Merge Subtitles

Merge Subtitles appends files in the queue from first to last. The first file keeps its original timestamps. For every later file, the applied offset is the previous merged latest cue end plus the chosen shared gap. That offset is added to every cue in that file.

Because the original timestamps inside each later file are retained before the offset is added, a local lead-in before its first cue remains intact. The tool does not reset each part to zero, sort cues globally, interleave simultaneous tracks, deduplicate repeated files, or guess content relationships.

Drag-and-drop reordering is supplemented by Move first, Earlier, Later, and Move last controls for keyboard and touch use. Duplicate files are accepted intentionally. Changing order or gap recalculates offsets from the original parsed files rather than stacking a previous result.

The live analysis reports each source encoding, cue count, local earliest and latest time, timeline from zero, overlaps, structural repairs, applied offset, and projected timing. Boundary cards show how the next offset is formed, while cue previews stay bounded even for large files.

Each SRT is decoded independently. UTF-8 and BOM-marked Unicode are detected automatically. If one file is not valid UTF-8, only that row asks for an explicit supported legacy encoding; other valid queue items remain ready and the merge waits until every row is resolved.

The tool accepts up to 20 files, 10 MiB per file, 50 MiB combined, 100,000 cues combined, and 40 million decoded character units combined. Files are prepared sequentially in a worker, and additions that would exceed aggregate limits are refused without clearing the existing queue.

Accepted cue text, multiline structure, Arabic and other Unicode text, RTL marks, emoji, punctuation, speaker labels, supported literal formatting text, cue durations, local gaps, overlaps, and source cue order are preserved. Only later-file timestamps change, while cue numbers and line endings use consistent output structure.

The download uses UTF-8 without a byte-order mark, standard HH:MM:SS,mmm timestamps, CRLF line endings, sequential cue numbers, one blank line between cues, and a final newline. The complete output is reparsed and compared with the exact projected cue sequence before download.

This version is intended for separate episodes, chapters, discs, or exported subtitle parts that should play one after another. It does not inspect video, calculate media durations, align dialogue automatically, merge bilingual tracks side by side, or resolve overlapping subtitle versions.

Source warnings and overlaps are reported rather than silently repaired. Use Subtitle Time Shifter when every cue needs one fixed delay or advance, Subtitle FPS Converter when synchronization drift grows proportionally, and Subtitle Editor for cue-by-cue changes.