No question is silly.
Higher numbered video tracks override lower, so if the speaker is on video and audio tracks 1, then an image placed on video track two will be seen instead. That's the simplest approach. Dissolves from the speaker to the video inserts can also be programmed...
In this example a fade transition has been added to the beginning and end of the 'face' graphic on track two, by using the f icons at the head and end of that clip
The graphic is a transparent PNG, so the lower track is still visible in the background. That's how superimpositions, such as text overlays, are composed.
If the track two content were not transparent and the same resolution as track 1, then it would fully obscure track 1.
Record_2021_06_21_14_34_31_648.mp4