Voice to text / dictation problem

Hello Community!

We have a strange issue. At least two of our users have reported a problem when they attempt to use the dictation feature of iOS, also known as voice to text, to add text to a text field. It begins to accept voice input and then about 2 seconds later, a tone is played and the device stops accepting voice input. This function worked fine previously, and the problem doesn't happen in other apps on the device. We were able to duplicate the behavior on a test device, and the users report this problem did not occur in the past.

Is it possible we accidentally disabled or hindered this function somehow? Is there a setting or function in OutSystems that might have resulted in this behavior that we could look at? Has anyone else seen a similar problem? If so, how did you resolve it? 

I look forward to any insight you might have. Thanks.

Hi,

Is this a mobile app or web app? are you using any mobile plugin?

Thanks.

It is a mobile app. We are using the ImagePicker Plugin.

Hi Jared,

So you are not using any plugin to get the voice to text. Are you using the mike icon found in the keyboard for voice to text?

Thanks.

We are experiencing this issue as well.  It surfaced for us in iOS 13.  It's any text field, when you use the dictation button, it only listens for about 2 seconds.  Is this an Outsystems bug from iOS 13?

Hi,

looks like it’s a Cordova issue and might be affected for devices running on iOS 13. Refer the bug in link below

https://github.com/apache/cordova-ios/issues/682

Thanks.

So, following the thread posted by Prasad, my best guess is that the apps that my current build of OutSystems is producing use a deprecated UIWebView API that is deprecated by Apple and is causing errors in iOS13. Apple recommends using the newer WKWebView, and other developers who have moved to that API report the issue does not happen in their apps. 

Can anyone tell me if upgrading to a newer version of the platform fixes this bug? We already plan to upgrade next week in order to take advantage of the new features we saw at the NextStep conference, so if nobody knows the answer, I will report back at that time.