The Speaker Identification Ability of Blind and Sighted Listeners: An Empirical Investigation

TY - BOOK

T1 - The Speaker Identification Ability of Blind and Sighted Listeners

T2 - An Empirical Investigation

AU - Braun, Almut

N1 - Doctoral thesis, Philipps-Universität Marburg, Germany 2015.

PY - 2016

Y1 - 2016

N2 - Previous studies have shown that blind individuals outperform sighted controls in a variety of auditory tasks; however, only few studies have investigated blind listeners’ speaker identification abilities. In addition, existing studies in the area show conflicting results.The present empirical investigation with 153 blind (74 of them congenitally blind) and 153 sighted listeners is the first of its kind and scale in which long-term memory effects of blind listeners‘ speaker identification abilities are examined. The motivation for this investigation is twofold: On the one hand, also blind individuals become victims of criminal conduct and may have to testify as earwitnesses in court. The data from the present project could serve as reference material in these cases. On the other hand, blind persons may also act as forensic analysts; the Belgian police already employ blind officers for the analysis of forensic voice samples based on the assumption that visually impaired people often develop a more acute sense of hearing which enables them, for instance, to better distinguish between intonation patterns of voices (Cleemput 2007). More extensive research with regard to the speaker recognition abilities of blind listeners is strongly needed to corroborate this assumption. For the present empirical investigation, all listeners were evenly assigned to one of nine subgroups (3 x 3 design) in order to investigate the influence of two parameters with three levels, respectively, on blind and sighted listeners’ speaker identification performance. The parameters are a) time interval; i.e. a time interval of 1, 3 or 6 weeks between the first exposure to the voice to be recognized and the speaker identification task (voice lineup); and b) signal quality; i.e. voice recordings are presented either in studio quality, cell phone quality or recordings of whispering speakers are used. Half of the presented voice lineups were target-present lineups in which the previously heard target voice was included. The other half consisted of target-absent lineups which contained solely distractor voices.Blind individuals outperformed sighted listeners only under studio quality conditions (Fisher’s Exact Test: p = 0.04). Within the blind group, speaker identification performance was significantly better under studio quality conditions and significantly worse under cell phone quality conditions (Chi-square Test, df = 2, chi² = 7.158, p = 0.028). No significant performance differences occurred within the sighted group regarding the three investigated signal quality conditions. Furthermore, for blind and sighted listeners no significant performance differences were found with regard to the three investigated time intervals of 1, 3 and 6 weeks. Blind as well as sighted listeners were significantly better at picking the target voice from target-present lineups than at indicating that the target voice was absent in target-absent lineups (Fisher’s Exact Test: p < 0.001). A marginally significant trend indicates that blind listeners exhibited more correct rejections than sighted listeners in lineups in which the target voice was not present. (Fisher’s Exact Test: p = 0.058). This trend became significant when only the studio quality lineups were considered for analysis (Fisher’s Exact Test: p = 0.003). Musically trained listeners performed equally well as listeners without (current) musical training and male listeners performed equally well as female listeners. A weak but significant positive correlation between identification performance and listeners’ confidence in their decision was only found for the sighted group (rank-biserial correlation = 0.23; Wilcoxon rank sum test with continuity correction: W = 1869.5, p = 0.013). Within the blind group, no significant correlations were found between identification performance and onset or duration of blindness. Late blind listeners performed equally well as congenitally blind listeners. Implications for the field of forensic phonetics are discussed.

AB - Previous studies have shown that blind individuals outperform sighted controls in a variety of auditory tasks; however, only few studies have investigated blind listeners’ speaker identification abilities. In addition, existing studies in the area show conflicting results.The present empirical investigation with 153 blind (74 of them congenitally blind) and 153 sighted listeners is the first of its kind and scale in which long-term memory effects of blind listeners‘ speaker identification abilities are examined. The motivation for this investigation is twofold: On the one hand, also blind individuals become victims of criminal conduct and may have to testify as earwitnesses in court. The data from the present project could serve as reference material in these cases. On the other hand, blind persons may also act as forensic analysts; the Belgian police already employ blind officers for the analysis of forensic voice samples based on the assumption that visually impaired people often develop a more acute sense of hearing which enables them, for instance, to better distinguish between intonation patterns of voices (Cleemput 2007). More extensive research with regard to the speaker recognition abilities of blind listeners is strongly needed to corroborate this assumption. For the present empirical investigation, all listeners were evenly assigned to one of nine subgroups (3 x 3 design) in order to investigate the influence of two parameters with three levels, respectively, on blind and sighted listeners’ speaker identification performance. The parameters are a) time interval; i.e. a time interval of 1, 3 or 6 weeks between the first exposure to the voice to be recognized and the speaker identification task (voice lineup); and b) signal quality; i.e. voice recordings are presented either in studio quality, cell phone quality or recordings of whispering speakers are used. Half of the presented voice lineups were target-present lineups in which the previously heard target voice was included. The other half consisted of target-absent lineups which contained solely distractor voices.Blind individuals outperformed sighted listeners only under studio quality conditions (Fisher’s Exact Test: p = 0.04). Within the blind group, speaker identification performance was significantly better under studio quality conditions and significantly worse under cell phone quality conditions (Chi-square Test, df = 2, chi² = 7.158, p = 0.028). No significant performance differences occurred within the sighted group regarding the three investigated signal quality conditions. Furthermore, for blind and sighted listeners no significant performance differences were found with regard to the three investigated time intervals of 1, 3 and 6 weeks. Blind as well as sighted listeners were significantly better at picking the target voice from target-present lineups than at indicating that the target voice was absent in target-absent lineups (Fisher’s Exact Test: p < 0.001). A marginally significant trend indicates that blind listeners exhibited more correct rejections than sighted listeners in lineups in which the target voice was not present. (Fisher’s Exact Test: p = 0.058). This trend became significant when only the studio quality lineups were considered for analysis (Fisher’s Exact Test: p = 0.003). Musically trained listeners performed equally well as listeners without (current) musical training and male listeners performed equally well as female listeners. A weak but significant positive correlation between identification performance and listeners’ confidence in their decision was only found for the sighted group (rank-biserial correlation = 0.23; Wilcoxon rank sum test with continuity correction: W = 1869.5, p = 0.013). Within the blind group, no significant correlations were found between identification performance and onset or duration of blindness. Late blind listeners performed equally well as congenitally blind listeners. Implications for the field of forensic phonetics are discussed.

KW - speaker recognition

KW - speaker identification

KW - blind listeners

KW - voice lineup

KW - voice parade

KW - blindness

KW - forensic phonetics

M3 - Book

SN - 9783658151973

BT - The Speaker Identification Ability of Blind and Sighted Listeners

PB - Springer

CY - Wiesbaden

ER -