The term ‘binaural’ was originally coined by Alan Blumlein:
‘On 14 December 1931, Alan Blumlein applied for his famous patent, No.394,325, Binaural Sound, in which he describes in great detail, an electronic method of reproducing sound from two microphones and two loudspeakers. He called this system ‘Binaural’ from the human factor of having two ears by which we hear sound.’ source http://www.doramusic.com/Stereo.htm
We now use the term ‘stereo’ to mean two channel audio recording and reproduction. However, our modern use of the term ‘binaural’ retains the principle of ‘…the human factor of having two ears by which we hear sound.’
So a binaural recording contains audio information as if the recording had been made using our head and ears, so that when heard on headphones with no crosstalk (the left and right ears are sonically isolated on playback) we perceive the three dimesional surrounding sounds as if the listener were at the location of the recording.
So how does it work?
Simply, the marvel of human hearing is that with only two channels (our two ears) we are able to perceive a full spherical (periphonic) surround sound perception of the world around us. The key to this ability relates to the size and shape of our head and the shape and position of our ears on either side of our head. Localization of sounds in space is achieved by two factors:
1) The separation of each ear, the shadowing effect of the head, and the time delays created between each ear of sounds coming from any point in space provides the means for the most powerful spatial elements in hearing. This fails when there are sounds directly in front or behind, where there is no delay between the left and right ears receiving the sound, and
2) The remarkable directional filtering effect of our external ears (pinnae). Our ears are assymatrical in three dimensions and function as directional acoustic filters causing specific phase shifts at certain frequencies depending on the _location_ of the source of the sound that we are hearing.
So as we are listening our brain actually receives a complex set of sounds – the actual sound itself and the time delay/ phase shifted sounds due to the physical effect of our head and ears, plus to some extent, the shadowing effect of our body, particularly with ‘below’ sounds.
Our brain has evolved to process the time delay/ phase information, which we hear as pure _location_ and the pure sound coming from that location.
This location / pure sound filtering and perception relies totally on the unique isolation and seperation of our ears, on each side of our head, such that there is no crosstalk between the left and right ears.
If we listen to a binaural recording on conventional stereo speakers, we introduce ‘crosstalk’, that is the ears are not isolated and both ears can hear the left and right channels, albeit at different volume levels. Because of this ‘crosstalk’ the time / phase components of the binaural audio are ‘exposed’ and localistion cues in the audio are heard as phasey or ‘comb filtered’ distortions mixed with the pure sounds. However, with high quality stereo speakers showing accurate phase matching, plus a wider positioning of the speakers, a high level of spatial localization can still be perceived and a very enjoyable ‘partial’ surround sound experience can be had from pure binaural recordings.
However for optimum listening binaural recording are headphone surround sound listening only.
And yet its not quite as simple as this would appear. Our head/ear/body shape (Head Related Transfer Function or HRTF) is unique to each of us, so if we record to one person’s head / ear characteristic (HRTF) only those with a similar match will hear the localisation cues perfectly, and if a ‘generalized’ characteristic is used some spatial detail will be lost by everyone. For reasons explained above the side spatial images are most stable, hence the most common effect of a mismatch is ‘front / rear reversal’ of centre front sounds, as this is the most fragile area in respect of left/right time delay cues as described above.
There is a vast amount of research and experimentation on this subject, the use of ‘dummy heads’, ‘pseudo binaural’ recording techniques, electronic synthesis and modeling of HRTF cues, as well as numerous binaural recordings, which you can access on the net, but will not be entered into for the purpose of this article.
My simple binaural recording technique is to insert high quality custom modified sub miniature microphones into my ears and thus sample my own hearing. ** On headphone playback, as described above, this will suit primarily those with my head/ear/body shape best, but I can make the spatial cues more robust by using ‘tracking cues’ – by creating or recording moving sound sources, which are easier to localize, and by not having stationary sound sources center front, which can easily be perceived as coming from behind. So with this sense of ‘spatial choreography’, my role as a binaural sound artist is to create a recording that is enjoyable and interesting for the majority of headphone listeners. It also provides a unique personal record of the human sonic perception of a sound environment.
**(Health Warning – do not insert any object into your ears without medical supervision).