The online racing simulator
Directional audio
(6 posts, started )
Directional audio
There is a project at my work using head related transfer functions to simulate directional audio using stereo headphones. I was thinking that we could use LFS to demo the technology.

My question is whether it is possible to get each incident sound source (including echoes from the environment) and its angle of incidence from LFS. We could then filter each sound source separately with the HRTF and mix them after the filtering.
-
(thisnameistaken) DELETED by thisnameistaken
Quote from thisnameistaken :Ooh. So does synthing this stuff create good results then?

if the hrtf is detailed enough (enough samples in time and space) and your ears are close enough to the standard ears used in measuring; with good headphoes it will sound better than any multispeaker arrangement (discounting wavefield synthesis systems)
-
(thisnameistaken) DELETED by thisnameistaken
Quote from thisnameistaken :Ooh. So does synthing this stuff create good results then?

Yes, very good results, provided that you're "average". The reason being that the HRTFs are created for a person with an average sized head/features. That said, I would say that most people find the effect very realistic and immersive.

We've got the freakiest looking dummy sitting in the sound lab right now. I get scared every time I walk past and catch him in the corner of my eye.
Quote from Shotglass :...with good headphoes it will sound better than any multispeaker arrangement (discounting wavefield synthesis systems)

You may know this already, but for the other interested readers: The weakness of wavefield synthesis systems is that they can only accurately reproduce a wavefield within a certain region of space. More speakers (more money) leads to larger regions over which the soundfield is accurate, but if you move outside the region then it all falls over.

The weakness of the HRTF filtering method is, of course, that you need to wear headphones. But then again, it's a lot cheaper.
Quote from Peptis :My question is whether it is possible to get each incident sound source

If its assumed that vehicles are the only sound source, excluding echos, then yes its possible to get the sort of data you're after (using InSim and potentially OutSim). I'm guessing (as I'm afraid I know very little about audio orientated math and the associated programming), that from both the relative positioning of the vehicles (from the listening point) and the positioning on the track, it would be possible to guestimate the where the echo "source" location orginates.
Quote from the_angry_angel :If its assumed that vehicles are the only sound source, excluding echos, then yes its possible to get the sort of data you're after.

I had already thought of that and it wouldn't be hard to do at all. I was just hoping to do something a little more immersive.

Directional audio
(6 posts, started )
FGED GREDG RDFGDR GSFDG