octaphonic project ambitions and binaural workarounds

mendel lee's icon

Hey there lovely community.

I have this project in my head to create a digital audio piece and/or framework that uses an octophonic sound setup. I basically want to create a piece that relies heavily upon the perception of a sound's space being anywhere in a room and moving from one place in a room to another.

problem is, i don't have an octophonic setup nor do i have easy access to one. There is one at the university, but that space is (and should be) a priority for the student body (i'm on staff). When I asked the question of if it was possible to simulate that sort of environment without buying my own setup, someone got me onto binaural setups.

When i was first looking into it, i thought "awesome. i can write the piece using a binaural setup and then just transfer that to octophonic once i have access to an octophonic space." but the more research i do on it, the less i'm convinced this is the way to go because of how the setup for binaural would be so vastly different than the setup for octophonic - i'm getting the impression that if i do the work to spatialize the piece binaurally, that has no practical translation to octophonic - i'd have to respatialize the piece "from scratch" as it were in octophonic, which makes the binaural part of it seem like a waste of time.

Does this seem like an accurate assessment? If so, can anyone recommend any alternatives? You're also free to tell me that writing an octophonic piece is a stupid idea, and i should do surround sound instead or something. I'm in very initial stages of this conceptually and it's the first time i'm trying this, so i acknowledge that the whole thing could be very ill-conceived.

Thanks!

Roman Thilenius's icon

one could write a patch for the 8-channel setup and add an additional function for stereo output to it, where all panning modulations fpr the 8 channel system are also translated into modulations for a binaural filter for the stereo version.
then you could compose and mix at home without the need a multi speaker setup but just using headphones and a bit of imagination.

my favorite binaural filter is two copies of prosonic ambisone plug-in with inverted positions for the azimuth parameter - something pretty easy to realize in maxmsp.

of course, in case you plan to virtually move sound sources which should already use more than 2 speakers, that stops working.

-110

mzed's icon

There are spatialization encodings that are not specific to the diffusion setup. I think ambisonics are worth looking at. There's a good bunch of objects from icst, and this looks interesting:
http://www.maxobjects.com/?v=objects&id_objet=4010&requested=ambisonic&operateur=AND&id_plateforme=0&id_format=&PHPSESSID=a8342021f03167d34dfc4b8afd9d4259 (although maybe too old).

IRCAM also has some stuff, although not free. VBAP is my other thought, although not really binaural.

Anyhow, you can think about your spatialization as a separate issue from your specific speaker setup. I've moved between 8 channels, transaural and binaural. It's a very different listening experience, of course, but the technology works.

mendel lee's icon

thanks guys. both posts are very helpful for me as starting points and i can hopefully take it from there.

Cheers!