Confirming Pseudo Stereo

[Illuminate]

I need to confirm that some stereo files in my possession have actually been made from a mono source (originally audiotape) as part of a forensic investigation. Can anyone assist and identify a definitive test?

I have read Dan McDonald's thread "Detecting Stereo vs Mono", which is helpful, but need to go a little further. Trying Craig Maier's suggestion of comparing a L-R version with the original, I could make-out no real difference, as he indicates is likely when listening to a mono converted sample. However for forensic purposes I need a less subjective test.

When viewing spectrograms of L compared with L-R, they are nearly identical, but with obviously less data in the L-R.

Any suggestions would be greatly appreciated.

[Craig Maier]

The key tool for determining stereo signals is the X-Y Display found under the View Menu. There is a description of it in the Users manual and some diagrams and descriptions to help you identify what a true stereo signal looks like compared to a mono one.

Craig

[Illuminate]

The key tool for determining stereo signals is the X-Y Display found under the View Menu. There is a description of it in the Users manual and some diagrams and descriptions to help you identify what a true stereo signal looks like compared to a mono one.

Craig

Thanks Craig for your reply. I've used this facility, and my signals do appear to be stereo, and not mono. But surely pseudo stereo will do that?

[Craig Maier]

A poor pseudo stereo signal will tend to produce elipses of varying major axis'. However, a good pseudo stereo signal (made with a program like Diamond Cut) is hard to discern from a true stereo one. So, it depends on the quality of the pseudo stereo synthesis.

When I do that, I use a combination of the comb filters found in the paragraphic, the time offset in the file conversion filter, a little echo effect and a very slight touch of reverb.

The result is something that looks and sounds like real stereo. But, I have seen commercial releases of pseudo stereo recordings that are easily identified as such using the X-Y display (elipses).

Craig

[Illuminate]

A poor pseudo stereo signal will tend to produce ellipses of varying major axis'. However, a good pseudo stereo signal (made with a program like Diamond Cut) is hard to discern from a true stereo one. So, it depends on the quality of the pseudo stereo synthesis.

When I do that, I use a combination of the comb filters found in the paragraphic, the time offset in the file conversion filter, a little echo effect and a very slight touch of reverb.

The result is something that looks and sounds like real stereo. But, I have seen commercial releases of pseudo stereo recordings that are easily identified as such using the X-Y display (elipses).

Craig

Thanks Craig. Using the X-Y vector analysis feature, I observe ellipses as you suggest, as the vector fluctuates from L to R. According to DCL's user manual, the signal is tending towards monophonic right and then monophonic left, which I assume can be described as "phase jitter".

Perhaps that is all I need to prove my hypothesis. However the "jitter" corresponds to individual speakers (ie the speech of individuals), and as I know the source audiotape was made on a 4 track conference recorder, I need to reassure myself that is not influencing the result. The problem is that I do not have a stereo source other than music to use as a comparison.

[Craig Maier]

Quoting Immuminate:

"The problem is that I do not have a stereo source other than music to use as a comparison."

----------------------------------------------------------------------

Just set up two mics in a room separated by a few feet and record a conversation between two people in said room. Look at that on the X-Y display and you shall see what a true stereo signal looks like. It will not be simple elipses.

Craig

[Illuminate]

Quoting Immuminate:

"The problem is that I do not have a stereo source other than music to use as a comparison."

----------------------------------------------------------------------

Just set up two mics in a room separated by a few feet and record a conversation between two people in said room. Look at that on the X-Y display and you shall see what a true stereo signal looks like. It will not be simple elipses.

Craig

I take the point, the problem is I do not possess two mics at the moment.

Below are some random screenshots taken from one of my suspect files.

[Craig Maier]

Hi,

I think that is a stereo signal - - - the average angle is changing alot which is a hint and the elipses are of variable foci. If you want to send me a short sample, I can analyze it myself and render my opinion about it. Send 10 secs of the file to me at

dctools@aol.com.

I will not get to it for a day or so as it is a holiday, but I will do it this week.

Craig

[Illuminate]

Thanks Craig, I appreciate that. I've sent you a short clip.