Siri is not speaking shortcut voice responses

Hi folks,

Last year I added Siri intents to my iOS app, providing class room information to the user. Worked nicely with iOS < 14: The user could define a shortcut with the corresponding action, e.g. "Next Lesson", and Siri read the result from my intent definition (voice only dialog response).

With iOS 14 Siri isn't reading responses anymore, but is only giving the feedback "Okay, viewing with...": So the shortcut is run, displaying the result provided from my app, but not reading it out loud. Unfortunately this was the main purpose of my intents, asking Siri just for some information w/o looking at the phone.

Is this a general design decision in iOS 14? What do I need to implement to make Siri (again) reading intent results? Currently I can only think of a bypass, putting the result text into the clipboard and adding a second action reading the clipboard (which is an iOS-provided one). Any thoughts?

Mattes

Accepted Reply

***! I just rechecked on my iPhone running iOS 14.5 and ... it just works as intended (and had with iOS < 14): Obviously Apple had introduced some bug with the major iOS update which has been resolved, now. Unfortunately I never found a corresponding information in the change logs.

I'm a happy camper, again...

Replies

Mattes,

i have the exact same problem although my intents are implemented using in-App Intent Handling (new in iOS 14).

i learned that if i turn the Show When Run option OFF for the shortcut, Siri will no longer speak the custom success responses, it will only say things like "Done" or "OK". Turning the Show When Run option ON makes Siri speak the custom success responses.

This behavior is acceptable if the intents are performing an action and "Done" or "OK" will suffice to let the user know that the request was competed. However, it defeats the purpose if your intent is returning information that the user requested. e.g. "When will my package be delivered?"


Hi,
Looks like there are some bad news. Look last paragraph in the article:
...
The voice-only dialog is spoken instead of the printed dialog in circumstances where the user is not looking at a screen, such as on HomePod, CarPlay, and AirPods.
...
But I'm sure that you already know that there is no way to force Siri to pronounce Result string just on iPad or iPhone. But you have mentioned that you have a bypass idea: to put string into clipboard and pronounce it. Does it work? And how to achieve it? I have had an idea to use AVSpeechSynthesizer from AVFoundation. But process fails with memory issue. Looks like it cannot be done from the background. So I'm looking for bypass as well.

But you have mentioned that you have a bypass idea: to put string into clipboard and pronounce it. Does it work? And how to achieve it?

Actually this topic has dropped from my todos - mostly because I didn't see any proper way forward. The idea via clipboard would require the user to set up a specific two-step workflow to (1.) run my app and propagate the clipboard and (2.) read its content via a factory action. I don't feel this is a viable solution.

Is anybody able to have Siri just speak a text provided by an app extension? It still looks like I'm missing a point...

Mattes
***! I just rechecked on my iPhone running iOS 14.5 and ... it just works as intended (and had with iOS < 14): Obviously Apple had introduced some bug with the major iOS update which has been resolved, now. Unfortunately I never found a corresponding information in the change logs.

I'm a happy camper, again...