Screen Recorder

Record a tab, a window or your whole screen with optional microphone audio, straight from the browser. The file is written locally and never uploaded.

🌐 Español

🔒 Private by design: everything is recorded and processed locally in your browser and never uploaded to any server.

Written and maintained by Last updated

No install, and no upload either

Screen recorders are one of the most reliably annoying categories of software. The free ones stamp a watermark across your video or cap you at five minutes. The good ones want a subscription. Several of the web based ones upload the recording to their servers as you record, which for a screen recording is an unusually intimate thing to hand over.

None of that is necessary any more. Browsers have a built-in capture interface and a built-in media recorder, and between them they do the entire job. This page wires the two together, adds a couple of options, and gets out of the way.

The consequence is worth stating plainly: no watermark, no length cap from any service, no account, and the recording never leaves your machine. Your screen might contain a client’s dashboard, a private message or a partly filled form, and none of that is something to upload for the sake of a convenience.

Recording a demonstration

  1. Tick Include microphone audio if you want to narrate. Leave it off for a silent capture.
  2. Set Recording quality to Smooth (30 fps, larger file) or Smaller file (15 fps).
  3. Click Screen Recorder. The page tells you it is waiting for permission.
  4. Your browser opens its own sharing dialog. Choose a tab, a window or a whole screen, and grant permission.
  5. Recording begins and a live preview appears with a recording indicator.
  6. Click Stop Recording when you are done. The button changes while the file is being finished.
  7. Play the result back in place, then use the download link, which is named for the moment you finished.
  8. Record another clears everything for the next take.

The sharing dialog belongs to the browser, not to this page. That is a security boundary rather than a limitation: no web page is allowed to decide what it captures, which is exactly the guarantee you want.

How audio is put together

Audio here is more involved than it looks, because a screen recording can have up to two sources and they arrive separately.

Alongside the video, the browser is asked for audio. Whether any arrives depends on your browser, your operating system and what you picked in the dialog. Sharing a tab in a Chromium browser will commonly offer that tab’s audio. Sharing an entire screen frequently offers none, and that is normal rather than a failure.

If you also ticked the microphone option, a second and completely separate stream is requested. When both exist, they are mixed into one audio track through the browser’s audio graph rather than being attached to the recording as two tracks. That distinction matters: a file with two parallel audio tracks plays back inconsistently across players and editors, and many will simply use the first and ignore the other.

When only one source exists, it is used directly with no mixing. When neither does, the recording is silent, which is the correct outcome rather than an error.

A denied microphone permission does not abort anything. The recording carries on with whatever audio it did get, on the reasoning that losing the whole take over an optional extra would be worse than a missing narration track.

Frame rate, and what the number really means

The two options set a hint for the browser rather than a hard target. Ask for thirty and a busy machine capturing a large screen may deliver fewer; ask for fifteen and you will get at most fifteen.

Thirty is right for anything moving. Scrolling, video playback, an animated interface, a game: at fifteen these look visibly stuttery.

Fifteen is right for a slide presentation, a code walkthrough, or any demonstration where the screen mostly sits still and changes in discrete steps. It roughly halves the file size and nobody watching will notice.

The container is WebM, using the best codec combination the browser supports, chosen from an ordered list that falls back gracefully rather than failing when a preferred codec is missing.

Afterwards

WebM plays in every browser and in most modern players, but it is still the format that occasionally gets refused by an upload form or an older editor. WebM to MP4 converts it.

For editing the take, Trim Video removes the fumbling at the start and end, Crop Video tightens the frame onto the part that matters, and Merge Videos joins segments if you recorded in pieces to keep memory use sane. Mute Video strips audio you would rather replace.

For narration without the screen, Voice Recorder captures the microphone alone. Teleprompter scrolls a script for you while you record, and Annotate Screenshot covers the case where a still image would have done the job. The rest is on the video tools hub.

Frequently asked questions

Where does the recording go while I am recording?

Into memory in your own tab. The browser's recording interface hands back chunks of encoded video as it goes, they are collected locally, and the finished file is assembled when you stop. Nothing is streamed anywhere, which is why there is no watermark, no time limit imposed by a service, and no account.

What gets recorded, exactly?

Whatever you choose in the browser's own sharing dialog when the recording starts. That dialog is part of the browser rather than this page, and it offers a single tab, an application window, or an entire screen. This page never sees your screen contents; it only receives the stream the browser gives it after you have picked.

Will it capture system sound as well as my voice?

It asks for audio alongside the video, and whether you get any depends on the browser, the operating system and what you selected in the sharing dialog. Sharing a tab in a Chromium browser commonly offers tab audio; sharing a whole screen often does not. If system audio comes back and you also ticked the microphone option, the two are mixed into a single track rather than attached as two separate ones, which is what keeps the result playable everywhere.

What if I deny the microphone permission?

The recording continues without it rather than failing. That is a deliberate choice, since the microphone is an optional extra and losing a whole recording because of a permission prompt would be a poor trade. You end up with screen audio only, or with no audio at all if there was none to capture.

What format is the file?

WebM, with the best codec combination your browser supports, tried in order from a modern video codec with audio down to a plain container as a last resort. The file is named for the moment the recording finished, so a folder of recordings sorts chronologically without any renaming.

Which frame rate should I choose?

Thirty for anything with motion, video playback, scrolling or an animated interface. Fifteen for a slide presentation or a static demonstration, where it roughly halves the file size and nobody can tell. The value is a hint to the browser rather than a guarantee, so a busy machine may deliver fewer frames than requested at either setting.

How long can I record?

There is no time limit in the tool, but there is a practical one, because the whole recording is held in memory until you stop. A long session at a high frame rate on a large screen will eventually become a problem. For anything over a few minutes, record in segments and join them afterwards.

Can I record a video that is playing?

Technically the browser decides. Some content is protected and comes through as a black rectangle, which is the protection working rather than a fault here. Recording material you do not have the right to copy is your responsibility either way.

Related tools