Since the built-in speech-to-text transcription plugin in Premiere Pro is only ~65% accurate (if you count the punctuation mistakes) — which in mid-2025 (the age of reasoning super AIs is quite unprofessional for such a large company), why don't you switch to using Open AI's free Whisper API app? All the processing is done on the user's computer too.
Free instructions on how to use it:
https://www.youtube.com/watch?v=ABFqbY_rmEk
You Could (but don't have to) also add an option to occasionally use the Cloud (like you do with the option in Photoshop to use local or Cloud-based Generative AI whenever it's needed) for more demanding audio transcription jobs.
AdChoices