temporal midpoin...
Poster
Full-text available
Most previous studies on speaker identification suggest that native listeners have an advantage over non-natives (Köster and Schiller 1997; Perrachione et al. 2009). Other investigations, however, suggest that it is possible to identify voices successfully when stimuli are random phonemes with no semantic meaning and not belonging to any language (...
Conference Paper
Full-text available
In forensic voice comparison, automatic, semi-automatic and phonetic methods are available for evaluating voice evidence. Across the world, the phonetic approach is used predominantly in casework. This is due, in part, to the 'black box' perception of automatic systems and the lack of direct links between the features extracted and the underlying p...
Poster
Full-text available
In forensic voice comparison (FVC), the strength of the evidence is dependent not only on the similarity between the suspect and offender with regard to the features present in the voices, but also the typicality of these features within the wider, relevant population (Aitken and Taroni 2004). Despite calls across forensic science (and from the UK...
Conference Paper
Full-text available
In forensic voice comparison, there is increasing focus on the integration of automatic and phonetic methods to improve the validity and reliability of voice evidence to the courts. In line with this, we present a comparison of long-term measures of the speech signal to assess the extent to which they capture complementary speaker-specific informat...
Conference Paper
Full-text available
In forensic voice comparison, it is essential to consider not only the similarity between samples, but also the typicality of the evidence in the relevant population. This is explicit within the likelihood ratio (LR) framework. A significant issue, however, is the definition of the relevant population. This paper explores the complexity of populati...
Article
Full-text available
Spanish and English naïve listeners judged the similarity of 5 pairs of Spanish speaking identical twins. Listeners rated speaker similarity in a comparable way irrespective of their L1. This is of forensic relevance in non-native earwitness evidence, as it suggests that similar listening strategies operate (i.e. holistic approach to voice quality)...
Conference Paper
Full-text available
We assess the potential improvement in the performance of MFCC-based automatic speaker recognition (ASR) systems with the inclusion of linguistic-phonetic information. Likelihood ratios were computed using MFCCs and the formant trajectories and durations of the hesitation marker um, extracted from recordings of male standard southern British Englis...
Article
Full-text available
This study investigates the evidential value of filled pauses (FPs, i.e. um, uh) as variables in forensic voice comparison. FPs for 60 young male speakers of standard southern British English were analysed. The following acoustic properties were analysed: midpoint frequencies of the first three formants in the vocalic portion; 'dynamic' characteris...
Poster
Full-text available
Forensic voice comparison (FVC) involves analysing the speech of an unknown offender and a known suspect to aid the court in determining whether the voices belong to the same or different speakers. In the UK, FVC casework is conducted using a combination of auditory and acoustic linguistic-phonetic analysis. For forensic phoneticians voice quality...
Article
We assess the potential improvement in the performance of MFCC-based automatic speaker recognition (ASR) systems with the inclusion of linguistic-phonetic information. Likelihood ratios were computed using MFCCs and the formant trajectories and durations of the hesitation marker um, extracted from recordings of male standard southern British Englis...
Conference Paper
Full-text available
This paper explores methods for characterising individual voices using different vocal tract output measures. Mel frequency cepstral coefficients (MFCCs), long-term formant distributions (LTFDs) and scores based on vocal profile analysis (VPA) of long-term supralaryngeal settings were extracted from the same corpus of recordings. Distances between...
Conference Paper
Full-text available
In likelihood ratio (LR)-based forensic speaker comparison it is essential to consider correlations between parameters to accurately estimate the overall strength of the evidence. Current approaches attempt to deal with correlations after the computation of LRs (back-end processing). This paper explores alternative, front-end techniques, which cons...
Conference Paper
Full-text available
We analyse dynamic formant data from a corpus of Derby English spanning three generations, focusing on the relationship between yod-dropping and GOOSE (/uː/)-fronting. Derby English exhibits a stable but variable pattern of yod-dropping in post-coronal position (e.g. new [njuː]~[nuː]), and an ongoing process of /uː/-fronting. The degree of /uː/-fro...
Article
Full-text available
The research presented in this paper builds upon a previous pilot study (Gold and Hughes 2012). This paper explores the correlation structure of speech parameters from a sociolinguistically homogeneous set of male speakers of Southern Standard British English using a series of segmental, suprasegmental and linguistic parameters. Data was extracted...
Article
Full-text available
This study assesses the extent to which likelihood ratios (LRs) are affected by analyst decisions regarding the number of reference speakers, number of tokens per speaker, and degree of linguistic match between the suspect-offender and the reference data. Using F1 and F2 trajectories from spontaneous /uː/ vowels, LRs were computed against a referen...
Conference Paper
Full-text available
This paper investigates the effects of different definitions of the relevant population with regard to regional background in LR-based forensic voice comparison using cepstral coefficients (CCs). GMM-UBM calibrated log likelihood ratios (LLRs) are computed using training and reference data in three conditions: (a) Matched, (b) Mismatched and (c) Mi...
Article
Full-text available
In forensic voice comparison, the expert is typically instructed to compare the voices in a pair of offender and suspect samples. To appropriately evaluate the strength of such evidence, it is necessary to consider both the similarity between the samples and their typicality in the wider, relevant population. This paper considers the effects of dif...
Thesis
Full-text available
Within the field of forensic speech science there is increasing acceptance of the likelihood ratio (LR) as the logically and legally correct framework for evaluating forensic voice comparison (FVC) evidence. However, only a small proportion of experts currently use the numerical LR in casework. This is due primarily to the difficulties involved in...
Article
Across forensic speech science, the likelihood ratio (LR) is increasingly becoming accepted as the logically and legally correct framework for the expression of expert conclusions. However, there remain a number of theoretical and practical shortcomings in the procedures applied for computing LRs based on speech evidence. In this paper we review ho...
Article
Full-text available
This paper focuses on the way that local social indexicality interacts with principles of vowel change. A combination of real and apparent time data from the northern English dialect of York indicate fronting of tense back vowels in the goat and goose lexical sets, and diphthongization of traditionally monophthongal mid-vowels in the face and goat...
Article
Full-text available
Research into the forensic performance of individual formants has offered considerable evidence to support the traditional acoustic-phonetic view that whilst F1 and F2 encode broad phonetic contrast, higher formants may offer greater speaker-discriminatory potential (Peterson 1959, Ladefoged 2006, Clermont and Mokhtari 1998, Rose 2002). However, th...
Article
Full-text available
This paper explores the effects of variability in the amount of reference data used in quantifying the strength of speech evidence using numerical likelihood ratios (LRs). Monte Carlo simulations (MCS) are performed to generate synthetic data from a sample of existing raw local articulation rate (AR) data. LRs are computed as the number of referenc...
Thesis
The likelihood ratio (LR) is the “logically and legally correct” (Rose and Morrison 2009:143) framework for the estimation of strength-of-evidence under two competing hypotheses. In forensic voice comparison these considerations are reduced to the similarity and typicality of features across a pair of suspect and offender samples. However, typicali...

Network

Cited By

Projects

Projects (3)
Project
The aim of the project is to compare the performance of different methods for forensic voice (or speaker) comparison – from linguistics and phonetics, acoustics, and automatic speaker recognition (ASR) – on the same set of recordings. We will explore the performance of the methods to assess their relative strengths, the consistency of their results and error patterns, and thus the potential for different methods to be integrated into a single framework. The ultimate aim is to improve methods in forensic voice comparison, taking a major step towards the development of a methodology that is more transparent, validated, and replicable. This outcome will benefit academics and forensic practitioners, the public, judicial systems, and investigative/security agencies.