On 2013-05-03, Iain Mott wrote:
Dave mentioned modelling of IRs as an alternative to interpolation - but this he mentioned would be based on not only the impulse responses taken, but on the architecture. These are wilderness environments - so I guess modelling won't be possible and I imagine the number of IRs needed would be high anyway.
There are some more or less systematic approaches you could use to extract what you need from a set of impulse responses, so that you could do some kind of analysis-resynthesis of the cues in a form that is safer to interpolate.
Most likely you'd try to extract some number of leading specular reflections, some low order approximation of the overall decay of the diffusive part of the reverb, and including some spectral envelope, directional prominence and average diffusivity information over the audio band. Many of the building blocks you'd need for something like that already exist, e.g. in echo cancellation work, room equalisation, Ville Pulkki's DirAC work with its frequency selective intensity/reactivity analysis, and speech coding (estimation of low order whitening filters as a starting point, including psychoacoustical optimization).
With that information you could drive a purpose built bank of conventional reverbs, or something working half-and-half in time and frequency domains to build the impulse response for convolution, or if you were really bold, DirAC-like STFT recontruction which isn't strictly speaking time-invariant even in steady state. In any case, I'm rather sure it's doable. Whether it has been done or how much progress towards such an implementation has been made is a question for the architectural acoustics crowd, parametric stereo coding gurus, and of course commercial reverb manufacturers who do try to mimic existing spaces without outright convolution, so that they have to have ways of extracting such higher level descriptions from their measurement data.
-- Sampo Syreeni, aka decoy - de...@iki.fi, http://decoy.iki.fi/front +358-50-5756111, 025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2 _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
