Create your video now·render in about a minute·no watermark on any plan
AI Poem Video Generator
Paste your poem, or name a subject and a form, and get it read slowly by a natural AI voice over a painted scene for every stanza, captioned as it is spoken.
01Paste the poem, or name the subject. Your own poem is kept word for word. A subject, a form and a length get an original poem written for you to edit.
02Read where it cuts. The scenes come back split at stanza breaks, before anything renders. Move a break, change a line, pick the voice.
03A painted scene per stanza. Each stanza gets its own watercolor scene, the voice reads it at its own pace, and small captions follow every word.
Voice:
Elli
Captions:
Subtle
Music:
Lo-fi
Format:
Vertical 9:16
What comes out
The settings this page ships with, and the shape they produce
Every example on this page is labelled with the exact configuration behind it, including the one that says it is a drawing.
Layout mock-up
A short rhyming poem about the first snow of the year, rea…
Style: Watercolor
Voice: Elli
Captions: Subtle
Ratio: 9:16
A mock-up of the frame and caption layout at this page’s settings — TikTok, Reels and Shorts, lo-fi music bed. It is a drawing, not a rendered video: we have no published render library to show you yet, and a stock clip dressed up as our output would be worth less than saying so.
Your input
“A short rhyming poem about the first snow of the year, read slowly, four stanzas”
Plus this page’s brief, verbatim
Make this a poem read aloud. If the user pastes a poem or says to keep their wording, keep every word, line and stanza exactly as written: do not fix a rhyme, smooth the meter, modernise a spelling or add a title, and never add a fact, a dedication or a line the user did not write. Split it into scenes at stanza breaks, or every two lines for a poem without stanzas, so each scene is read as one unit. If the user gives only a subject, write an original poem in the form and length they ask for (otherwise three or four short rhyming stanzas), in plain modern language, with concrete images rather than abstractions. Never recite a published poem from memory; use another poet's words only when the user pastes them. Each scene's picture shows the image in its own lines — a place, an object, weather, light — with no text in the frame and no real person on screen.
That string is not marketing copy — it is the instruction the script writer is given before it reads a word of yours, and it is the only thing separating this page from the other forty-seven.
The shape that comes back
Scene 1 — the hook
One line that works with no context, because the viewer arrived mid-scroll.
Scenes 2–6 — the body
One idea each, one or two spoken sentences, each with its own image prompt and its own duration.
Final scene — the close
A line that lands the point rather than asking for a follow.
Per scene — the shot
The exact image prompt the model would receive, with this page’s style suffix already appended.
This is the structure, not a saved example. The box at the top of the page writes the real one — with your wording, in about ten seconds.
5–8 scenes
each with its own line, its own frame and its own length
Under 60s
cut to the narration, not to a fixed per-scene timer
Word-timed
captions burned from the voice model’s own alignment
1080-class
vertical, square or landscape — no watermark on any plan
01
Your poem, word for word, cut where the stanzas break
A poem is the one kind of text where a tool should change nothing. Say to keep your wording and the brief keeps every word, line break and stanza exactly as you wrote them: no rhyme is fixed, no meter smoothed, no title added. The poem is only split into scenes — at each stanza break, or every two lines if it has none — and you see the split before a single frame is painted.
Each scene runs exactly as long as the voice takes to read it, a length that comes back from the voice model with the audio rather than being guessed, so a long stanza holds the screen longer than a short one. Captions are small and low in the frame, timed to the words as they are read, so the picture and the line carry the video rather than the text on top of it.
02
Or a new poem, written to be read aloud
Name a subject and a form — rhyming couplets, a sonnet, a run of haiku, unrhymed lines, a limerick for a birthday — and an original poem is written in plain modern language with concrete images, because a listener cannot reread a line. Without a form it writes three or four short rhyming stanzas. It never recites a published poem from memory: a model will misquote it, so another poet’s words go in only when you paste them.
A written poem is a first draft, not a finished one. Language models rhyme easily and count syllables badly, so read it before you render: change a line, cut a stanza or rewrite the ending at no cost, then say to keep your wording and the rest of the video is built on the poem as you have left it.
03
What it reads, what it paints, and in which voice
Every scene is a watercolor painting of the image in its own lines — a kitchen at night, snow on a bus shelter, the last star over a bakery — with no words in the frame and no real person on screen. Choose another look in the composer if the poem wants it. The voice reads, it does not sing: a poem set to music is a song, and this is the wrong tool for one. There are 5 voices, all recorded in American English. There is no British, Irish or Australian voice and no voice cloning. The narration and captions can be in any of 29 languages — French, German, Spanish, Italian, Dutch and Polish among them — chosen in the composer; the voices read them, but a native listener may hear the American accent, so listen to the first one before you post a series.
Someone else’s poem is theirs: in the US, poems published more than 95 years ago are in the public domain, and in the UK a poem is out of copyright 70 years after the poet’s death. Anything newer needs the rights holder’s permission before you post it. Your own poems, and poems written for you here, are yours. Every video is made on a plan: Mini is $7 a month, and the first payment on any plan has a 7-day money-back guarantee.
Who this is for
Four people this was built for
Poets on Instagram and TikTok
Each new poem as a read-aloud video in the same voice and painted look, without recording it on your phone.
English teachers and students
A poem the class is studying, or one a student wrote, read slowly with a picture for every stanza.
