AI Presenter
Give your book a face: a presenter pitches it on camera, scene by scene.
The Presenter turns your script into a video where a person looks into the camera and says your text word for word, scene by scene. You can also film a two-person podcast: the hosts discuss your book around a table, and the camera moves to whoever is speaking. You end up with a vertical, landscape or square video, ready for TikTok, YouTube or your sales pages.
How it goes
Paste what needs to be said
Write your script, one line per scene, or insert the built-in example to see the format. For a podcast, just give the topic and the tone: the writing engine drafts a dialogue of 6 to 10 lines, each sized to fit one clip. If you like, add a photo of your presenter and the cover of one of your books, pulled from your account in one click.
Set up the shoot
Pick the project format (a single creator on camera or a two-person podcast), the scene length (6, 8, 10 or 15 seconds), the frame (vertical, landscape or square) and the image and video models. Every option shows its price in credits, and the full estimate appears before you launch.
Launch and follow along
Generate the plan and confirm production, or tick the full automatic run: everything chains together without further approval and you track progress in My productions. You can leave the page at any point, the work carries on server-side and your plan stays saved on your account.
What this feature does
Your text, word for word
The presenter says your script to the camera, in the language you wrote it in. No voice-over layered on top: they are the voice of the video, backed by light sound design.
The same face from start to finish
If you provide a photo, it is used as a real reference and never redrawn. Without one, a credible, consistent creator is defined for you, matched to your script's subject.
Your cover in one click
The cover of any book created on the platform can be pulled straight from your account. The presenter holds the real book; in podcast mode it stays closed, cover visible, and passes from hand to hand.
Two hosts, eleven profiles
For the podcast, pick who hosts and who is hosted from eleven profiles, from a young man to a woman of experience, robot and animals included, or describe your own. The camera moves to whoever speaks while the other listens in frame.
The cost known before you launch
The price in credits sits on every option, and the full per-scene estimate appears before launch. Mid-run, a button stops generation: only the work already done is charged.
Scenes timed to the second
Each script line becomes a clip of the length you chose, with a per-scene gauge that flags lines running long. Switch the pace and the script is re-cut on the spot.
Three situations where it helps
Selling a recipe book on TikTok
Camille publishes a book of family recipes and wants vertical videos for her Reels. She pastes her sales script, pulls her book cover from her account and picks the in-action mode. Her presenter holds the actual book, shows it to the camera and says every line on screen. She then spins out several hooks with the same face.
A filmed podcast about a guide
Julien, a course creator, wants a calmer video for YouTube. He picks the two-person podcast, describes his hosts and lets the engine write a short dialogue about his guide. The book sits on the table, passes from hand to hand, and the camera follows whoever is speaking while the other listens. He gets a believable slice of conversation with no studio to book and no guest to find.
An episode series with one recurring face
Aminata, a coach, releases one commented chapter a week. She provided a photo of her character, so the same face returns in every scene and is never redrawn. For each new episode she reuses the universe of the previous project, and references already produced are neither regenerated nor billed again. Her series keeps a stable identity week after week.
Who it helps
Common questions
How much does a video cost?
The price depends on the models you pick, and it is announced before every step: each option displays its cost in credits, and the full estimate (image plus video per scene) appears before launch. Bible references (characters, settings) come on top and are itemised at the production stage.
Can I use my own face?
Yes: upload a photo and it serves as the reference in every scene, never redrawn. The result is still a generated video, not a recording of you, so judge the look on one short video before committing to a series.
Which formats and lengths are available?
Three frames: vertical 9:16 for TikTok, Reels and Shorts, landscape 16:9 for YouTube, and square 1:1. Scenes run 6, 8, 10 or 15 seconds; some video models do not cover every length and are then shown greyed out, with the reason.
What if I have no dialogue for the podcast?
Say in one sentence what your two hosts are talking about, add the tone if you wish, and the engine writes an alternating dialogue of 6 to 10 lines, each sized to fit one clip. You read it over and edit anything you like before launching.
Can I add background music?
Not in this format: speech stays front and centre, with light sound design but no music. That choice keeps a soundtrack from drowning out the presenter's voice.
What happens if I close the page?
Nothing is lost. The plan keeps running server-side, tracking resumes on its own when you return, and your plans stay saved on your account, reachable from any device. You can also stop generation at any time: only the work already done is charged.
Fifty credits free when you sign up
No card required. A browser is enough, on desktop or on your phone.
Create my free account