On the Use of A utom atic Speaker Verification System s in Forensic Casework

In forensic applications of speaker recognition it is nec­ essary to be able to specify a confidence level for a de­ cision th a t two sets of recordings have been produced by the same speaker (or by different speakers). Foren­ sic phoneticians are sometimes incriminated because they find it impossible to provide ’hard’ estimates of the confidence level of an expert opinion. In this pa­ per it is investigated to what extent the problem can be solved by deploying autom atic speaker verification algorithms, to work alone or to support the work of forensic phoneticians. It is shown th a t for several rea­ sons hard estimates of the confidence of an opinion in a specific case cannot be provided by autom atic speaker verification either.