Hi there
I know that 100$ for the latest AES67 revision can be taxing depending on personal circumstances, especially just for that single doc. I still mention, if you can afford so, I'd recommend becoming an AES member which - if I recall correctly - is something like 220$ and gives you access to a lot more documents.
I think I once found an older revision published with a request for comments or so, either way, I think this is it:
Either way, Ravenna sort of is based on AES67 but extends upon it and adds a bit, most noticably on the discovery and management side of things.
at its core, according to standard, the audio stream uses RTP with L16 or L24 encoding (also see RTP RFCs/docs), SDP, practically relevant are SAP and possibly mdns (if you wanna do ravenna style).
Personally, I think the ravenna approach is quite nice, but I hear NMOS (for discovery and management) will be a thing, especially in broadcast.
Also personally, I'd recommend having a look at the relevant smtpe st2110 specs because I also consider it a meaningful and pragmatic adaptation and will likely find wider use (was it 2110-30/31 or so and something for clock?); think they cost also, but you can find some presentations showing differences/particulars. Amongst others: no RTPC, only multicast, some clock-option = 0
I started out on a framework but have been a bit lacking capacity; somewhat at the point where a PTP service would have to be implemented (or integrated), the actual audio/RTP aspects respectively. Sort of trying to make the core-codebase low-level and as usable on embedded devices as possible but also to provide some generic tools/functionality around AES67/RTP/PTP/etc (ie without having to rely on some fancy xy-systems-library). - >
https://github.com/tschiemer/aes67
Hope this helps, cheers
Philip