Abstract
The present paper describes a statistical modeling procedure that was developed to account for the fact that, in a forensic voice comparison analysis conducted for a particular case, there was a long time interval between when the questioned- and known-speaker recordings were made (six years), but in the sample of the relevant population used for training and testing the forensic voice comparison system there was a short interval (hours to days) between when each of multiple recordings of each speaker was made. The present paper also includes results of empirical validation of the procedure. Although based on a particular case, the procedure has potential for wider application given that relatively long time intervals between the recording of questioned and known speakers are not uncommon in casework.
Highlights
The present paper describes a statistical modeling procedure that was developed to account for the following situation in a forensic voice comparison analysis
The procedure we describe in the present paper has been further developed and refined since its application in the original case, and, similar, the forensic voice comparison system used for the research reported here is not identical to that used for the case
We have illustrated that if there is a large time-interval between when the questioned-speaker recording is made and when the knownspeaker recording is made, but one trains and tests using pairs of samespeaker recordings that have a short time-interval between when each member of each pair was made, :
Summary
The present paper describes a statistical modeling procedure that was developed to account for the following situation in a forensic voice comparison analysis:1 There is a long time interval (e.g., years) between when the questioned-speaker recording was made and when the knownspeaker recording was made.2 In the sample of the relevant population used for training and testing the forensic voice comparison system there is a short interval (e.g., hours or days) between when each of multiple recordings of each speaker was made. originally developed for a particular case, the procedure has potential for wider application; in the first author’s experience it is not uncommon to receive requests to conduct casework involving2 “Questioned speaker” and “known speaker” are used as abbreviations for the speaker of questioned identity and the speaker of known identity respectively.time intervals of several years between when the questioned-speaker and known-speaker recordings were made. The present paper describes a statistical modeling procedure that was developed to account for the following situation in a forensic voice comparison analysis:. In the sample of the relevant population used for training and testing the forensic voice comparison system there is a short interval (e.g., hours or days) between when each of multiple recordings of each speaker was made. Time intervals of several years between when the questioned-speaker and known-speaker recordings were made. The procedure we describe in the present paper has been further developed and refined since its application in the original case, and, similar, the forensic voice comparison system used for the research reported here is not identical to that used for the case. The description immediately below of the case conditions and of the system used to conduct the analysis in the case is deliberately terse
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have