Okay I get it. Let me show you this. This is something I'm working on for a short film coming up in the near future for my AI band. I posted this on another thread to somebody who says you can't build camera control into AI clips, but you can. I'll show my work below. The nuance: that's the struggle with AI media, capturing the nuance, and if you're good at it, it looks better.
Prompt is included for reference. You can get a lot more complicated than this but that's the idea. Notice the "two shot" and "medium" and "close" verbiage. That's the method R02b, six shots, cuts at 4/9/13/17/21. Business + tics written in.
1 two shot 0-4: Jenny folds her arms again, looks at the floor, then at him. JENNY: "We're not a band that adds people."
2 Bear medium 4-9: hands on his knees, glance at the door and back to her. BEAR: "Yeah, well. ... He wrote one with a fiddle in it."
3 Jenny medium 9-13: she looks away at the glass and back. JENNY: "...We'll cut it without the fiddle."
4 Bear medium 13-17: leans back in the chair, a short laugh through his nose. BEAR: "He won't."
5 Jenny close 17-21: she exhales, a small shake of the head. JENNY: "Everything else has banjo and, uh... nobody plays that either."
6 two shot 21-25: Bear stands up out of the chair. BEAR: "That's a different problem." Jenny, after a beat, walking toward the door: JENNY: "Tuesday. ... I'll go. I'm not promising anything."