Whitepaper

This whitepaper consists of 3 parts.

Part 1. Discrepancies between common measurement methods and human hearing.

Part 2. Directional vs omnidirectional stereo sound reproduction.

Part 3. A novel practical solution for the creation of omnidirectional loudspeakers.

Discrepancies between common measurement methods and human hearing

Imagine the following situation. You are listening to a live orchestra performing in a world class concert hall and have the best seat in the house. Or you are at a organ concert in a cathedral, a chamber concert in a fine chamber, a choir performance in a church or any other acoustic performance in a space known for the finest sound.
You record this performance with a pair of perfect flat response omnidirectional microphones and then you visit an anechoic room, set up a pair of full range linear phase low distortion loudspeakers which measure perfectly flat in frequency reponse and play back the music you’ve recorded.
You’ll quickly notice something once you hit play. It sounds absolutely nothing like the real life performance!
And it is not dissimilar in a good way. It sounds harsh, small, unrealistic, uninvolving and quite unenjoyable compared to the real performance.
And as a check we can record the individual channels again with the same microphone by playing them back in this anechoic room through the speakers and get a perfect matching identical recording again. We have a seemingly perfect reproduction. But then why doesn’t it sound anything like the live performance?
There are several reasons for this but we can relate almost all of them back to the fact that our hearing is nothing like an omnidirectional microphone.
We can demonstrate and measure this discrepancy with a modification to the above experiment:
We’ll run the experiment as before but this time we record the live performance with both omnidirectional microphones and a special dummy head binaural microphone (like for instance a B&K 5128). We then play back the omni microphone recording in the anechoic chamber as before and record this listening session again with the dummy head.
We can now compare the dummy head recordings of the live performance with the dummy head recording of the playback in the anechoic room. And when we do this we’ll see that indeed they’re nothing alike. So whereas the omnidirectional microphone would show an exact match of the original recording and the (mono) playback in the anechoic room, this does not hold at all for human hearing.
And hereby is also shown that this discrepancy is something we can (and should) measure and quantify.

(under construction)

Directional vs omnidirectional stereo sound reproduction

(Under construction.)

A novel practical solution for the creation of omnidirectional loudspeakers

(Under construction.)