The one spatial audio trick nobody seems to know about!

Uncover the secret behind 3d mid-side processing for immersive and binaural audio.
Join my Discord Server: / discord
Links:
Plogue Bidule: www.plogue.com...
Kushview Element: kushview.net/
Ambisonics vs. Dolby Atmos: What's the difference? • Ambisonics vs. Dolby A...
Ambisonics in Ableton Live 10 Tutorial Part 1: Introduction • Ambisonics in Ableton ...

[@AudioLemon]

I just wanted to say that it is great to have a place to go to and learn more about spatial audio - so thanks for all your efforts!

[@michaelgwagner]

Thanks! You’re very welcome!

[@ampmodclips]

At last someone is coving this ! many thanks :)

[@michaelgwagner]

You’re very welcome!

[@jurgendeblonde5003]

Great tutorial and idea! I just managed to do the same thing in Reaper without using Bidule. I basically have a binaural decoder (IEM) to monitor. I send an instance of Odin2 into the ambix i2_o1 encoder by Kronlachner and send that into the wigware ambix freeverb (to have reverb) that then goes into an instance of mcfx_gain_delay and I have the same thing as you, where I can change the gain, mute channels et all... Sadly, i can't include a screenshot here ;-) Thanks for you great videos in general!!! (But as usual, things are a lot simpler in Reaper ;-) )

[@michaelgwagner]

Correct! There is a lot you can do with Reaper. Maybe I should do another Reaper video.

[@tikkorome5752]

I'll take Nuendo first every time, but Reaper is a great DAW for experimentation. I always recommend Reaper for students. Otherwise, ReaPlugs is the only bundle a student should ever touch for the first year (if an actual student) or the first 5 or 6 years if they're trying to learn from KZbin.

[@DanielBrandl]

I use the sennheiser foa microphone for live dance productions. The audience has headphones on and I send them a binaural signal. In processing, I can access all 4 channels of the mic individually and loop them. Like this I can not only loop a sound but also the direction it came from. It’s very handy. The approach in this video is of course a little different, but my idea also worked with the idea of working with the ambisonic channels by themself. I know, that the mic does not produce the channels, that are used in this video. Still, the idea is in the same way

[@michaelgwagner]

Interesting approach!

[@DanielBrandl]

@@michaelgwagner I'll have to thank you in the end. Your Videos on Ambisonics were the door Opener for the whole immersive Audio topic

[@agusleonardi4957]

this is what i'm looking for, thx sir

[@michaelgwagner]

You’re very welcome. 😉

[@SoundSleuth]

I am in process of building multiple MS Microphones for use in Spatial Arrays. You are a great resources and I bought Fiedler Audio Atmos plugins because of you. Would love to chat sometime.

[@michaelgwagner]

Great to hear. Thanks! Hop over to our Discord.

[@SoundSleuth]

@@michaelgwagner Will do

[@SamHocking]

Cool! For Mach1 I do similar and pull off the 4x stereo pairs into mid/side processing. Great for when you have mono stems, stereoize them using a k-stereo / haas delay technique and then mid/side process that back into Mach1 8. Little more difficult to do in Atmos as limited to only mid/side on the object input unless you work with object beds or beds instead and use the Plogue Bidule / Element to split the bed up.

[@michaelgwagner]

Yes, object based audio works differently. This trick would only be applicable to beds.

[@noelcrowder3708]

I’m trying to visualize how this may enhance or be applicable to more static recordings such as rock band situation. 🤔

[@QFXmusic]

This is great and i would say very useful and i learned something new well done michael

[@michaelgwagner]

Thanks! Happy to hear it is useful!

[@tapdapy]

probably, BRS Flare will be more demonstrable to see the sound spreading in 360. Thank you, vety useful experiment!

[@immerseaudio]

now were talking! awesome :)

[@immerseaudio]

so glad i found this video - especially the comment about eqing the higher orders. thank you

[@michaelgwagner]

@immerseaudio You’re very welcome! 😊

[@luckyjapan88]

Hi Michael, Thanx again for a good video with a different idea/approach to study spatial audio and sound design. BTW: I still have to say sorry for not being able to follow-up inside your discord community as I had too much work left to follow-up first.😒 But now I'm back to follow-up my work again in spatial audio design...😁 Your video is something what I'm looking for as I'm researching (I'm not a professional researcher 😉😎), or in other words, looking for ways to create spatial sounds in a more precise way within a DAW. It's really a pity that neither Cubase nor Nuendo (so, I like Nuendo) does support a detailed manual routing in Ambisonics (downmix is more flexible then upmix in terms of channel count, even with creating groups and converting the tracks around and using audio sends). According to your video, you are combining each order to one group, which is, of course, useful in terms of the specification/design/signal processing of Ambisonics, but am I right, that it would work also for a single channel only in a higher order than 0? For example, using the "side channels 1 or 2 to control this direction of the sound different and more in detail between channel 1 and 2? Either I need to search for different plugins (like Plogue bidule) or maybe I need to consider to change to reaper which seems to be more simple and flexible in terms of spatial sound routing details... I'm still not satisfied with current available 3D sounds (360゜, etc.), that's the reason why I'm looking for different approaches and ideas. So, for music or video/cinema, current technology is nice to create space and cool effects, when you have the right physical speaker assignment to the the physically available audio channels. Not talking about H/Ps with some more or less 3D sound image, depending on the used HRTF... Ambisonics is promising for me (so it's already old but unfortunately not really further evolved), as it is not speaker layout dependent, but, as all current available technologies, as far as I do know them, are still not enough for the real "wow, that's great!" effect... Whatever H/Ps, HRTF's, binaural, or whatever rendering is used... So, I'm still waiting for a video from you, where I can really say "Wow, that's f... cool! This is the direction to go I was searching for!" (If I should find a solution earlier, I'll let you know...) 😁😁😁😎😎😎

