We are attempting to update our app to use the PTT framework, as it has been made clear that this will be required in a future iOS version as opposed to using the Unrestricted VoIP entitlement we are using for several features of our app.
However, the behavior of this framework poses some problems with implementing our app's functionality:
-
It is not possible to programmatically join a channel when the app is not in the foreground. This hinders our ability to implement the Automatically activate radio stream feature of our app, which allows users who have opted into this feature to immediately begin hearing live PTT audio from their agency following an incident alert. Having the app constantly "joined to a channel" and using the restoration delegate could potentially work, however this is not ideal as this would result in the PTT UI needing to be displayed at all times, even when no radio stream is activated.
-
We have a "Text to Speech" option that, when enabled, reads out the content of an incident alert after the alert sound has played. This currently happens by triggering an AVSpeechSynthesizer in the PushKit incoming push callback. It may be possible to render TTS audio on the fly in a Notification Service Extension and assign it as the notification's sound, if that is possible this is less of a problem.
-
We also use the PushKit callback to, again if the user has enabled it, activate a "Shake to Respond" feature, allowing a short period of time after receiving an incident alert in which the user can shake their device to indicate that they are responding to the incident. There does not appear to be any way to have the level of background execution required to implement this using an NSE, and this is of course beyond the scope of the PTT framework.
What options do we have to be able to continue to provide this functionality, without risk of it being disabled in a future iOS version?