As far as I know, Youtube uses first order Ambisonic (FOA) rendered to binaural.
You can create such FOA content via Ambisonic microphone(s) and/or via synthesis.
Both are supported by Panoramix.
For synthesis :
- create a FOA bus; send your (mono, stereo, etc.) tracks in the bus; then export the FOA encoded stream to Master (“Routing…” button on the bus)
For microphones :
- when the mic signals are not encoded, this is called A-format.
So, create an A-format track, and send it to the FOA bus.
- if the mic signals have already been encoded (e.g. via Ambeo plugin), create a “Direct to Bus” track with 4 channels, and send it to the FOA bus.
In Panoramix, you can preview your mix in binaural (“monitoring” button on the FOA bus), however the rendering might differ from what Youtube does.
Last but not least : be aware that there are various “flavors” of Ambisonic in use.
So, you need to check which convention Youtube expects, and make sure everything is produced accordingly.
Things to check :
- channel ordering (ACN, SID, FMH)
- Ambisonic normalization (SN3D, N3D, FuMa, etc)
- coordinate system : Spat uses Y-axis as forward direction; most other tools use X-axis instead
(see e.g. spat5.tuto-bformat.maxpat for further details)
Conversions between Ambisonic “flavors” are easily done (spat5.hoa.sorting~, spat5.hoa.converter~, etc); just be careful about it.
I believe (to be confirmed) that Youtube uses ACN, SN3D, X-forward.
Let us know how it goes.