Showcase: a talking head with words behind the speaker

A presenter in a black hoodie talks to camera while huge white words appear behind him, not over him. Captions run along the bottom in karaoke style, key words in red. Halfway through, the frame flips to full red with "API + MCP" in giant type, then back to the speaker.

To be clear about what this is: the presenter is a generated person, not a customer or a team member, and he is a set of still images, not filmed video. There is no lip sync in this cut.

How we made it

  1. A few generated portraits of the same presenter, at slightly different angles and lighting.
  2. Each portrait is cut out from its background, so the giant words can sit between the backdrop and the person. That layering is the whole trick.
  3. A slow push on each still, synced cuts on the beat.
  4. A stock studio voice reads the script, and the captions highlight word by word.

How your agent can make one

Create the presenter once, so every image is the same person, then generate the shots:

POST /v1/characters
{ "name": "Presenter", "description": "man in his late twenties, short dark hair, black hoodie, dark studio backdrop" }
POST /v1/images
{ "kind": "scene", "aspect": "1:1", "characterIds": ["<presenter id>"], "brief": "talking to camera, dark studio, soft key light" }

Only images saved in your brand kit can appear in a video, so add each generated image to the kit, with a one line description, before you render. If you want a real talking face rather than stills, lip sync on your own clip starts at 15 credits and is 1.25 credits a second after that.

POST /v1/explainers
{
  "brandId": "<your brand kit>",
  "aspect": "1:1",
  "captions": true,
  "voiceover": true,
  "brief": "20 second social cut. Presenter images with big words behind, captions on, a full screen card that says API plus MCP, end on a call to action."
}

What it costs

StepCredits
Presenter character4
Three presenter images12
Script2
Render with voice and captions, under a minute12
Total30

That is $3.00 on the $10 starter pack.

Get started if you want your agent making these.

talking head kinetic captions social voiceover generated images

Keep reading

Made with Second Studio

Every one of these came from a brief

Real renders from our showcase. Hover, or scroll on your phone, to watch.

See the whole showcase