I will work on multiple projects at different times in the day; an ideal workflow would be to have a voice memo for each project that day and I progressively add to that projects memo through the day, then process the memo at the end of the day with notes for that project. Additionally, I have found that, for long recordings (90s+) the screen shuts off on my phone and the superwhisper recording gets lost.
I think a simple way of enabling this would be (similar to the Mac App) to allow selection of an on-device M4a/memo to be transcribed and processed.