[@michaelgwagner]

Welcome back! Yes, you are right, having individual channel controls would be very helpful. But there are already developers out there who are developing plugins that can be used that way. I think the development of immersive audio is a gradual process. I will take time for the "wow" experiences to emerge. ;)

[@luckyjapan88]

Hi @Michael G Wagner thanx. Do you have some more examples than the ones in your video? Would save me some time... 😁

[@michaelgwagner]

There will be a few in the next couple of videos.

[@przymrozek]

This is brilliant! I wonder if one could mix 1OA down to traditional stereo (not binaural) as 3 M/S techniques, with doubling the side channel, reversing the phase etc. and then control the somewhat three-dimensional space of a stereo recording in this way.

[@michaelgwagner]

Not sure if it would have much of an effect if you don’t use an immersive format but give it a try. There are plugins that use immersive techniques for stereo audio.

[@Cam-Obi]

@@michaelgwagnergot any examples of some of those plugins?

[@michaelgwagner]

Sure. DearVR Music or the THX Spatial plugin are examples.

[@RobertSuchorski]

Hi Michael Great video 👍 It would be great if you make a video how all this works using cubase with Mir Pro 3D ? If you’re familiar with Mir ….

[@michaelgwagner]

Thanks for the suggestion! MIR is probably not a good choice since that is used for hyperreal room simulations and this trick is on the creative end. I’ll think about that.

[@RobertSuchorski]

@@michaelgwagner Thank you !

[@uniteone4189]

thank you , super interesting topic

[@michaelgwagner]

Glad you liked it!

[@i.a.m-poornasrikarm]

Excellent ✨

[@michaelgwagner]

Thank you! 🙂

[@solarion33]

any guidelines for basic mixing principles for ambisonic signals ? how much headroom each voice\sound needs ? distortions can also arrive from unexpected corners :-)

[@michaelgwagner]

Very good question! Ambisonics is not speaker based, so as long as you stay in the Ambisonics domain, headroom should not be an issue at all. It‘s only when you decode it but hat this becomes a concern. And i could be wrong but I believe that the decoder ensures that everything stays in check.

[@jaydobeatzz]

Can you make a video on where you place audio, for example bass in center channel, vocals on left and right, etc.

[@michaelgwagner]

I'm not really a professional audio engineer, so I usually try to avoid these type of tutorials. My main expertise is on the technical end of things.

[@gabrielwolfcolor]

Super tutorial! Wäre es möglich mit ffmpeg und den ersten beiden Kanälen einen einfachen Stereo-Downmix auf der Kommandozeile zu machen?

[@michaelgwagner]

That I don’t know. Ffmpeg should be able to work with 1st order Ambisonics because that is what is used in KZbinVR, but I’ve never used that.

[@sonar_worlds]

I have a question, not sure it makes sense 😅 If one where to copy a mono sound source to 4 identical mono tracks (A-format?), and then adjust gain levels (or add fx) to these individual tracks _before the encoding to ambisonics B-format (let’s say FOA). The omni and side 1,2,3 directions would not give the same result like when changing the levels of already encoded ambisonics signals, right?

[@michaelgwagner]

Correct

[@gwsound]

Is this the technology that Atmos uses in their player to generate 3D space on your headphone?

[@michaelgwagner]

No, Dolby is object based.

[@gwsound]

@@michaelgwagner yes I know but at the end of the day you have to create the space in software that you hear on the headphone. This converter is based on an algorithm. That was what I was thinking of.

[@michaelgwagner]

This technique is based on Ambisonics which is essentially a mathematical representation if the sound pressures around the listener position. You can convert Atmos into Ambisonics, but when you convert it back, all object data is lost. The algorithms are very different.

[@murraywebster1228]

How do you find out which plugins can be used?

[@michaelgwagner]

They need to be capable of handling multichannel audio, ideally on a per-channel basis. Tonebooster plugins can do that, for example.

[@SHAIKHAZIALTHAFCSE19-23VelTech]

can we create spatial audio that produces silence? Is it possible?

[@michaelgwagner]

Silence is spatial by definition. ;)

[@roberthiggins2162]

Would have been better to have a different sample to play back. All the echo kind of ruins the tutorial.

[@michaelgwagner]

Valid point. Thanks!