First of all, there are many substantive errors. Upon the first reading, I found 2 mistakes per page of 11-point text.
Secondly, it's the quality of conversational English since it's for YouTube, meaning it should be spoken on camera. The translation came out very literal, technical, like sentence by sentence. I constantly caught myself thinking that it sounds somehow unnatural. It's fine on paper, but in real life, people don't speak like that. Of course, I can't make edits like "sounds bad" because how do I justify them? So the text needs to be carefully proofread and revised.
As a result, after Google Translate or chat GPT, I would be refining, which is what I'm doing here. Well, maybe a little less.
I've come to the conclusion that it's better to pay 2 or 3 times more but have the performer work, not me.
I just have something to compare it to - the previous script was translated by another performer, just a couple of corrections and it's ready. The text sounds on camera as if it flows.