Can it remove hardcoded subtitles burned into a video?

Yes — that's exactly what it's built for. When subtitles are part of the pixels and there's no caption track to switch off, the AI locates the text region in each frame, erases the lettering, and rebuilds the moving background, then re-encodes the clip so the captions are gone for good.

How is this different from the image text remover?

The image tools clean a single still photo. This processes a whole video — it tracks the caption region across every frame, removes the text consistently as the scene moves, and re-encodes the result with the original audio and timing preserved. For a single screenshot, use the Remove Text from Image tool instead.

What video formats and sizes are supported?

MP4, MOV, WebM, and most common containers, up to 100 MB and ten minutes per clip. The upload goes straight to secure storage from your browser, so large phone recordings transfer without hitting a server size limit.

How much does it cost?

It runs on credits and is charged by the length of the clip — the estimated cost is shown before you confirm, so there are no surprises. New accounts start with free welcome credits, and you're only charged when a job succeeds; failed jobs are refunded automatically.

Do I need to sign in?

Yes. Because video processing is credit-based, you sign in with Google so the credits are tied to your account. Sign-in takes a couple of seconds and your first credits are already waiting.

Can it remove text from anywhere in the frame, not just the bottom?

Yes. Choose “all on-screen text” and it targets burned-in text wherever it appears — top, center, or corner bugs and labels. If your captions are only ever along the bottom, the bottom-strip region keeps everything else perfectly untouched.

Will the audio and length stay the same?

Yes. Only the picture is processed; the original soundtrack, length, and timing are carried through to the output unchanged. You get the same video back, minus the on-screen text.

How long does it take?

It depends on the length and resolution of the clip. Short clips finish in well under a minute; several-minute videos can take a few minutes. The page shows live progress and the finished video appears in place when it's ready.

Is the quality of the rest of the video affected?

The model focuses its changes on the text region and rebuilds only what the captions covered. The video is re-encoded once on the way out; choose best-quality mode to keep that re-encode as clean as possible, or smaller-file mode when you need a lighter download.

How long is my result kept?

The cleaned video stays available for seven days, then it's deleted. Download it once it's ready if you want to keep a copy — there's a Download button right under the result.

Video tool · AI subtitle removal

Remove Subtitles from Video

Erase hardcoded subtitles and on-screen text from a clip with AI — no frame-by-frame masking. Upload a video, choose what to clear, and download a clean version with the audio untouched.

In short

To remove subtitles from a video: sign in, upload your MP4, choose “subtitles only” or “all on-screen text,” and tap Remove subtitles. The AI tracks the text region across every frame, paints it out, rebuilds the moving background, and re-encodes the clip with the original audio intact. It works on hardcoded captions, burned-in labels, and bugs that no caption track can switch off. Charged by clip length, refunded if a job fails. Note: it erases text, it does not translate it.

Why hardcoded subtitles are so hard to remove

There are two kinds of subtitles. Soft subtitles live in a separate track — a sidecar SRT file or a stream you can toggle in the player — and disappear with a click. Hardcoded subtitles are different: they have been rendered permanently into the video’s pixels. Every frame carries the lettering as part of the picture, so there is no track to switch off and no menu that makes them go away.

That is how most re-shared and downloaded clips arrive. Short-form videos, screen recordings, ripped scenes, and exported social posts routinely bake the captions in so they survive re-compression and show up on every platform. Once the text is part of the frame, the only way to remove it cleanly is to rebuild whatever was behind it — across the whole clip, frame after frame, while the scene moves.

Doing that by hand means masking the caption in every single frame and painting in plausible background — hours of rotoscoping for a clip that lasts seconds. This tool does it automatically: it finds the text region, tracks it as the shot changes, erases the glyphs, and reconstructs the moving background underneath, then re-encodes the result so the captions are gone and the rest of the video looks untouched.

