About Clapr

What Clapr is

Clapr is a set of video and audio tools that run inside your browser tab. You pick a file, the work happens on your own machine, and you get the result back. There is no upload step, no queue, no account, and no watermark.

That is the whole product. It is deliberately not a video editor, not a cloud service, and not a subscription.

Why it exists

Most "online video converter" sites work the same way: you upload your file to someone's server, wait in a queue behind everyone else, download the result, and trust that the copy on their disk eventually gets deleted. For a 2 GB wedding video, a client's unreleased footage, or a screen recording with your inbox visible in the corner, that trade is worse than it looks.

Browsers can now do this work locally. FFmpeg — the tool that underpins most video processing anywhere — compiles to WebAssembly and runs at usable speed in a modern browser. Clapr exists because once that is true, uploading is no longer the price of admission.

How it actually works

1. You select a file. The browser reads it from disk into memory. No network request carries your file.
2. Clapr runs FFmpeg, compiled to WebAssembly, inside your tab. For multi-threaded encoding it uses SharedArrayBuffer, which is why tool pages are served with Cross-Origin-Opener-Policy: same-origin and Cross-Origin-Embedder-Policy: credentialless.
3. The result is written back to memory and handed to you as a download.

The FFmpeg WebAssembly core is self-hosted from this domain rather than pulled from a CDN, so using Clapr does not hand a third party a log of what you process.

What this approach cannot do

Being honest about the limits matters more than the pitch:

• Your device is the bottleneck. A long 4K file on an old laptop will be slow, because the encode is running on your CPU instead of a server farm. A cloud service with a GPU will beat it on raw speed.
• Very large files can exhaust browser memory. Browsers cap how much memory a tab may use, and that ceiling is lower than your actual RAM.
• Browser support is not uniform. The multi-threaded path needs cross-origin isolation and SharedArrayBuffer. Where those are unavailable, Clapr falls back to a slower single-threaded path, and some tools may not run at all.
• Anything requiring a server is a separate thing. Transcription, for example, needs a speech model that cannot run locally at reasonable quality. Where Clapr uses a server, the tool page says so plainly — and the audio is extracted locally first so only a small audio file is ever sent, never the full video.

Who builds it

Clapr is built and maintained by a single developer. You can reach me at hello@clapr.app — bug reports and "this file broke it" emails are genuinely welcome, since browser-based processing has more edge cases than a server pipeline does.

Contact

Email: hello@clapr.app

For what data Clapr does and does not collect, see the Privacy Policy. For terms of use, see the Terms of Service. To browse everything Clapr can do, see all tools.