Along this same line I'll use MacWhisper to dress things up to keep the actor or narrator from sounding buffoonish.
Create SRT file. Import into project. Search for those ever so popular but annoying... you know or like or uhhh cow pies in your video pasture and silence (video) or remove them (audio). On video you won't notice since uttering these generally don't involve obvious lip or jaw movement.
The beauty of this approach is you need not burn time scrubbing or dutiful listening. You may miss a few but you won't need to clean your boots using a stick after you're finished with the end result being the talent sounding like Alistair Cook.