Forensic audio enhancement

Forensic audio enhancement is the scientific analysis and improvement of audio clarity, typically to improve intelligibility.

Although the term is "enhancement", the process is almost entirely composed of filtering away unwanted sounds in order to leave a more understandable version.

Process

To accomplish enhancement, the most common process is to follow these steps:

1. Reliably digitize audio recordings onto a computer system, typically in WAV (lossless) or MP3 (lossy) formats.
2. Filter out hiss, click and pop sounds. This is common with recordings that may have suffered partial data loss with age.
3. Block out^[1] or quiet ranges and specific frequencies (such as 60Hz wall power, and its multiples, within the USA and Japan).
4. Isolate unwanted constant tones or patterns and then removing those items from across the entire recording.
5. When the recording is in stereo, reference the channels to each other to perform directional noise cancelling.
6. Normalize voices so that each audible person is heard at equal volumes.
7. Smooth sounds that border nearly inaudible gaps. These gaps include those created during the filtering process.
8. Prepare a transcript of audio and words heard
9. Testify in court about the enhancement done and the audio and words heard

Admissibility as evidence

• The processes applied must adhere to the standards of evidence and so it is admissible in a court of law. The audio expert must be able to document the steps they took, these steps must be replicable, no data can be created, and no judgment based edits or deletions can occur. Changes in pitch or speed can also render a recording inadmissible.
• The courts may require proper chain of custody documentation with this evidence. This includes appropriate precautions to ensure that the evidence is properly protected and stored.
• If the enhancement is performed using commercially available software, proof of vendor specified training and proficiency may be required.

References