|
|
|
Lamberto Tronchin*,**, Angelo
Farina**, Michele Pontillo*, Valerio Tarabusi* |
|
|
|
* DIENCA - CIARM , University of Bologna, Viale
Risorgimento 2 - 40136 Bologna Italy |
|
** Ind. Eng. Dept., University of Parma, Viale
delle Scienze - 43100 Parma, Italy |
|
|
|
|
Creation of synthetic filters for the
auralization of the acoustic response of a room |
|
Computation of binaural impulse response
starting from the results of room acoustic programs |
|
Use of experimental HRTF sets coming from dummy
heads (Kemar, B&K, Ambassador, Sennheiser) |
|
Development of the algorithm for creating the
binaural impulse response in standard WAV format |
|
|
|
|
The room acoustic program employed here
(Ramsete) produces two output files: |
|
the complete energetic impulse response
(typically with a resolution of 1ms) without any directional information |
|
The discrete energy arrival of low-order
reflections, with exact arrival time and director cosines |
|
The new algorithms processes these two files,
together with a third WAV files containg the experimental HRTFs of the
chosen dummy head |
|
|
|
|
The complete set of binaural IRs measured on the
Kemar dummy head at MIT-Medialab |
|
|
|
|
Number of measurements at each elevation |
|
|
|
|
A local cartesian reference system is assumed
solidal with the listener head; X’ axis is pointing forward (nose), Y’axis
is pointing on the left ear, and the Z’ axis towards the top of the head. |
|
|
|
|
The Ramsete program saves the coordinates of
three points in the absolute reference system: |
|
- Pprov º (xprov, xprov, xprov)
= provenience point of the ray; |
|
- Orec º (xrec, yrec,
zrec) = receiver origin; |
|
- Ptarg º (xt,
yt, zt) = receiver target point. |
|
|
|
First of all, the position of the
provenience point of the ray is recomputed in the local reference system: |
|
Pprov º> P’prov (x’prov,
y’prov, z’prov) |
|
|
|
|
|
|
Elevation angle j : |
|
|
|
|
|
Azimuth angle q : |
|
|
|
|
|
the weight Pi (i = 1,2,3) relative to
each HRTF is obtained calculating the opposite triangle area and dividing
it for the total initial triangle area; so we have P1+P2+P3
=1. |
|
|
|
|
At the exact arrival time of the received ray,
an averaged binaural IR is added to the global impulse response. |
|
The averaged IR is obtained by a frequency
domain interpolation between the three HRTF complex spectra: |
|
|
|
|
|
The interpolation is actually done separately on
the modulus and phase of the complex spectra |
|
FFT and IFFT are used for converting between
time domain and frequency domain |
|
|
|
|
Once the energetic impulse response has been
converted into a binaural pressure impulse response, it can be used for
auralization by means of a fast convolution algorithm (Select-Save) |
|
The auralized sound can be listened by means of
an headphone, or it can played on a pair of loudspeakers thanks to a proper
cross-talk cancelling set of inverse filters (Stereo Dipole) |
|
An example of the auralization of a sound field
in a virtual room is provided at the end of this presentation |
|
|
|
|
The selection/interpolation algorithm developed
allows for the use of HRTF data-bases measured with uneven angular
resolution, as those provided by MIT-Medialab |
|
The reconstructed binaural IRs proved to produce
realistic, high quality auralization |
|
Already available software allows for
manipulation of the computed IRs (convolution, cross-talk cancellation),
making the whole auralization process feasible on a low cost PC without
dedicated hardware. |
|
|
|
|
S.Domenico Church - Udine |
|