I built a recording studio 30 years ago, before digital audio technology was available. I had always been interested in computers so as they became more involved in the audio industry, my usage of them in the creation and editing of audio increased as well. I was a beta tester for a number of different software companies over the years with respect to digital audio and video editing. I have done significant amounts of audio clean up and recovery, as well as redaction and format conversion for the district attorney’s office at a county local to me. I was just presented with a case where a telephone call was recorded. There is an extreme amount of noise in the file, and while the voice of the person using the phone is relatively understandable, the person he is talking to is not. I’ve been asked to see what can be done to make the second voice audible. In general I am capable of cleaning up audio distortion and providing an opinion as to whether a voice on multiple recordings belongs to the same person, however as I have not heard the file yet, I will not state it can be done in this situation. For example the word distortion is used. What kind of distortion are we talking about? Is it truly distorted or is it a case of noise drowning out the voices in question? Noise and distortion are two different things. In many cases noise can be handled and even completely eliminated depending on the severity of the noise and the signal strength of the audio we want to recover. Noise is “separate” from the voices. Distortion is the degradation of the audio we wish to recover itself. It would be similar to a smudged finger print. To what degree is the voice distorted? If we can remove some of the noise, and some of the frequencies creating the distortion we may be able to hear what was being said. Voice identification is based on spectral analysis of the audio recording. Unfortunately, unless the two or more voices are saying the same thing, in a controlled environment, it would only be an opinion, not proof.