The correct term for this is ADR (automated dialogue replacement) or, it is also referred to as “looping” which goes back to the day of film and mag reels for sound, when they would create a loop of the lines in question and load that on the projector so the actor could keep going over and over until they got it right.
Anyway the standard way of setting up for the actor is to have a set of 3 beeps placed at even intervals before the line, so the line would start on the 4th beat. To keep it simple, create a 1khz tone 2 frames long then place that at 1 second intervals before the line, so the line would start on the 4th second. You need to be able to send that to some headphones for the actor to hear so they get the pacing of the beeps, so as to know when to start speaking. You could also create what is known as a “streamer”, which is a visual cue. It is a solid vertical line that travels across screen from left to right. The main thing is to have some sort of reference or count in for the actor so they will know where to start speaking.
[Espnetboy3] “(Perfect Storm I heard was shot with no audio at all until they got into the studio Post)”
That would be incorrect! They would have still shot with audio even though they knew most of it would be unusable due to the large fans and water noise on set. If you dont record the sound, no matter how bad you know it will be, you will not have any reference audio to sync up the ADR to later in post. Also the editorr wouldn’t know what the hell was being said if all they had was a set of moving lips. Unless of course they could lip read-:)