Aural and automatic forensic speaker recognition in mismatched conditions

Abstract

In this article, we compare aural and automatic speaker recognition in the context of forensic analyses, using a Bayesian framework for the interpretation of evidence.We use perceptual tests performed by non-experts and compare their performance with that of an automatic speaker recognition system. These experiments are performed with 90 phonetically untrained subjects. Several forensic cases were simulated, using the Polyphone IPSC-02 database, varying in linguistic content and technical conditions of recording. We estimate the strength of evidence for both humans and the baseline automatic system, calculating likelihood ratios using perceptual scores for humans and log-likelihood scores for the automatic system. A methodology analogous to the Bayesian interpretation in forensic automatic speaker recognition is applied to the perceptual scores given by humans in order to estimate the strength of evidence. The degradation of the accuracy of human recognition in mismatched recording conditions is contrasted with that of the automatic system under similar recording conditions. The conditions considered are fixed telephone, cellular telephone and noisy speech in forensically realistic conditions. The perceptual cues that the human subjects use to perceive differences in voices are studied, along with their importance in different recording conditions. We observe that while automatic speaker recognition shows higher accuracy in matched conditions of training and testing, its performance degrades significantly in mismatched conditions. Aural recognition accuracy is also observed to degrade from matched conditions to mismatched conditions and in mismatched conditions, the baseline automatic systems showed comparable or slightly degraded performance compared to the aural recognition systems. The baseline automatic system with adaptation to noisy conditions showed comparable or better performance than aural recognition. The higher level perceptual cues used by human listeners in order to recognise speakers are discussed. We also discuss the possibility of increasing the accuracy of automatic systems using the perceptual cues that remain robust to mismatched recording conditions.

Forensic Identification Reporting Using A GMM Based Speaker Recognition System Dedicated to Algerian Arabic Dialect Speakers

“…Finally, we should mention that our preliminary results confirm exactly what was presented in literature on other databases especially in [10].…”

Forensic voice discrimination by lay listeners: The effect of speech type and background noise on performance

“…However, Leemann, Kolly, and Dellwo () showed that variations in suprasegmental temporal features are stable across changes in speaking style (spontaneous vs. read). Listeners may therefore be able to rely on stable, high‐level features of speech, such as mannerisms, speaking rate, and pauses when making matching decisions (Alexander, Dessimoz, Botti, & Drygajlo, ).…”

An empirical estimate of the precision of likelihood ratios from a forensic-voice-comparison system

“…Fig. 2 provides an example Tippett plot [9,10,21,[45][46][47][48][49][50][51][52] illustrating the improvements due to calibration for /a/ comparison set a. After calibration, the likelihood ratios from /i/ were more accurate than those from /a/, which were in turn more accurate than those from /e/ (compare C llr cal values in Table 1).…”