In virtual acoustic imaging systems, the perception of a sound source anywhere in space can theoretically be reproduced for one listener by using only two loudspeaker and an appropriate set of filters, called crosstalk cancellation (CTC) filters. However, the ideal position of the loudspeakers in such systems is still a question that is currently addressed by many researchers. Several analytical models with varying degrees of complexity have been used to analyze this matter. The fact is that, so far, all implemented CTC systems were built for a single listener use. When it comes to two listener CTC systems, the conditioning of the matrix of transfer functions between the loudspeakers and the listener's ear seems to be more irregular than the conditioning of the matrix for a one listener CTC system. Using a simplified model made it possible to verify that the position of the sound sources also plays a major role as far as the two listeners system's performance is concerned. Now, the influence of the presence of the listener's head in this model is evaluated. The free-field model of a two listener CTC system is expanded by modelling the listeners' heads as two rigid spheres. Using this new model an optimal source displacement is calculated based on two potential source distribution arrangements, namely a linear and a circular source distribution.
展开▼
