I've tried Whisper and Sayboard. Both are very cool for speech to text.

One issue is that I can't share pre-existing audio files into them to get the text.
Are there any foss apps that have that feature?

I've noticed Notely's feature of recording audio and transcription, which is very nice. The notes are not saved in the accessible storage, but the app's private storage. So, I currently have to export the audio and text. It also doesn't support importing audio recorded using a regular recording app.

So, is there any FOSS app that has the feature of being able to create trasncripts for pre-existing audio files? Or is too specific and would it be better to look into Termux?