Paper short abstract:

This paper examines how judges have decided on the admissibility of forensic voice and speech evidence and the impact that these decisions have for experts working in this field.

Paper long abstract:

Advances in audio technology have turned courtrooms into noisy places. Police wiretaps, taped 911 calls, and messages left on answering machines are just a few examples of the kinds of voice recordings that have made their way into criminal and civil proceedings. This has been fuelled by a growing industry aimed at providing lawyers with the technical resources needed to transform these 'raw' auditory materials into legal evidence that can help win a case. Not only can these audio forensic experts enhance, authenticate, and render intelligible analogue or digital recordings for courtroom presentation, but they can also analyse and, in some instances, identify the person to whom the voice and speech belongs. However, as a relatively new field of forensics, this type of analysis - commonly referred to as 'speaker identification' - has been admitted as a credible technique in some jurisdictions, but not in others. This paper examine how judges have decided on the admissibility of forensic voice and speech evidence and the impact that these decisions have for those working in this field. Rather than seeing law as a separate and distinct field from science and technology, I use the adversarial setting of the courtroom as a site where scientific facts and technological artefacts are 'black boxed' or opened up for scrutiny. Drawing on theoretical tools from Actor-Network Theory (ANT), I document the work that audio forensic experts have done - inside and outside the courtroom - to gain legitimacy and credibility for their field and the tools and techniques they employ.