We want to update some screen recordings of a coding project with some fixes to the current recordings. We can edit the screen shots but the original speaker is very busy.

To get the correct timing, it is useful to do the audio recording along with the typing/mouse movements. I'd like to record my voice along with the screen re-recording and the overdub my voice with the original speaker. We have an AI voice of the original speaker.

Text to speech using an AI voice of the original is ok - not perfect, but ok, so this seems possible?