Transcription of indistinct forensic recordings : Problems and solutions from the perspective of phonetic science

Resumo

Covert recordings (speech captured on audio or video without the knowledge of the speakers) can provide powerful forensic evidence. Unfortunately, however, since it is dicult to control their recording conditions, covert recordings
are often indistinct, to the extent the speech is unintelligible to listeners who do not already have knowledge or expectations about their content. This raises the question of how to present the speech evidence in court so that the trier of fact (jury, magistrate, judge, etc.) can make use of the information contained in the recording. In many jurisdictions, the answer is to have a transcript made by those who know the context (typically police working on the case), and provided to the trier of fact as an aid to perception of the speech. The present article outlines several problems with this approach, then suggests some solutions that allow maximal value of the intelligence contained in covert recordings, while reducing the risk of injustice through biased perception of indistinct audio. A key part of the suggested solution is to make a clear distinction between investigative and evidentiary uses of indistinct covert recordings, and to ensure that transcripts of evidentiary recordings be produced by professional transcribers independent of the case.