How to remove subtitles in 3 steps

1
Upload your video
Drag an MP4, MOV, or WebM into the upload card, or tap to browse. Clips up to 100 MB and ten minutes work directly in the browser — the file uploads straight to secure storage, so even a long phone recording goes up without choking on a server limit. You stay on the page while it transfers; a progress bar shows exactly how far along the upload is.
2
Choose what to erase
Pick “Subtitles only” to target the caption track the way a viewer reads it, or “All on-screen text” to also clear signs, logos, and burned-in labels anywhere in the frame. If your subtitles always sit along the bottom, switch the region to “Bottom strip” so the model leaves the rest of the picture completely untouched. Choose best-quality or smaller-file output depending on where the clip is headed.
3
Download the clean clip
Tap Remove subtitles. The AI tracks the caption region across every frame, paints over the lettering, and rebuilds the moving background underneath, then re-encodes the whole video. Longer clips take a few minutes; when it finishes, the cleaned video plays right in the page and a Download button saves the MP4. The original audio and timing are preserved.

What people use it for

Hardcoded subtitles you can't toggle off

Downloaded clips, screen recordings, and re-shared social videos often have subtitles burned permanently into the pixels — there is no caption track to disable. This erases them so you can re-subtitle, translate, or repost cleanly.

Repurposing footage across languages

Strip the original-language captions off a video before adding your own translated subtitles, so two sets of text don't stack on top of each other in the same lower-third.

On-screen text, logos, and bugs

Switch to “all on-screen text” to clear watermarked usernames, channel bugs, sponsor lower-thirds, and burned-in labels — useful for clean B-roll and reaction or commentary edits of material you have the right to use.

Clean masters for editing

Editors who receive a render with captions already baked in can recover a text-free master to recut, re-time, or relayout without the old text fighting the new edit.

Formats, limits & output

Input formats	MP4, MOV, WebM and most common video containers
Max file size	100 MB per upload
Max length	Up to 10 minutes per clip
Audio	Original soundtrack and timing are kept intact
Output	Re-encoded MP4 — best-quality or smaller-file mode
Retention	Result stays available for 7 days — download to keep it

Tips for cleaner results

Use the bottom strip when you can

If captions only ever appear along the bottom of the frame, choosing the bottom-strip region tells the AI to ignore the rest of the picture. That means faster, cleaner results and zero risk of it touching faces or detail higher up.

Quality mode for re-uploads, size mode for messaging

Pick best-quality output when the clip will be re-edited or posted to a platform that re-compresses anyway. Pick smaller-file when you need to send it through chat or email without hitting an attachment cap.

Plain or moving backgrounds rebuild best

Captions over sky, a wall, a blurred bokeh background, or smooth gradients vanish almost perfectly. Text sitting over fine, fast-moving detail is harder, and may leave faint residue on a few frames.

Trim before you upload

Charges and processing time scale with the length of the clip, so cut the video down to just the part you need before uploading rather than processing minutes you'll discard.

What it can and can’t do

It erases text — it doesn't translate it

This rebuilds the video underneath the captions; it never reads or converts the words. If you need a translation, remove the old subtitles here, then add a fresh translated track in your editor.

Hidden content is reconstructed, not recovered

Where a caption covered part of a face or object, the AI predicts a plausible continuation from the surrounding frames. Over a plain or moving background that's seamless; over intricate detail it's an educated guess.

Only use it on footage you have rights to

Removing captions, credits, or watermark text from someone else's video can carry its own legal weight beyond the underlying copyright. Clean your own content, or material you're licensed to edit.

FAQ

Questions about removing subtitles from video

Related tools

Remove Text from Video

The sibling tool for on-screen text anywhere in the frame — channel bugs, watermarks, lower-thirds, and labels.

Remove Text from Image

The single-image version — erase captions, date stamps, and writing from one photo or a video screenshot.

Remove Watermark from Photo

Remove Sticker from Image

Clear emoji, chat bubbles, and overlay graphics from screenshots and social exports.