Use OpenAI's Whisper neural network

Open AI recently open sourced their neural network that is incredibly good at making transcriptions and translations in english including english spoken in a variety of accents

Here is a link to Open AI's blog post on Whisper https://openai.com/blog/whisper/

There is already a script to auto translate any videos that work with youtube-dl https://github.com/m1guelpf/yt-whisper

and here's a fork of Whisper made to make it simple to use which might be helpful as well https://github.com/Stage-Whisper/Stage-Whisper