I am active in forensic voice analysis, working with my colleagues at Forensic Research Associates.  FRA are the designated voice analysts of the British government's Forensic Science Service.

FRA provide the following services:

SPEAKER COMPARISON
Speaker comparison (a more accurate term than the more familiar 'speaker identification') involves analysis, by ear and computer, of two recorded speech samples.  One sample is the speech of an unknown perpetrator (e.g. threatening calls; blackmail; fraudulent calls; nuisance calls; hoax calls to emergency services).  The other sample is the speech of a suspect.  We report on the degree of consistency and distinctiveness in the samples.  It is important to bear in mind that voices have not been proven to be unique; therefore voice comparison can never result in 100% positive identification.

CONTENT DETERMINATION
This involves analysis and transcription of a recording where the content is uncertain, due to factors such as poor recording quality.  We produce not only a detailed phonetic transcript, but also an analysis of other acoustic material in the recording.

SPEAKER PROFILING
Where there is a recording of a criminal’s speech but no suspect has been identified, we can compile a profile of the speaker based on his/her speech.

VOICE LINEUPS
Where a witness has heard a perpetrator’s voice but not seen their face, it is possible to construct a lineup of voice recordings parallel to visual ID parades.

We offer forensic consultancy for both prosecution and defence, and for clients outside a court of law.  The findings of our analysis are detailed in a report and formulated as a statement, both of which can be subsequently used as expert evidence in a court case.