Panorama

How can I import my own HRTF data?

See this document on the Panorama HRTF file format.

These matlab functions facilitate the manipulation of Panorama format HRTF files. Functions are included to create and open HRTF files, and to read and write individual HRTFs from a file. To get started, look at the documentation in the h_new function, and read the HRTF format documentation.

Why does the sound move when I first start processing?

The sound is getting panned from its default initial location to the target location when you start processing. Panorama implements motion filtering, hence the panning will take fractions of a second to complete.

Why doesn’t the sound source just start in the proper start position?

This is a particular issue with the DirectX plug-in format and the host implementations. If we were to fix this problem on one host, it will cause problems with other hosts. For now you can try:

1) insert 1 sec silence before sound

2) preview the effect once before applying the effect, allowing Panorama to complete the motion pan.

How is Panorama different than WaveSurround?

Panorama is a general purpose 3-D panning plug-in with environmental effects, while WaveSurround is specifically intended to be a stereo widening effect. Both plug-ins are based on similar HRTF and crosstalk-cancelling technology, but Panorama is far more advanced.

Panorama allows you to move the sound anywhere in the space, while simulating Doppler motion effect, distance cues, early room reflections and late reverb. WaveSurround constrains the virtual source positions to be symmetrically positioned on opposite sides of the listener at a fixed distance and on the horizontal plane, like virtual stereo speakers. Hence WaveSurround is very limited as a 3-D panner.

Panorama can implement WaveSurround style stereo widening by positioning a stereo source at 0 degrees azimuth and adjusting the stereo width parameters. However, WaveSurround has a subtle difference in the way it processes reverb. In WaveSurround, the reverb output is spatialized by the 3-D HRTF filters, whereas in Panorama, the stereo reverb output is not spatialized. Hence, WaveSurround is capable of producing very wide sounding reverb effects. This effect could be produced in Panorama by preceding Panorama with MasterVerb, and having Panorama position MasterVerb’s output.

With a monophonic source and 0 deg transformation, why are the left and right results different?

For sources on the median plane (0 degrees azimuth), the left and right channels can be processed differently when using the “Human” HRTF set or the CIPIC HRTF data, because those HRTF data are not perfectly symmetric. Most likely this is because the human subjects had slight asymmetries in their head and ears, or perhaps were not

facing perfectly forward when the HRTF was measured. The MIT KEMAR data is perfectly symmetric, thus a source on the median plane will have the same HRTF for left and right ears.

Another factor is room reverberation. The early reflection pattern will be different for the left and right ears if the distances to the left and right walls are different. In fact it sounds bad to have equal distances to the left and right walls with a centered source – the resulting early reflections are monophonic and have no sense of space. All of the Panorama presets have unequal left and right wall distances.

The late reverberation is always different for the two ears in order to produce a sense of space.