One person tells their story, looking at an interviewer beside the camera. Use it for portrait documentaries, brand stories, and craftsperson introductions.
In Shot 2, place the person on the right third line with their gaze toward the interviewer to the left of the camera. Space is left in the direction of the gaze, so the frame doesn't feel cramped and it feels like a conversation with someone.
To recreate it: Write the position and gaze direction together, as in 'placed on the right third, speaking to someone just off camera left'.
Light from a high window on the left falls on one side of the face, leaving a small triangle of light on the shaded cheek. The face gains dimension, and the speaker looks serious and thoughtful.
To recreate it: Write 'small triangle of light on her shadow-side cheek' as is, so the AI doesn't light the whole face flat.
In Shots 2 and 3, blur the workshop behind so the eye goes only to the speaking face and the working hands.
To recreate it: Include 'softly blurred workshop behind', and decide the placement of the background objects to be blurred in the start frame.
On the first interview shot, show the name and occupation for 2-3s so viewers know who is speaking. Since the person is on the right, put the text in the empty lower left.
To recreate it: Don't generate the text. Add it in editing.
Keep the voice as is and cover only the picture with the hands in Shot 3 and the bowl in Shot 5. It cuts the boredom of a long talking face and shows the story in pictures.
To recreate it: Lip sync tends to drift, so keep the talking-face shots short and lay the rest of the voice over hand and prop shots.
Carry the end of the previous sentence's sound into the next picture after the cut, so the story doesn't break.
To recreate it: Leave the voice track about 1s longer than the video and let it run over the next shot.
In Shot 4, approach a thoughtful face slowly to stress the moment the emotion deepens. Before this shot, pause the speech briefly and let only the sound of the workshop play.
To recreate it: Instruct only 'slow push-in' and use it under 3s, so the face changes less.
Cinematic multi-shot sequence, 15 seconds total. Shot 1 (3s): static wide shot of a small pottery workshop at dusk, a potter at the wheel, warm work lamp; Shot 2 (4s): static medium close-up of a woman potter in her sixties speaking off camera left, placed on the right third, window light leaving a small triangle on her shadow-side cheek; Shot 3 (3s): close-up insert of wet clay rising between her weathered hands; Shot 4 (3s): slow push-in on her thoughtful face as she looks down; Shot 5 (2s): static close shot of finished bowls on a windowsill, backlit dust floating. Consistent woman and workshop, soft window light, 35mm film grain, sincere quiet mood.
Light from one side and above, plus a blurred background, gives the face depth and keeps attention on the speaker. Lay the voice over shots of working hands, and the story becomes visible.
Wide static shot of a small pottery workshop at dusk, shelves of unglazed bowls, a single warm work lamp and soft window light, a potter sitting at the wheel in the midground, deep focus, soft 35mm film grain, quiet honest mood.
Medium close-up of a woman potter in her sixties speaking to someone just off camera left, placed on the right third of the frame, window light from high on the left leaving a small triangle of light on her shadow-side cheek, softly blurred workshop behind, static camera, 35mm film look, sincere mood.
Close-up insert of wet clay rising between two weathered hands on a spinning potter's wheel, static camera, warm work lamp from the side, shallow depth of field, rich earthy tones, fine 35mm film grain, calm focused mood.
Close-up of the same potter's face as she pauses and looks down thoughtfully, slow push-in, soft window light from the high left with a small triangle of light on her shadow-side cheek, dark background, 35mm film look, reflective mood.
Static close shot of finished ceramic bowls lined up on a windowsill, late afternoon backlight glowing through the edges of the glaze, dust floating in the air, shallow depth of field, warm muted tones, 35mm film grain, peaceful closing mood.
On the first interview shot, show the name and job as a lower third for 2–3 seconds. When the story gets long, keep the voice as it is and cover only the picture with B-roll overlay / cutaway shots of hands and the workshop. Even after the cut changes, carry the end of the previous sentence over with an L-cut, so there is no break. Before the pensive push-in shot, pause the speech briefly and let only the workshop sound be heard.
AI faces change slightly from shot to shot, so choose one portrait image and make the start frames of all shots by editing that image. Use only a locked camera or one slow push-in. Lip-sync, matching the mouth to the voice, easily slips, so keep talking shots short and lay the voice over hand and prop shots chosen from 480p drafts. It looks like a real person speaking, so say in the video description that the person is AI-generated.