New approaches to Head Related Transfer Function (HRTF) based artificial spatialisation of audio are presented and discussed in this paper. A brief summary of the topic of audio spatialisation and HRTF interpolation is offered, followed by an appraisal of the existing minimum phase HRTF interpolation method. Novel alternatives are then suggested which essentially approach the problem of phase interpolation more directly. The first technique, based on magnitude interpolation and phase truncation, aims to use the empirical HRTFs without the need for complex data preparation or manipulation, while minimizing any approximations that may be introduced by data transformations. A second approach augments a functionally based phase model with low frequency non-linear frequency scaling based on the empirical HRTFs, allowing a more accurate phase representation of the more relevant lower frequency end of the spectrum. This more complex approach is deconstructed from an implementation point of view. Testing of both algorithms is then presented, which highlights their success, and favorable performance over minimum phase plus delay methods.