Collection-image

TOPICAL COLLECTIONS

Robust Acoustic, Segmental and Prosodic Parameters in the Task of Speaker Comparison in Forensic Phonetics: Native and Foreign Languages
Submit Abstract

Guest Editor

Sandra Madureira

Sandra Madureira

Pontifical Catholic University of São Paulo

Sandra Madureira holds a degree in Portuguese/English Languages and a specialization in Linguistics from the Catholic University of Santos. She also obtained a master's and doctorate in Applied Linguistics and Language Studies from the Pontifical Catholic University of São Paulo (PUCSP). She completed a postdoctoral internship at the University of Campinas. Currently, she serves as a full professor at the Department of Linguistics at PUCSP and as a professor at the Postgraduate Studies Program in Applied Linguistics at PUCSP. Sandra has extensive experience in the field of Linguistics and Applied Linguistics. She is the founder and leader of the Research Group on Speech Studies (GeFALA) and the founder and coordinator of the Integrated Laboratory of Acoustic Analysis and Cognition (LIAAC). Additionally, she is the editor of the journal Intercâmbio em Linguística Aplicada at the Pontifical Catholic University of São Paulo and serves on the editorial committees of scientific journals. Her research in Experimental Phonetics encompasses various areas, including Speech Expression, Sound Symbolism, Voice Qualities, Intonation, L2 Sound Acquisition, Phonetic Variation, and Speech Disorders.

Collection Overview

In Forensic Phonetics, the two most relevant acoustic-related analyses are: (1) analyzing the similarity between voices recorded in different contexts (questioned and reference samples) and (2) analyzing the typicality of these voices within the sociolinguistic community of the individuals under criminal investigation. Both analyses require robust acoustic parameters that demonstrate significant voice similarities only when the samples come from the same individual. Additionally, for a successful Speaker Comparison task, the acoustic parameter values should be atypical, making them distinguishable from those of other individuals within the same linguistic community. This necessitates research on acoustic measures with low intra-speaker variability and high inter-speaker variability, which justifies this Methods Collection proposal. The study of novel measures derived from both segmental and prosodic domains, particularly those that address robustness across different speaking styles, languages, attitudes, and emotions, will be highly appreciated. Furthermore, research on the connection between perceptual protocols (e.g., Laver's VPA) and acoustic analysis is also encouraged.

Articles

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
9:09

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

0 Views

•

2024

•