Talking-head videos
Use your own on-camera recording as the presenter and narration for a video.
On this page
#Bring your own take
A talking-head upload is a video of your presenter speaking to camera. Kinema uses the recording’s audio as the narration and the recorded person for on-camera sections. You do not need to generate a narration voice or a photo-based avatar for that take.
#Upload and configure
Attach the video
On Dashboard, choose + → Talking-head video, or drag the video onto the page.
Check the recording
Wait for its duration to appear. Hosted uploads must be 6–40 minutes and no larger than 4 GB.
Confirm the spoken language
Select the matching language using the flag at the top of the composer.
Review the source selection
Attaching a talking-head video switches the supporting visual selection toward AI media. Open Sources and adjust it for your project.
Set coverage and create
Use High on-camera coverage if you want more of the presenter in the film. Check the summary and credit estimate before creating.
#Prepare the recording
- Use a take with clear speech and a consistently framed presenter.
- Trim unwanted material before attaching the file so the recording represents the video you want to build.
- Check that you have permission to use the person’s appearance and voice.
- Keep a copy of your original recording. Remove attachment in the composer if you need to choose a different take.
#Review the presenter and supporting visuals
The first cut may combine the presenter with supporting visuals. In the editor, review full-frame, split-screen, and corner-presenter scenes. Replacing supporting media in a composite presenter scene should preserve the presenter and the layout.
A real talking-head upload does not incur the AI presenter generation fee. Its video production and selected extras are still reflected in the confirmation estimate.