Forensic Audio Examinations
Below is a list of the forensic audio examination services that are preformed by Forensic Science Services, Inc. These are generalizations and should you require any additional info on the services we offer please feel free to contact us durring our regular business hours.
General sound analysis
Spectrographic analysis of the recorded sounds in order to determine the:
- The origin of the sound, i.e. sounds of animals and birds, articulated and unarticulated human sounds, jets, automobiles, bells and so on..
- The distance of the sound source.
- The echoes and reverberation.
- The sequence of recorded sounds.
- Background noise profile.
Voice Identification
The spectrographic voice identification analysis has two steps. The sound of speech is first transformed into a three dimensional (time - frequency - volume) graphic pictures which do reveal numerous acoustical features of an individual’s voice. The second step involves the pattern comparison of the same phrases/sentences from the unknown sample and the suspect’s sample. The results of analysis are expressed as:
- Probably the same speaker (high level of confidence).
- Possibly the same speaker (intermediate level of confidence).
- Inconclusive(due to the insufficient number of comparison words, poor quality of recordings, too high variability of the voice, possible disguise).
- Possibly not same speaker (intermediate level of confidence).
- Probably not the same speaker (high level of confidence).
The results depend on quality of recordings, the total number of comparison words, speakers’ condition, and individual speakers’ voice variability. There is a requirement for a minimum number of 20 comparison words in a ‘connected speech’. The suspect should provide the comparison sample by reading three times the transcript of the unknown voice sample.
Gunshot sound analysis
The spectrographic analysis of the submitted recording identifies and locates gunshot sounds. It may be determined if more then one gun was involved in shooting. The sequence of shots may be determined.
Testing for alterations of audio recordings
Any changes to the audio content subsequent to the original recording constitutes alteration regardless if it is done
on purpose (tempering or editing) or by accident.On the original recording the alteration may be done on purpose or by accident. On the copy alteration is usually done on purpose by adding something to the original content or omitting something from the original content. The testing for alteration is done by waveform, spectrographic and magnetic development analysis.
Enhancement of audio recordings
The excessive noise can obstruct the recorded speech thus reducing the intelligibility. Spectrographic analysis is used to determine the noise profile. Subsequently digital filtering is used to suppress the noise without significantly affecting the recorded speech. If the recorded speech signal is at very low level the noise suppression cannot improve it.
Transcription of audio recordings
Most of the transcriptions are done by court reporters and clerks and specialized agencies. These transcripts are not always accurate for court purposes in particular when intelligibility is low or when there are numerous breaks in the recording induced by VAR (Voice Activated Recording). On each break a part of the conversation (of undetermined length) may be dropped from the recording. For that reason it is important that new line of transcript is started on each break in the recording. By ignoring the missing part of the recording two parts of different sentences or paragraphs are
spliced together producing something (if meaningful) that was not a part of the original conversation. These types of breaks may have an audible click, however often there is no audible signal associated with them.
Transcription of a continuous recording may have the breaks in the form of unintelligible parts. The notation “(unintelligible)” or “(inaudible)” used in the middle of the sentence may be quote misleading. If the unintelligible words can be counted that should be noted in the brackets, for example “(unintelligible, 3 words)”. If the number of unintelligible words cannot be determined, time in seconds for unintelligible interval should be entered, for example, “(unintelligible, 4s)” and a new line/paragraph should be started.
Thus all transcriptions (and their source recordings) made for court purposes should be evaluated for accuracy by a forensic audio expert.
Top of Page |
