Old 05-04-2021, 01:09 PM   #1
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default Seeking greater directivity

Hello,

I am using SPARTA/IEM and Blue Ripple plug-ins to compose in 3rd-7th order for binaural reproduction and have been disappointed with the directivity of the signals. I would love some help understanding the issues and identifying methods to increase directionality of signals. Sitting in my car and listening to raindrops hit the roof convinced me that my workflow is missing something or there's a significant ways to go.

1. Using a mono signal of a simple, timbrally-rich transient, I directionalize it to 3rd (or 7th) order using an encoder (typically O3A but I have tried IEM and SPARTA).

2. I decode that to headtracked binaural for the headphone output.

3. Verify the required track count is satisfied and instantiate an Energy Visualizer on the Mix Monitor track.

With this setup the output responds to source placement (panning) in the encoder and responds to my headtracked movements.

Where I am struggling is achieving MORE directionality. I've tried using Directivity Shaper in conjunction with the Room Encoder according to the documentation (https://plugins.iem.at/docs/directivityshaper/) but the placement of sounds is still diffuse to my ears.

I have tried loading custom SOFA files but find it crashes SPARTA/Reaper more often than my liking. One possible avenue of interest is with exotic HRTF SOFA files. I would like to hear through a bat's ears some day.

Anyways, if anyone has methods they use to increase directivity I am all ears.

Best
Cotally
Cotally is offline   Reply With Quote
Old 05-05-2021, 01:25 AM   #2
jm duchenne
Human being with feelings
 
jm duchenne's Avatar
 
Join Date: Feb 2006
Location: France
Posts: 858
Default

Hello,

It seems to me that you have already done everything possible ...

The ambisonic encoding configured in order 7 considerably reduces the spatial blurring provided by this technique, and it can be considered that its degree of precision is in this case quite acceptable.

The problem of binaural processing is more complicated, and at this level I suppose that only the use of personalized hrtf could allow an improvement, especially regarding the elevation!

You could also try to not go through the ambisonics processing and conversion, if your purpose is to produce binaural.
Some binaural spatialization plugins are very good at it :
http://anaglyph.dalembert.upmc.fr/index.html

Last edited by jm duchenne; 05-05-2021 at 01:30 AM.
jm duchenne is offline   Reply With Quote
Old 05-05-2021, 09:46 AM   #3
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,039
Default

It's a bit expensive but DearVR Pro seems to walk this line well for me. I use it for spatialized mono sources, objects basically. Because it outputs to speaker arrays, binaural or ambisonics I can easily fit it into whatever format I'm mixing for.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
mypodcast - Ultimate Sound Test
plush2 is offline   Reply With Quote
Old 05-05-2021, 12:15 PM   #4
Kewl
Human being with feelings
 
Join Date: Jan 2009
Location: Montreal, Canada
Posts: 119
Default

Cotally, I would be curious to check the Reaper project (without the audio files): maybe there is some routing problems?
Kewl is offline   Reply With Quote
Old 05-05-2021, 01:16 PM   #5
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by Kewl View Post
Cotally, I would be curious to check the Reaper project (without the audio files): maybe there is some routing problems?
Please see attached. My master template is over 300 tracks with complex routing so I have created a basic one for you to review. Thanks for taking a look at it. It only requires the SPARTA plugins.

The directionality is OK but what I really want is, simply, an off-axis attenuator. Specifically I want sounds coming from behind to be more attenuated.

I started the reddit.com/r/HRTF in an effort to get a better understanding of the HRTF/SOFA file resources out there but still find the subject daunting (as an amateur). I found a matching ear/shape and size in the CIPIC database but its directionalization was backwards using SPARTA's binauralizer. What I'd really like is some exotic HRTFs to be able to use the decoding of the soundfield for artistic purposes as opposed to solely accurate reproduction.

Best,
Cotally
Attached Files
File Type: rpp Basic O3A sparta example.rpp (42.5 KB, 107 views)

Last edited by Cotally; 05-05-2021 at 09:38 PM.
Cotally is offline   Reply With Quote
Old 05-05-2021, 09:53 PM   #6
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by jm duchenne View Post
You could also try to not go through the ambisonics processing and conversion, if your purpose is to produce binaural.
Some binaural spatialization plugins are very good at it :
http://anaglyph.dalembert.upmc.fr/index.html
Thanks for your reply. I'm interested in head-tracked binaural. Am I mistaken or is that beyond the purview of Anaglyph?

