How subtitle matching works
Instead of pairing captions by their sequence numbers, this tool finds the nearest unused cue in the second track based on start time. This supports tracks with slightly different numbering.
A default tolerance of 700 milliseconds pairs captions whose start times are reasonably close. Unmatched cues from either file remain in the output.
This is not automatic speech synchronization: if two subtitles have substantially different timing, adjust the offset first or revise them manually.
Common questions
Why do some lines stay separate?
Their start times are too far apart. Try a larger matching tolerance, or shift one file first.
Is it safe to use on already translated files?
Yes. It only combines existing cues; no translation service is used.
Can this match every subtitle perfectly?
No. A timestamp heuristic needs manual review when subtitle segmentation differs significantly.