How to Replace Subtitles for Translation: Remove the Old Band, Then Add the New Track

Replacing subtitles for a new language is a two-part job: getting the old captions off the picture, and putting the new ones on. ClearPix handles only the first part — it rebuilds the pixels behind burned-in text locally in your browser. The new subtitle track is added in any video editor or captioning tool. Knowing where each tool stops prevents the most common workflow mistake.

Pick the cheapest source first

Before processing any pixels, rule out the two paths that preserve the original background perfectly: a removable soft-subtitle track (most players and editors can disable it) and a textless master from the original editing project. AI inpainting is the fallback for a video that only exists with captions burned into the frame. For the full decision table, see the hardcoded subtitle removal guide.

Steps

1. Play the video and check its subtitle or CC menu. If the captions toggle off, export without that track and skip pixel removal entirely.
2. Ask for the original project or a textless export if the video is not yours to edit. A clean master beats any reconstruction.
3. Open the subtitle remover and load an MP4, MOV, WebM or MKV file you own or may edit. Drag the band over the old caption area and keep it tight — everything inside the mask is treated as pixels to replace, including faces and scene details.
4. Process the clip and review more than the first frame: moving backgrounds, skin, detailed textures and scene changes are where reconstruction artifacts appear. If a trace remains, tighten the mask over that area and re-run — no need to start over.
5. Download the cleaned video as your intermediate master, then create the new language track in your editor. ClearPix does not translate audio or generate subtitle files; it prepares the picture so the new captions sit on a clean background.

Quality boundaries to plan around

The rebuilt background is a plausible estimate, not recovered footage. Keep the cleaned master at the original resolution — if the clip also needs more pixels, run the video upscaler after removal, knowing that upscaling will not hide inpainting artifacts. Frames wider than 1280 px are processed proportionally smaller, so check legibility of any remaining scene text before publishing.

The video stays on your device throughout; only the model and runtime files are downloaded and cached. See what browser-based AI processing uploads for the exact distinction.