Thanks
Cotally is offline   Reply With Quote
Old 05-05-2021, 09:55 PM   #7
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by plush2 View Post
It's a bit expensive but DearVR Pro seems to walk this line well for me. I use it for spatialized mono sources, objects basically. Because it outputs to speaker arrays, binaural or ambisonics I can easily fit it into whatever format I'm mixing for.
Have you tried SPARTA and compared? I've read anecdotal sources that say DearVR has desirable processing but am wondering if it's actually superior to IEM or SPARTA. Also it seems the head-tracking is limited to their Spatial Connect system. No way I'm wearing a clunky headset longer than I need to...
Cotally is offline   Reply With Quote
Old 05-06-2021, 10:41 AM   #8
Kewl
Human being with feelings
 
Join Date: Jan 2009
Location: Montreal, Canada
Posts: 119
Default

So,

- Maybe a setting you forgot, but the AmbiBIN is set for 7th order, but the project is 3rd order.

- To attenuate certain directions, maybe you should have a look at Matthias Kronlachner's ambiX plug-ins, namely the "ambix_directional_loudness": http://www.matthiaskronlachner.com/?p=2015

Kronlachner's mcfx plug-ins are also quite useful: http://www.matthiaskronlachner.com/?p=1910
Kewl is offline   Reply With Quote
Old 05-06-2021, 10:46 PM   #9
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by Kewl View Post
So,

- Maybe a setting you forgot, but the AmbiBIN is set for 7th order, but the project is 3rd order.
http://www.matthiaskronlachner.com/?p=1910
Yeah, I can monitor in 7th order but realistically I think 3rd order has the best possiblity of reaching an audience at the moment so I'm primarily testing that.

Quote:
Originally Posted by Kewl View Post
- To attenuate certain directions, maybe you should have a look at Matthias Kronlachner's ambiX plug-ins, namely the "ambix_directional_loudness": http://www.matthiaskronlachner.com/?p=2015

Kronlachner's mcfx plug-ins are also quite useful: http://www.matthiaskronlachner.com/?p=1910
Thanks, I just had a go at it and I expect it will be somewhat useful but not quite the silver bullet I'm seeking. It seems decent at focusing a sound by controlling reflections but it doesn't seem to attenuate the sound source arriving from the rear.

My current thought is that it will take modifying an HRTF to attenuate sounds coming arriving at the rear of the head. If anyone has the ability to assist me in this effort I would be grateful. I've got Octave and the SOFA API installed but am outside my comfort zone.

Last edited by Cotally; 05-06-2021 at 11:11 PM.
Cotally is offline   Reply With Quote
Old 05-07-2021, 02:39 AM   #10
Kewl
Human being with feelings
 
Join Date: Jan 2009
Location: Montreal, Canada
Posts: 119
Default

Quote:
Originally Posted by Cotally View Post
I can monitor in 7th order but realistically I think 3rd order has the best possiblity of reaching an audience at the moment so I'm primarily testing that.
Not to be nitpicking here, but, when you say that you "can monitor in 7th order", is that actual audio you're monitoring in 7th order?
Kewl is offline   Reply With Quote
Old 05-07-2021, 08:32 AM   #11
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,039
Default

Quote:
Originally Posted by Cotally View Post
Thanks, I just had a go at it and I expect it will be somewhat useful but not quite the silver bullet I'm seeking. It seems decent at focusing a sound by controlling reflections but it doesn't seem to attenuate the sound source arriving from the rear.
Knowing nothing about your background or the specific goals of your project I'm wondering, do you have comparable audio production that you are aiming for? I only ask because there is a point at which the problem is not technological, it's physiological. Your mention about listening to the rain falling on the roof of your car is quite visceral, I can pretty much place myself in that exact scenario in my mind. The problem, if that's your end goal for perfect reproduction, is that there are many other phenomenon going on that have nothing to do with hearing. I guess I'm just trying to gauge what your expectation is based on.

Quote:
Have you tried SPARTA and compared? I've read anecdotal sources that say DearVR has desirable processing but am wondering if it's actually superior to IEM or SPARTA. Also it seems the head-tracking is limited to their Spatial Connect system. No way I'm wearing a clunky headset longer than I need to...
Actually you can select the output format for the DearVR Pro panner to be nearly anything, including TOA which can be sent to any Sparta or IEM etc. rotator or decoder.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
mypodcast - Ultimate Sound Test
plush2 is offline   Reply With Quote
Old 05-07-2021, 09:27 AM   #12
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by Kewl View Post
Not to be nitpicking here, but, when you say that you "can monitor in 7th order", is that actual audio you're monitoring in 7th order?
Specifically, I take a mono timbrally-rich source and use either IEM or SPARTA or Kronlachner's AmbiX plug-ins to place that in a 7th-order-ambisonics space (64 channels), then I decode via the same plug-ins to head-tracked binaural. I haven't opted to work in 7th order much as I notice more ticks/drop-outs given the heavy workload. My plan would be to work in 3rd order then finish/export master at 7th order.

