# Searching for an algorithm that can detect the stereophonic recording method used for a given stereo recording

I am trying to automatically detect, whether a given stereo recording was done by the X-Y (intensity) or A-B (time-of-arrival) recording method.

Does anybody know, if there is research on the topic? I could not find any.

• Not familiar with either term; from my very amateur look at things, stereophonic recordings are done with two synchronously sampled microphones. In which of your two categories would just two microphones fall? Sep 22, 2021 at 13:30
• @Marcus Müller: A-B, as well as X-Y are done with two microphones. X-Y is more common, but a lot of classic music is recorded A-B. There is a bunch of spacial algorithms (e.g. Surround Upmixing) that only work with X-Y, therefore it would be convenient to know the method used for a recording. (en.m.wikipedia.org/wiki/Stereophonic_sound) Sep 22, 2021 at 15:51

It's easy to tell if you can find passages in the track that are "hard panned". If there is mostly time difference, it's A/B, if it's mostly level difference it's X/Y.

You can also look at the average power spectrum of the sum and difference channel. I'm suspecting that A/B will show some signs of comb filtering correlated to the microphone spacing. That's the main reason why A/B isn't particular popular: it doesn't sum well to mono.

Keep in mind, that almost all commercial recordings these days are pos-processed mixed & mastered so they will be neither X/Y nor A/B.

If time of arrival differences are significant for some sources in some frequency bands, then one might speculate that this should be detectable using something ala crosscorrelation or its frequency domain equivalent.

For a perfectly aligned microphone pair, I would assume that such time differences are fewer and smaller.

-k

• So, I just calculated cross-correlations over a moving window on my test-data. The result looks very promising. My A-B recorded classical concert shows almost no energy at zero-offset. My X-Y test recording on the other hand shows a massive bulge. I think I will able to work with this. Sep 24, 2021 at 14:45