Forensic voice ID is regarded as the procedure of analyzing as well as comparing different voice samples to determine a speaker’s identity. And this process is accomplished through various techniques that include spectrographic analysis. In spectrographic analysis, a visual representation of the acoustic profile of the speaker is formed besides an aural-linguistic analysis, and it involves measuring some aspects of a speaker’s voice, like prosody and articulations. In the matter of forensic investigation, voice identification becomes useful for comparing various voice samples that are gotten from a suspect along with samples that are received from other sources like police interrogations or conversations.
Kinds of voice identification
Forensic voice identification is classed into a couple of types:
Earwitness identification – Earwitness identification is utilized when a witness hears the voice of the speaker. The reliability of an earwitness is similar to the identification of an eyewitness as it is dependent on several factors. Some of them are the conditions of the witness when he hears the voice and his age, race, and sex. How much time the witness was hearing the speaker, too, is an important factor. Again, the time that has passed between the identification and the witnessed event is vital too. After evaluating all these situations, experts become capable of assessing the reliability of the witness.
SPID or Forensic speaker identification – SPID is utilized when forensic experts get access to a voice recording that an unknown person makes. The experts use that recording to determine whether the voice eliminates or matches a suspect. This process is known as Aural-Acoustic Speaker Identification. Since the early 1990s, this testing method has been utilized for meeting the Frye & Daubert standards in the courtrooms of the US. This determination seems to be hugely dependable, and even the Intl. Assn. of Identification accepts this methodology.
Recording in two worlds
While recording in an ideal world, when you utilize a recording device, it brings pristine audio. Hence, the speech becomes audible clearly. People can use the recordings for their purpose.
And when you record in the real world, you will come across some issues that would interrupt masking your audio, like street or background sounds, interference, low levels of recordings, hum, buzz, and hiss. At times, the masking turns out to be intentional too. So, many filters are customized to correct all these issues. And most often, it enables the recordings to get utilized for investigation purposes.
The method of voice analysis
In forensics, voice analysis involves many techniques. Commonly, this process includes one or more than one techniques, such as verifying the authenticity of a recording, interpreting noises, profiling speakers, or transcribing dialogues from voice recordings. Some fragments of a conversation are habitually received from phone conversations that include hoax calls, making calls asking for ransom, etc. Again, voicemails, too, seem to be vocal information’s common sources besides secretly recorded conversations as well as voices that are captured in a video. Hence, voice forensic experts seem to be an indispensable part of an investigation as these experts can recover, repair, analyze, and improve video and audio recordings using many scientific techniques and tools.