Is ANYONE using audio transcription (Whisper AI import) in sentence mode?

Audio transcription via WhisperAI is one of the premium features
Is anyone using it in sentence mode?
For me it is completely broken. I mean, nine times out of ten the audio and the text of a sentence don’t fully match.

The following link which says “Filen” is actually a video that demonstrates the problem. This was from importing an mp3 file.
I have an open ticket to support about it, but I’m really curious to hear if ANYONE is able to use this paid feature the way it was supposed to work