Last edited by Cotally; 05-07-2021 at 09:39 AM.
Cotally is offline   Reply With Quote
Old 05-07-2021, 09:38 AM   #13
Cotally
Human being with feelings
 
Join Date: Jan 2021
Posts: 13
Default

Quote:
Originally Posted by plush2 View Post
...do you have comparable audio production that you are aiming for?
I don't have a reference production for my request and haven't found much accessible ambisonics content (I'd enjoy a link to some ambisonic content). I've been working in surround sound for a long time so cinema is my main reference. I have hands-on experience with most of the surround and ambisonic mics including Eigenmike 32. One could say I'm looking to reproduce the experience of hearing the soundfield while wearing something that blocks sound coming from behind the listener. I want this because I'd like to place other sound content there. Presently I would need to lose head-tracking and bake to binaural in order to accomplish this.

Quote:
Originally Posted by plush2 View Post
Actually you can select the output format for the DearVR Pro panner to be nearly anything, including TOA which can be sent to any Sparta or IEM etc. rotator or decoder.
Noted with the caveat that it doesn't presently support higher than 3rd order.

On a side-note: I have been listening to your podcast btw but am wondering if you have a downloadable version as my youtube experience isn't head-tracked.

Last edited by Cotally; 05-07-2021 at 10:10 AM.
Cotally is offline   Reply With Quote
Old 05-07-2021, 03:15 PM   #14
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,039
Default

Quote:
Originally Posted by Cotally View Post
I don't have a reference production for my request and haven't found much accessible ambisonics content (I'd enjoy a link to some ambisonic content). I've been working in surround sound for a long time so cinema is my main reference. I have hands-on experience with most of the surround and ambisonic mics including Eigenmike 32. One could say I'm looking to reproduce the experience of hearing the soundfield while wearing something that blocks sound coming from behind the listener. I want this because I'd like to place other sound content there. Presently I would need to lose head-tracking and bake to binaural in order to accomplish this.
I've been playing around with using spectral editing and acon noise reduction (2) on some recent recordings and was rather astonished by how well it pulled out the ambisonic encode of foreground sounds while eliminating the background. Since you come from a film background (it sounds like) it wouldn't be unpalatable to you (I assume) to get clean spatialized focus sounds and add in the ambience as it were after the fact. This would give you near absolute control of what is in front of the listener and what is behind.

Quote:
Noted with the caveat that it doesn't presently support higher than 3rd order.
That is true. I'm not sure that higher order is going to get you the universal resolution performance you desire. I think there's always a bit of trickery. There's a reason why they shake a box of matches while rotating it around your head and don't play a violin. The violin would sound better but wouldn't be as definite.

Quote:
On a side-note: I have been listening to your podcast btw but am wondering if you have a downloadable version as my youtube experience isn't head-tracked.
That's an interesting question. I would be happy to provide you with an audio render of one of them, or would you like the video directional cards/visualization as well? If there's more interest I could probably put up a page with all of it. I'll do some looking into that.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
mypodcast - Ultimate Sound Test
plush2 is offline   Reply With Quote
Old 02-01-2022, 11:18 PM   #15
sephirious
Human being with feelings
 
sephirious's Avatar
 
Join Date: Sep 2020
Posts: 16
Default share

ive been at this for a minute now the best combo i got was


dear vr to pan stuff (mono source)
sparta ambienc to pan stuff (multi source)
ambipan hd (to pan stuff)

compass decoder (will not run in realtime on MUSIC setting (the best one) so i usually substitute with the sparta ambibin during production

if you want more pinpointed sounds bias the compass decoder towards the direct setting .

also flux evo channel as a eq compressor for up to third order i believe

good luck man

consider contacting me if you find something more powerful.. im dying for sparta/compass to release a 7th order compass decoder !





Quote:
Originally Posted by Cotally View Post
Hello,

I am using SPARTA/IEM and Blue Ripple plug-ins to compose in 3rd-7th order for binaural reproduction and have been disappointed with the directivity of the signals. I would love some help understanding the issues and identifying methods to increase directionality of signals. Sitting in my car and listening to raindrops hit the roof convinced me that my workflowa is missing something or there's a significant ways to go.

1. Using a mono signal of a simple, timbrally-rich transient, I directionalize it to 3rd (or 7th) order using an encoder (typically O3A but I have tried IEM and SPARTA).

2. I decode that to headtracked binaural for the headphone output.

3. Verify the required track count is satisfied and instantiate an Energy Visualizer on the Mix Monitor track.

With this setup the output responds to source placement (panning) in the encoder and responds to my headtracked movements.

Where I am struggling is achieving MORE directionality. I've tried using Directivity Shaper in conjunction with the Room Encoder according to the documentation (https://plugins.iem.at/docs/directivityshaper/) but the placement of sounds is still diffuse to my ears.

I have tried loading custom SOFA files but find it crashes SPARTA/Reaper more often than my liking. One possible avenue of interest is with exotic HRTF SOFA files. I would like to hear through a bat's ears some day.

Anyways, if anyone has methods they use to increase directivity I am all ears.

Best
Cotally
__________________
http://axisproject.net
sephirious is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 11:59 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, vBulletin Solutions Inc.