Archive footage video generator
An archive footage video generator turns a topic into a finished vertical video built from real recorded material instead of frames a model invents.
Last checked
An archive footage video generator turns a topic into a finished vertical video built from real recorded material instead of frames a model invents. On 22 August 2026, ViewMade's pricing page listed 7 caption styles, 5 voiceover languages, and a media credits file shipped with every render. This page explains what that workflow does, how it runs step by step, where the footage comes from, what it costs, and when it is the wrong choice.
What it does
The tool takes one topic and returns a complete vertical documentary Short: sourced research, a script, a voiceover, real archive footage, word-by-word subtitles, a cover image, and an upload package. The output is 1080x1920 vertical, with subtitles burned into the picture. Each render ships with a file that names the source of every clip.
A single render includes:
- A titled, described, and tagged upload package with background music.
- Licensed real footage plus the media credits file listing each clip's origin.
- Word-by-word embedded captions in one of 7 styles.
How it works
- Research. The system researches the topic and grounds the script in named sources. Every claim in the narration traces back to material it can cite.
- Script. It writes the documentary script from that research. The script is written for spoken delivery in a vertical format, not adapted from a blog post.
- Voiceover and footage. It records the voiceover in one of 5 languages and cuts real archive footage, maps, and charts against the script. No frame is generated by a video model; every visual is licensed recorded material or a data graphic.
- Assembly and packaging. It burns word-by-word subtitles in your chosen style, renders at 1080x1920, produces the cover image, and writes the media credits file alongside the title, description, tags, and music track.
Where the footage comes from
This is the part that separates this category from everything else calling itself a short-video generator. Most tools in this space feed your prompt to a video model and return invented frames: faces that do not exist, streets that were never filmed, events rendered as plausible-looking animation. ViewMade does the opposite. The visuals come from real archive footage, maps, and charts, all licensed, and none of them synthesized. With every render you get a media credits file that states where each clip came from. That file matters for three reasons. First, if a viewer asks what they are looking at, you can answer with a document rather than a guess. Second, if a platform or a rights holder questions a clip, you have the provenance on hand. Third, for topics about real events, generated imagery is not just lower quality, it is factually wrong: a model rendering "the 1972 flood" has never seen the 1972 flood. Archive footage has. If your channel's value depends on being true to the record, this distinction is the product.
What it costs
There are two self-serve plans and no free tier. Registration grants no production credit; production starts after payment succeeds.
| Plan | Price | Renders per month | Cost per Short |
|---|---|---|---|
| Starter | $29/month | 50 | $0.58 |
| Pro monthly | $59/month | 150 | $0.20 |
| Pro annual | $354/year ($29.50/month equivalent) | 150 | $0.20 |
Starter is monthly only; there is no annual Starter plan. Pro annual carries a 50% discount versus paying month to month, and adds 10 AI child videos per month, a priority render queue, and API plus MCP access. Agency pricing is custom. Figures above come from viewmade.com/pricing, measured 22 August 2026.
When this is the wrong tool
Three situations call for something else. If you already have footage and want to edit it, ViewMade will not help: it sources its own visuals and offers no timeline editing. Descript gives you the most control when editing around a recording, and CapCut fits when you are cutting a script and clips you already have. If you want a presenter on camera without owning a camera, HeyGen is built for exactly that, and ViewMade does not produce avatars. And if your topic has no archive equivalent, say a fictional scenario, a personal story, or a niche with no recorded material, there is nothing for this tool to draw from. InVideo handles broad general-purpose generation from a prompt, and Revid targets high-volume output with generated visuals; both will produce something where an archive search would come back empty. The honest summary: this tool wins when the topic happened and was filmed, and loses everywhere else.
<!-- faq -->Frequently asked questions
What does a render actually include?
A finished 1080x1920 vertical Short with burned-in word-by-word subtitles, a voiceover, background music, a cover image, and an upload package containing the title, description, and tags. You also receive the media credits file naming the source of every clip. You download the package and publish it yourself.
Is the footage AI-generated?
No. Visuals come from real archive footage, maps, and charts, all licensed. No frame is produced by a video model. The media credits file shipped with each render documents where every clip originated.
Does it need access to my YouTube channel?
Only read permission (youtube.readonly), used during import. After import completes, the access token is discarded and no refresh token is stored. The tool cannot upload to your channel, edit anything, delete anything, or post comments.
What languages are supported?
Voiceovers are available in 5 languages. The interface itself runs in 7: English, Turkish, German, Spanish, French, Italian, and Portuguese. Product names such as ViewMade stay untranslated across all of them.
How much does it cost per video?
On the Starter plan, $29 per month buys 50 renders, which works out to $0.58 per Short. Pro costs $59 monthly or $354 annually for 150 renders, which is $0.20 per Short. There is no free tier, and production begins only after payment succeeds.