Skip to content

How to turn a PDF into a lesson video

By VitrinAI Digital Studio ·

A recording of the live panel · five steps

On this page13 sections · 9 min read

Short answer: To make a lesson video from a PDF, you take the text out with its page and line, choose the sentences the video will say from those lines, put the scenes in a scene plan and approve it, and then make the narrated, subtitled video. In Zıpır your part is to upload the PDF and approve the scene plan; the 28-page 6331.pdf became a video of 8 scenes this way.

Why is a PDF a hard source?

A PDF is designed to be read on a screen, so getting the text out is hard. Page headers, columns, footnotes and words split at the end of a line get mixed into the text. For someone who wants to cite sources the real problem is this: the text has to come out without losing which page and which line a sentence is on. Zıpır extracts the text line by line and ties every on-screen line to its place in the document.

1. Upload the PDF

You drop the document on the “New video” screen in the panel. The video title is optional; if you leave it empty, the document's name is used. The example in this guide is a PDF with a text layer: 6331.pdf (28 pages). The code has text recognition for scanned PDFs, but it has not yet been verified end to end in the live product.

2. Choose a format

The product has twenty-one visual formats: Cartoon, Isometric, 3D cutaway, Paper cut, Whiteboard, Branded workplace, Map story, Illustrated explainer, Clay, Mascot host, Cause and effect, Panel presentation, Chalkboard, Balance, Showcase, Pictogram, Timeline, Brand colors, Document walk-through, Summary presentation and Host-led cartoon. See all of them on the formats page. For example, Summary presentation shows one principle per scene in large type, and Document walk-through shows the page itself with the quoted lines marked. You can leave the choice to Zıpır. The example video was made in the Summary presentation format.

How do I set the length of the video?

Short answer: Under the format choice you pick a target length: 1, 3, 5 or 10 minutes; the default is 3 minutes. The scene plan is built to that length, and you see the estimated length on the approval screen.

The same document with a 1-minute target says fewer sentences; with 10 minutes it covers more of the document; either way, every sentence on screen stays a quote from the source. If the plan goes over the target, the video is cut back to under the target; if the document is not long enough to fill the target, the video stays under it and says so.

3. Approve the scene plan next to the source

Zıpır first brings a scene plan: for every scene, the quote it will show and where that quote is in the PDF. The sixth scene of the example video:

“Tehlikeli olanı, tehlikesiz veya daha az tehlikeli olanla değiştirmek.”

6331.pdf, s. 4, satır 17

In English (our rendering): “Replacing what is dangerous with what is not dangerous or less dangerous.” Source: 6331.pdf, p. 4, line 17.

If a quote cannot be found in the source, its scene is not made. If everything is right, you approve; before approval not a single frame of video is made.

4. Get the video and the source list

After approval the narrated, subtitled video is made. In the example the video ran 1 minute 13 seconds. Below the video, the source of every sentence is listed; in the example video 16 of 16 on-screen lines were found in the document. The last scene was:

“Bu madde kapsamında verilecek eğitimin maliyeti çalışanlara yansıtılamaz.”

6331.pdf, s. 12, satır 1

In English (our rendering): “The cost of the training given under this article cannot be passed on to employees.” Source: 6331.pdf, p. 12, line 1.

What do I need to make a training video from a PDF?

Short answer: A PDF and a Zıpır account. Zıpır reads the document, drafts a narrated video plan (the scene plan) and does not make the video until you approve. The result is a narrated MP4: it carries the voice-over and a subtitle track, and every scene shows its source line. You do not need to record your voice, use editing software or write a script.

There are four steps: upload the document, approve the scene plan, get the video, and, if you want, edit a frame and share. Opening an account is free and asks for no card. To start, sign in at app.zipirapp.com or write to [email protected]. You can download the video and get the proof report separately.

Why is the scene plan approved before the video is made?

Short answer: Making the video is the step that takes minutes and is hard to undo; it is cheaper to see a mistake first than to look for it after publishing. Zıpır first prepares the scene plan, which shows the scenes and the source line of each, and does not make the video until you approve.

Approval is tied to the plan as it is at that moment: if you change a scene after approving, the old approval lapses and you approve again. So the published video never contains a sentence you have not seen. If a quote cannot be found in the source, its scene is not made, and a second model searches the source again for every sentence in the scene plan. If you do not like a scene you can ask for a revision, and a revision does not take minutes from your plan. In the example video, 6331.pdf (28 pages) became 8 scenes and a video of 1 minute 13 seconds. The number of scenes and the length of your document will come out differently, depending on what it says.

How do I fix a wrong frame in a finished video?

Short answer: Pause the finished video, circle the frame you want to fix and write what you want. The change applies only to that scene; the other scenes are not made again. Editing a frame and revisions do not take minutes from your plan; minutes are spent only when you make a new video.

After fixing, you can create a share link if you want. A link is closed until you create one, and you choose its validity as 7, 30 or 90 days. When the video is done a proof report is also made; it gives a table of which page and line each on-screen text comes from. The panel also has an assistant: asked something like “where is the source?”, it answers with citations, but it only suggests and never writes changes for you. You can download the video as an MP4; the file comes with the voice-over and a subtitle track together.

How much does a training video from a PDF cost?

Short answer: Studio is ₺990 a month for 15 finished video minutes and Team is ₺3.490 a month for 60 minutes, in Turkish lira with VAT included. Buying is open.

Subscriptions are billed in Turkish lira for now, with VAT included: Studio is ₺990 a month for 15 finished video minutes, Team is ₺3.490 a month for 60 minutes. Paying yearly, Studio is ₺9.504 and Team ₺33.504.

If your minutes are short, the video does not start; you can shorten it, split it into parts or buy an extra pack. Extra packs are sold in US dollars (10 minutes for $17, 30 minutes for $45, 90 minutes for $115), are valid for 24 months from purchase, are not sold on the free plan, and plan minutes are used first. Buying is open; the pricing page has everything in one place.

For a better result

  • One topic, one document. When a video has a single source, both the scene plan and the source list stay simple.
  • Upload the current version. Because the video is made from the document, when the document is updated you only make the video again.
  • Take sensitive data out. Models are not trained on your documents; still, it is a good habit to remove personal data that does not need to appear in the video before you upload.

You can read why citing matters in how to make a video lesson that cites its sources, and the points specific to legal texts in training video from legislation.

Frequently asked questions

How do I make a lesson video from a PDF?

You take the text out with its page and line, put the scenes in a scene plan and approve it, and get a narrated, subtitled video. Your part is to upload the PDF and approve the scene plan.

Are scanned PDFs supported?

The code has text recognition for scanned PDFs, but it has not yet been verified end to end in the live product.

What do I need to make a training video from a PDF?

A PDF and a Zıpır account. You upload the document, approve the scene plan and get a narrated, subtitled MP4; no voice recording, editing software or script writing is needed.

Why do I approve the scene plan before the video is made?

Making the video is the step that takes minutes and is hard to undo; you see problems first instead of looking for them after publishing. Approval is tied to the plan as it is at that moment, and no video is made until you approve.

How much does a training video from a PDF cost?

Studio is ₺990 a month for 15 finished video minutes and Team is ₺3.490 a month for 60 minutes, in Turkish lira with VAT included. Buying is open.

How do you show where a sentence in the PDF is?

Zıpır extracts the text line by line and ties every on-screen line to its place in the document.

Try it with your own document

Send a document and we will make its first video together.

Get started