Street interview / vox pop: structure, script and how to clone it

Street interview / vox pop is a proven short-video structure. Here is why it holds attention, how its beats are timed, a script you can start from, and the command that clones it for your own subject.

Why this format works

Man-on-the-street formats work because the viewer is waiting for the awkward answer. Curiosity about other people does the retention work for you.

Structure breakdown

  1. 1. Cold open (0–2s)

    Start mid-answer, not with a setup. The question appears as an on-screen caption instead.

  2. 2. Question–answer loop (2–40s)

    Three or four people, 6–10 seconds each. One question repeated; answers escalate in surprise.

  3. 3. Best answer held back

    Save the wildest or most relatable answer for the final slot.

  4. 4. Text payoff (last 3s)

    Repeat the question as a caption so the clip works with sound off.

A script you can start from

Segments, takes and selection anchors — this is the part of the project that changes when you re-run the format:

<script>
<segment> id="ask">
  <take>How much do you think || @rent-amount this flat costs? || (to camera) Watch. || She says || @too-cheap eight hundred. || <D | Dee> It is || @actual-price four thousand.</take>
</segment>
</script>

What you need

  • — 3–5 short interview takes (or a single presenter voicing both sides)
  • — B-roll of the location
  • — Caption track for the repeated question

Run it

hypit run clone --reference ./reference/vox-pop.mp4 --script ./vox-pop.svs --run --keep

The reference video stays fixed. Change the script, re-run, and you get a new edit in the same format — that is the whole point of cloning instead of prompting.

Keep reading

FAQ

What if I do not have real interviewees?

You can run the format with a single presenter playing both question and answer, or generate the replies from a script. The format is the question–answer rhythm, not the crowd.

Why start mid-answer?

Because the first second decides whether the clip is watched. A cold open skips the setup and drops the viewer into the payoff.

How do captions fit this format?

Captions carry the question and the numbers, which is what makes the clip legible without sound. The caption track re-renders with each take.