Did you try to verify if the microphone permissions are explicitly granted in the iOS Teams app settings (device-level settings)? You can check them using the navigator.permissions.query
API.
Run the speech recognition code on Safari (iOS) to verify that it works independently of the Teams environment.
If getUserMedia
doesn't work as expected, consider implementing native iOS speech recognition using the Apple Speech framework within a custom integration.
As an alternative, you can create a bot that can join the meeting and capture audio for processing by Azure AI Speech.
If critical, consider integrating native iOS functionality (using Swift or Objective-C) and calling it through Teams SSO or an OAuth flow.