Occasions
A poem for a birthday, a wedding toast, an anniversary or a memorial, played on a screen or sent to the family.
Writers and small presses
A poem from a new collection as a short video that points readers to the book.
Compared
What this replaces
Six things somebody has to arrange to make one short, and who arranges them.
Input
The usual way: Record yourself reading the poem, then find or shoot a picture for every stanza.
With Veedio: Paste the poem, or name a subject and a form.
Visuals
The usual way: Stock photos that never quite match the lines, or an evening of drawing.
With Veedio: A watercolor scene painted from each stanza’s own images.
Voiceover
The usual way: Several takes in a quiet room until one has the right pace.
With Veedio: 5 AI voices with previews, read at each stanza’s own pace.
Captions
The usual way: Auto-captions that drift, then an hour nudging keyframes.
With Veedio: Burned in from the voice model’s own word alignment, so they land on the syllable.
Turnaround
The usual way: Days, and most of it waiting on someone else.
With Veedio: About a minute, and you watch it fill in scene by scene.
Changing it
The usual way: Re-record the line, re-cut the timeline, re-export, re-upload.
With Veedio: Edit the line and only its voice is re-recorded, free. A new still is free too; a new motion clip is free once per video.
The usual way
With Veedio
Input
Record yourself reading the poem, then find or shoot a picture for every stanza.
Paste the poem, or name a subject and a form.
Visuals
Stock photos that never quite match the lines, or an evening of drawing.
A watercolor scene painted from each stanza’s own images.
Voiceover
Several takes in a quiet room until one has the right pace.
5 AI voices with previews, read at each stanza’s own pace.
Captions
Auto-captions that drift, then an hour nudging keyframes.
Burned in from the voice model’s own word alignment, so they land on the syllable.
Turnaround
Days, and most of it waiting on someone else.
About a minute, and you watch it fill in scene by scene.
Changing it
Re-record the line, re-cut the timeline, re-export, re-upload.
Edit the line and only its voice is re-recorded, free. A new still is free too; a new motion clip is free once per video.
Related tools
Same engine, different preset
Twenty more pages over the same pipeline. Each one is a different brief, a different look and a different page — and each one prints its brief the way this one does.
Paste text you already wrote — a post, a newsletter paragraph, a caption, your notes — and get it rewritten for the ear and made into a voiced, captioned 9:16 TikTok.
Not if you say to keep your wording. Every word, line break and stanza is kept exactly; the poem is only split into scenes at its stanza breaks, and you see the split before anything renders.
Can it write the poem for me?
Yes. Name a subject, a form and a length and an original poem is written in plain language. Read it before rendering — models rhyme easily and count syllables badly — and edit any line at no cost.
Can I make a video of a famous poem?
Paste the text and it is read as written. In the US, poems published more than 95 years ago are in the public domain, and in the UK a poem is out of copyright 70 years after the poet’s death; anything newer needs the rights holder’s permission before you post it. It never recites a poem from memory, because a model will misquote it.
Can the poem be sung, or set to music?
It is read, not sung. A quiet music bed plays under the voice, and you can change it or turn it off in the composer, but there is no singing voice and no song.
Is there a British voice?
There are 5 voices, all recorded in American English. There is no British, Irish or Australian voice and no voice cloning. The narration and captions can be in any of 29 languages — French, German, Spanish, Italian, Dutch and Polish among them — chosen in the composer; the voices read them, but a native listener may hear the American accent, so listen to the first one before you post a series.
Can I use my own voice?
No. There is no voice cloning; every narration is one of the stock AI voices. No video or audio can be uploaded: every scene and every voice track is generated. If you already have footage and want a voice laid over it, this is the wrong tool, and we would rather say so before you pay.
What does it cost?
Plans are monthly: Mini $7 (200 credits), Starter $19 (600 credits), Pro $49 (2,000 credits), Studio $149 (7,500 credits). A five-scene short in pictures costs 13 credits; with AI motion in every scene it costs 68. You see the price before you generate, and the first payment on any plan has a 7-day money-back guarantee. Mini ($7 a month, 200 credits) covers about 15 five-scene shorts in pictures, or about 2 with AI motion in every scene.
Can I post it to Instagram and TikTok?
Connect TikTok, Instagram and YouTube once and publish to TikTok, Reels and Shorts from the app, on every plan. Mini connects 1 account, Starter 3, Pro 6 and Studio 25. Or download the video and post it anywhere. You own the videos on every plan and can use them commercially, paid ads included. There is no watermark on any plan, the export is 1080p, and the music beds are licensed for commercial use with no attribution.
Do I have to sign up to try this?
Yes, and pick a plan: Mini is $7 a month, and the first payment has a 7-day money-back guarantee. On a plan, the box at the top writes the script and the shot list for no credits — that is the part that decides whether the video is worth making.
What does it cost?
Mini is $7 a month for 200 credits, Starter $19 for 600, Pro $49, Studio $149. A credit is a unit of pipeline cost rather than a video, so a five-scene short costs less than a nine-scene one, and you see the estimate before you spend anything.
Do I own the videos, and is there a watermark?
You own them, on every plan, and there is no watermark on any plan, no resolution cap and no platform withheld. Four things do follow the plan, because they are what a video costs to make: AI motion and sound effects need Starter, and a presenter avatar and the choice of video model need Pro. Everything else — every voice, every language, every caption style, every aspect ratio — is the same on all of them.