I already did it years ago. But I used my mixer-encoder to produce the test signals.
Benjamin Bauer did the SQ version of the tests. I did the QS version.
View attachment 75541
And Bauer made a circuit that moves that F signal from the top of the head to the front. I built one. Using that, I got a straight decoding for QS for headphones.
But what is a "binaural extraction"? It is an encoding that works with the human ear. Since the ear can decode it, there should be an electronic way to decode it.
There are two ways to do a downmix: Straight combing and matrix encoding.
There are two ways to do an upmix: Designed panning and matrix decoding.
This ear decoding is essentially what dichophony is (above).
