[PlanetCCRMA] Any 3d midi controlled audio mixer using HRTFs?

Niels Mayer nielsmayer at gmail.com
Sun Mar 14 15:34:25 PDT 2010

Seems like you could have a 127x127x127=2,048,383 element matrix of wav
files, each
representing the FIR/IIR of a point source in a reference "performance
space". Then use three midi controllers to select X/Y/Z axes in this matrix
of WAV files. Each time a controller moves, you tell qconvolver to switch to
a difft FIR/IIR WAV in the "matrix", mix out the old qconvolver at the
previous position, and digitally mix in the new one at the new position.
Having multiple sound sources moving around would involve having multiple
instances of jconvolver, all digitally mixed in stereo in the end.

All you'd need to do is precompute 2 million FIR/IIRs and have on the order
127*127*127*32848 =67,285,284,784 bytes storage. You could easily fit that
amount of data on a hard drive or SSD.

Would it work? Are today's processors up to the challenge? How do you go
about precomputing 2 million positional FIR/IIR's in a virtual acoustic

And what of tech issues like "zipper noise" problem as you swept midi
parameters and invoked a series of separate filters, unless you figured out
a way of nicely "inbetweening" the sound. And then of course, if there's
velocity, there needs to be a doppler effect too.

-- Niels

PS: http://space-net.org.uk/?q=node/50

Developments in measuring the acoustic characteristics of concert halls and
opera houses are leading to standardized methods of impulse response capture
for a wide variety of auralization applications. The work presented here are
some of the results from a recent UK survey of non-traditional performance
venues focused in the field of acoustic archaeology. Sites have been
selected and analyzed based on some feature of interest in terms of their
acoustic properties. As well as providing some insight as to the
characteristics and construction of these spaces, the resulting database of
measurements has a primary use in convolution based reverberation and
auralization. The initial survey work has been funded by the Arts Council
England <http://www.artscouncil.org.uk/> and the
AHRC<http://www.ahrc.ac.uk/> as
part of their ACE/AHRC Arts/Science Research Fellowships. Damian Murphy was
appointed as one of these fellows and is the lead researcher in this work
that now falls partly under the remit of SpACE-Net. Later work was sponsored
by the York Renaissance Project <http://www.renaissanceyork.org.uk/> in
conjunction with artist and archaeologist John Oxley and artist Mark
Hildred. More details relating to this work, including details relaing to
the measurement techniques used can be found in the following publications:

   - D. T. Murphy, "Multi-channel Impulse Response Measurement, Analysis and
   Rendering in Archaeological Acoustics", 119th AES Convention, Paper No.
   6532, New York, USA, Oct 7-10, 2005. Available
   - D. T Murphy, "Spatial Audio Measurement, Modeling and Composition,"
   Leonardo, Vol. 39, No.5 (2006) - Forthcoming.
   - D.T Murphy, "Archaeological Acoustic Space measurement for convolution
   reverberation and auralization applications", Proc. of the 9th Int.
   Conference on Digital Audio Effects (DAFX’06 <http://www.dafx.ca/>),
   Montreal, Canada, September 18-20, 2006. Available online

In addition, the following links have been invaluable in providing direction
for this work:

   - Angelo Farina's Home Page <http://pcfarina.eng.unipr.it/>
   - Preserving Acoustics for Posterity -
www.acoustics.net<http://www.acoustics.net/> -
   associated with Waves <http://www.waves.com/>.

PPS: Above link mentions"96kHz versions are available at 5deg intervals for
the complete measurement circle through negotiation with the authors" seems
like that also would form the basis for a real positional acoustic space
simulator. Seems like they are using this software etc for similar
purposes: http://www.kokkinizita.net/ambisonics/casadelsuono/chandelier.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://ccrma-mail.stanford.edu/pipermail/planetccrma/attachments/20100314/555c4143/attachment.html 

More information about the PlanetCCRMA mailing list