Speaker Comparison

This 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.

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. 

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.

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.