Subtitle stitcher

A subtitle stitcher that finds the caption band for you

Marking the caption band on every frame is the tedious half of stitching subtitles. This one compares the frames first, works out which strip actually changes, and keeps only that strip.

What a subtitle stitcher does

A subtitle stitcher joins several screenshots of the same film, episode or clip into one tall image. The first frame stays whole so the picture still reads on its own. Every frame after it keeps only the subtitle strip, so the result is a column of dialogue rather than a stack of near-identical stills.

The output is a plain PNG. It drops straight into a post, a group chat or a review, with nothing to crop out first.

Where the work actually goes

Joining images is easy. Deciding what to keep is the hard part, because the caption band sits in a different place on every film. The manual method answers that with a fixed default slice and a handle you drag. On a run of ten screenshots that is ten corrections, and the tenth is the one that gets missed.

How this one finds the band

Every screenshot is downscaled and converted to greyscale. The tool compares rows across all the frames and scores how much each row changes. The contiguous run of rows with the highest score is the band. Because the score comes from your own frames, no fixed percentage is involved — a scene with two-line subtitles needs no different setup from one with a single line.

You can still nudge the band afterwards. The difference is that you are correcting a measurement rather than supplying one.

From a video file instead of screenshots

QuoteStitch also takes the video itself. Drop in an MP4, MOV or WEBM and it samples up to 320 points across the clip, then keeps one frame per line of dialogue. That removes the slowest part of the job — pausing the player and screenshotting by hand.

It does this without reading the text. It looks for the rows that carry dense horizontal edges, because burned-in subtitles are sharp near-vertical strokes. Camera motion does not confuse it: two different lines share a brightness envelope but not a glyph pattern, so the shape of the strokes decides where one line ends and the next begins.

• Video inMP4, MOV, WEBM — anything the browser can play
• Images inPNG, JPG, WEBP
• Long image outPNG at 1x or 2x, at original width or 720 / 1080 / 1440 px
• SamplingUp to 320 points across the clip, so seek time stays bounded
• LimitNo fixed image count. Practical limits come from device memory.
• NeedsSubtitles burned into the picture, not a separate .srt file

A subtitle stitcher is not a subtitle file tool

Search for a subtitle stitcher and you will also get tools that merge, convert or resync .srt and .vtt files. Those work on subtitle files. This one works on pictures and on video frames. If your subtitles live in a separate file, a subtitle file tool is what you want, and this page will not help.

Frequently asked