AirPlay 2 disappears from newest iOS 11.3 & tvOS 11.3 betas
Despite prior support in Apple's development track, AirPlay 2 appears to have been removed from the third iOS 11.3 and tvOS 11.3 betas, issued earlier on Tuesday.

Beta testers can no longer take advantage of features like multi-room audio, or including an Apple TV in the iOS Home app, iDB noted. Apple has yet to provide an explanation.
It could be that AirPlay 2 support is being postponed until 11.3.x or later, since it had a rough implementation in recent betas. Alternately it could be reintroduced in upcoming 11.3 beta seeds, and was simply taken out briefly for polishing.
Once it's finished AirPlay 2 should provide an Apple-based alternative to platforms like Sonos or Google's Chromecast Audio. It was originally announced at WWDC 2017, but didn't make it into the initial release of iOS 11 in September.
It also failed to materialize in time for this month's launch of the HomePod, for which the standard was likely conceived.

Beta testers can no longer take advantage of features like multi-room audio, or including an Apple TV in the iOS Home app, iDB noted. Apple has yet to provide an explanation.
It could be that AirPlay 2 support is being postponed until 11.3.x or later, since it had a rough implementation in recent betas. Alternately it could be reintroduced in upcoming 11.3 beta seeds, and was simply taken out briefly for polishing.
Once it's finished AirPlay 2 should provide an Apple-based alternative to platforms like Sonos or Google's Chromecast Audio. It was originally announced at WWDC 2017, but didn't make it into the initial release of iOS 11 in September.
It also failed to materialize in time for this month's launch of the HomePod, for which the standard was likely conceived.
Comments
Randomly disconnects while streaming YouTube videos.
It's fine using Airfoil, however.
And I am increasingly annoyed by Apple announcing features that never make it to the intended iOS release.
A few examples: above mentioned Airplay 2, iMessages in the Cloud, People sync in iOS10...
It would be nice if the iOS had a release schedule independent of hardware, that updates were to actually fix things, and that new features are introduced progressively, instead of announcing them, putting them in betas, and then remove them or postpone them to the next iOS x.0 release...
For me Airplay didn't prove reliable enough.
But Apple did announce a new approach to software development and release, which means the company realises that there is a problem it needs to fix. I really appreciate the fact that they are looking at the problem and at ways to fix it. Hopefully it will also fix the increasing issue of announcing new features but then not delivering them in what seems like a reasonable timeframe.
I actually do recognise that Apple is having to scale its operations to meet the demands of millions and millions of users and devices. I take these issues as just growing pains and, overall, they don’t make me think that Apple has « lost it » and is swirling around the drain.
So isn’t it a bit odd to complain about Apple (or any company) announcing a feature before putting out the beta versions, and then being impatient and upset while they take the time to get it right? Isn’t it doubly odd to complain about beta software being buggy?
Seriously, how would you have Apple (or any company) work through this process? You know, besides getting everything perfect in alpha stage, skipping beta, and going from announcement to final release immediately, with all features included and working perfectly.
2) Wouldn't it be transparent to the user. It's not like the customer needs to know that an app was made using Swift 4.0 or Swift 3.0.
FWIW I pretty much agree with you, take time to get it right, but by announcing last year all the wonderful things Airplay2 was bringing to the Apple community they created high expectations for it and being right around the corner, expectations that are taking a bit long to materialize. Silence on the matter isn't helping.
Right about now, there likely are probably some fairly intense conversations in the new building about beefing up in-house alpha testing of new features and products. That approach is subject to diminishing returns, however, because it's the act of putting something out in the wild that turns up the variables no one in-house thought of or could simulate. That brings us full circle to the need to make public announcements and do the dog-and-pony show before putting out something significant in a beta release. People will notice it if they're testing it, and a dog-and-pony show done later about a feature that's been tested and discussed in public for months really doesn't have the same impact, does it?