In these subdirectories, HRTF measured anechoically for various dummy head are stored.

The data were measured at the anechoic chamber of Centro Ricerche Fiat, in Turin, Italy.
A MLSSA system was employed. The loudspeaker was a Quested F11 self-powered monitor, 
in "flat" mode both at low and high frequency. The data actually are NOT compensated
for the loudspeaker/microphone response, although it was measured and I am doing the processing.

The directory name is the name of the dummy head.
The binaural impulse responses are stored in 32-bits float files (new Microsoft .WAV format)
at a sampling rate of 80 kHz (useful bandwidth is around 25 kHz).
CoolEditPro is suggested for reading properly these files.
Each HRTF is exactly 2048 points long. They are stored with constant angular separation (10° or 5°),
starting from front and going to the left of the listener.
In a few days, I shall post also filtered and shortened versions (at 44.1 kHz, 16 bits).

Angelo Farina - 8 march 1999