The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Increasing depth perception/immersion in Binaural audio? Virtual Instrument Plugins
Old 16th January 2018
  #1
Gear Addict
 
Oop's Avatar
 

Thread Starter
Increasing depth perception/immersion in Binaural audio?

Hi guys,

I just recently started exploring binaural/ambisonic audio mixing for VR and have a few questions regarding the platform.

I've been using Reaper and after exploring all the plugin options, I've been testing out AudioEase's 360pan (I like the embedded grid and pucks on the video) and the dearVR 360 plugin.

After working on a few post-production demos for a client of mine, I realised I was having a difficult time with the front-back movement perception. Sounds panned to the back, very rarely sound correctly placed. From watching videos on Youtube I realised, only recordings done with binaural microphones had that immersive depth, especially ones with the ambiences being recorded in 360. A lot of other audio I heard was quite flat and 2D, more like slightly enhanced stereo.

In my experience, dearVR's plugin had a bit more depth in panning back than the AudioEase plugin, but I wasn't too happy with some level/sound changes it applied.

Anyways, after showcasing the demo to my clients, I was sent an audio clip as reference, found in this link:

I was quite blown away by the immersion it provided. The depth, the ambience, distance information, all seemed very 3D. The Youtube version seems slightly degraded compared to what I was sent, but you can hear the sounds at the back, especially distance information.

A bit of research showed that it used something known as the 'Cetera Algorithm', used in hearing aids. I'm not sure if QSound Labs developed it, but it also seems like the technology is a decade or more old. It seems to handle audio at the back of your head quite well. In fact, the processing seems to add a kind of reverberation, so not sure how sounds directly in front of you would sound.

I was wondering if anyone had tips on how to create this level of immersion in post production within a DAW, using standard mono/stereo files, mixing in VR. I even looked around for binaural recordings of ambiences to use as a base layer, but didn't find much. Automating reverb etc. didn't really give the effect I was looking for.

TLDR:How to increase depth perception in binaural audio, similar to the link posted.
Old 9th February 2018
  #2
Lives for gear
 

how were the sounds recorded that you are mixing?

in my experience, mics that have their capsules very near together (such as the neuheiser ambeo vr) provide not enough spatial information for mixes that require very large depth or movement.

yet you may use the same techniques as when 'expanding' from stereo to pseudo surround: use different reverbs for front and rear.
Old 9th February 2018
  #3
Here for the gear
 
allengriswald's Avatar
 

I think it sounded quite flat with some stereo panning. To get depth you have to have dynamic range. Before the digital destruction of quality of sound, before those record producers destroyed the music to over compress for the cheap consumer gear, We actually had something called dynamic range. Please discover this again.
Old 10th February 2018
  #4
Gear Head
 
nonnaci's Avatar
 

Good sound depth in the far-field is about having the direct sound-source sit with the early reflections and late reverb in a spatialized environment (see our Zephyr plugin demos and sound-samples).

For near-field type depths, the direct-source dominates and requires additional modeling of low-frequencies diffracting around the head, perturbing the HRTFs; see Muze plugin.
Old 15th February 2018
  #5
Gear Nut
 
johannburkard's Avatar
Quote:
Originally Posted by Oop View Post
[I]I was wondering if anyone had tips on how to create this level of immersion in post production within a DAW, using standard mono/stereo files, mixing in VR.
Roland RSS or Prosoniq Ambisone. Supposedly there's a different algorithm of RSS in the Roland Integra 7.
Old 18th February 2018
  #6
As you suggested, the best way is with original binaural, or surround, audio files. You should also investigate using surround plug-ins, especially reverb.

Spatial placement gets freakishly esoteric when it comes to binaural and surround microphones, and I haven't heard a plug in that can accurately duplicate all the space that is created by a good surround or binaural microphone. (The digital haircut would be easy with a binaural mic.) I have also noticed height information in some of my binaural studio recordings which was very interesting - we positioned the head a little low in the room relative to the surround speakers and it was obvious in the binaural mix.

So FWIW, here's a no-plug-ins example of a recording that used a Holophonic microphone to record the voices, and a Neumann KU100 binaural microphone to capture the final binaural mixdown from the original surround mix in-studio:



Old 28th February 2018
  #7
Depends on your project, but you could bring your VR360 footage into Unity and place your sounds in 3d space. Built in audio engine works really well.
Old 28th February 2018
  #8
Lives for gear
You need to learn a lot about HRTFs to answer your question. - literally far too much for a message board. You might have something recorded on a binaural mic but it wouldnt have any effect at all on some people. Note the haircut vid doesnt work at all on a lot of people and less well on other suntil it works really well on most - you have to be on the bellcurve of "normal" headshape for it to work - I assume it was recorded on a Kemar dummy head (which is supposed to be the "average" shaped head) All to do with head/pinna shapes and since everyone has a different morphology there is no "one binaural filter fits all". Also look at front back reversals. Nearfield HRTFs are an area of ongoing reesearch the best we can do at the moment is head shadowing but theres a lot more to it than that.
Until personalised HRTFs come into easy existance this will not be solved - until then the best we can do is interploiate between different HRTF datasets. Id love to hear a plugin that can accurately reproduce neafield HRTFs without using personalised HRTFs....
Old 28th February 2018
  #9
Lives for gear
Quote:
Originally Posted by nonnaci View Post
Good sound depth in the far-field is about having the direct sound-source sit with the early reflections and late reverb in a spatialized environment (see our Zephyr plugin demos and sound-samples).

For near-field type depths, the direct-source dominates and requires additional modeling of low-frequencies diffracting around the head, perturbing the HRTFs; see Muze plugin.
Looks interesting Ill try them out. Do you interpoliate between HRTF datasets in any of your plugins - I assume you are taking the "best fit" approach ?
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump