Head Related Transfer Functions

The crux of binaural synthesis is the amount and quality of transfer filters, which indicate a given sound's journey towards the receiving ear. These filters are called "Head Related Transfer Functions," or HRTF´s. The HRTF database; the measurements of transfer filters from which the binaural synthesis is being made, determines the quality of the 3D Sound. The 3D Sound products by AM Production Multimedia are developed on the basis of the world's best HRTF´s.

In principal an infinite number of HRTF´s are needed to synthesize genuine 3D Sound - one for all possible directions and distances. In practice a smaller set of HRTF´s is necessary. This reduction is achieved by applying a number of acoustical simplifications on the wave propagation model. Also the human localisation performance allows the use of a reduced set of HRTF´s. It should be emphasised that the quality of the HRTF´s used by us are evaluated through extensive listening tests.

HRTF´s are usually obtained through measurements.

Once a HRTF database has been obtained the basic binaural synthesis is obtained with the following signal processing:

The HRTF´s are individual since the body; head and ear shape are unique for each human. Ideally the used HRTF database should origin from the listener. However, it is possible to use non-individual HRTF´s but with a lack of localisation performance. Also the sensation that the virtual sound source is outside the listener's head will degrade.

The HRTF database used by AM3D is chosen from 40 human HRTF databases as being the most "average". That is, most people localize the best with this particular HRTF database. This work has been done at the Department of Acoustics at Aalborg University. The HRTF measurement method is patented and the Department of Acoustics at Aalborg University have shown that this method is the optimum with respect to obtaining non-individual HRTF´s.