Hi there,
I’m currently developing an App Playground for the Swift Student Challenge, and a core feature of my playground uses SpeechAnalyzer with SpeechTranscriber from Apple’s Speech framework to transcribe microphone input to text.
I understand that SpeechTranscriber performs transcription on-device and does not use the internet. However, from the dev documentation for AssetInventory, it looks like the required language/model assets may first need to be downloaded from Apple’s servers for devices that may not have it installed by default.
Since Swift Student Challenge submissions are evaluated offline, I’m unsure how this should be handled.
My main questions are:
-
If the device on which my playground is tested does not already have the required SpeechTranscriber asset installed, will transcription fail when the submission is run offline?
-
Is it safe for my submission to depend on SpeechAnalyzer / SpeechTranscriber as a core feature, or should I redesign it to work even if it is unavailable?
Thank you!! :)
So the Swift Student Challenge is not currently running, and AFAIK there’s been no announcement about a future challenge, and that includes what its rules might look like. Given that, I can’t give you an answer here.
However, you raise a good point and I’ve brought it to the attention of the folks who run the challenge.
Share and Enjoy
—
Quinn “The Eskimo!” @ Developer Technical Support @ Apple
let myEmail = "eskimo" + "1" + "@" + "apple.com"
