Merge Subtitle Cues
Or paste subtitle text
Options
Two cues with a gap of this many milliseconds or less become one cue running from the first one's start to the last one's end. The test is inclusive, so a gap of exactly this many milliseconds merges. 1000 ms (one second) is a good starting point for a file whose lines were split mid-sentence; a small value like 40 ms closes only the flicker between cues that were meant to be one.
A space runs the two texts together on one line, which is what you want when the original file split a sentence across two cues. A line break stacks them, so the cue comes out as a two-line subtitle that stays on screen for the whole span.
Which frame rate the timings are counted in. On the format pages this is what MicroDVD needs, because it stores timing as frame numbers. On the frame rate pages it is the rate the file was timed against — the version the subtitles were made for — and the page's second box takes the rate of the video you are going to use them with. 23.976, 24, 25, 29.97, 30, 50, 59.94 and 60 are the common ones. Getting it wrong shifts the whole timeline, and on the frame rate pages it shifts it by a ratio rather than by a fixed amount.
This page joins neighbouring cues into one when the gap between them is small enough. You set the limit — 1000 ms to start with — and any two cues separated by that much or less become a single cue. It is the tool for a file that flickers: a sentence split across two cues, a word-per-cue export from a transcription service, or a set of lines whose gaps are so short that the subtitle appears to blink on and off.
The combined cue runs from the first cue's start to the last cue's end, so the pause between them is filled with the merged text. That is what removes the flicker, and it is also the one thing to be sure you want: a gap that the original subtitler left on purpose — a beat between two speakers, a pause before a punchline — is now covered by text. Nothing else in the file moves. No cue outside the merged group shifts by a millisecond, and the cue numbers stay exactly as they were, so cue 3 can follow cue 1 in the output.
The last decision is how the two texts meet. On one line runs them together with a space, which is what you want for a sentence that was split: the reader gets one continuous line for the whole span. On separate lines stacks them, which is what you want when the two cues were meant to be read as two lines — a short exchange, a caption over a translation. A file that needs a custom separator between the two — a dash, a slash — is a bilingual merge rather than a timing fix, and lives in the text group.
The limit stops at 60 seconds on purpose: past that the tool is not merging cues any more, it is collapsing the timeline, and a file with one cue in it is not a subtitle file. As with every page here, nothing is uploaded — the file is read, merged and written again inside this browser tab, and SRT, WebVTT, ASS, SSA, SBV, SAMI, MicroDVD, TTML, JSON and CSV are all read the same way.
Examples
One sentence, split across two cues
Input
1
00:00:01,000 --> 00:00:02,500
She walked into the room
2
00:00:02,600 --> 00:00:04,000
and closed the door.
Output
1
00:00:01,000 --> 00:00:04,000
She walked into the room and closed the door.
A 100 ms gap, well inside the one-second limit, and the sentence is whole again. The fade-out and fade-in that a player performs between two cues is gone: the line stays up for the full three seconds instead of disappearing for a tenth of one. The cue keeps the number 1 because it is the cue that survived; nothing renumbers the file after a merge.
The same two cues, kept on separate lines
Input
1
00:00:01,000 --> 00:00:02,500
She walked into the room
2
00:00:02,600 --> 00:00:04,000
and closed the door.
Output
1
00:00:01,000 --> 00:00:04,000
She walked into the room
and closed the door.
Identical timings, different output: choosing separate lines writes the second cue's text as a second line of the same cue rather than running it on. Useful when the two halves were written to be read as a pair, and useful for bilingual files where the two cues were already one line each. A cue can hold as many lines as the format allows, and this is a hard line break in the file, not a wrapped line.
A gap the tool is right to leave alone
Input
1
00:00:01,000 --> 00:00:03,000
What time do you call this?
2
00:00:09,000 --> 00:00:11,000
Sorry. The bus never came.
Output
1
00:00:01,000 --> 00:00:03,000
What time do you call this?
2
00:00:09,000 --> 00:00:11,000
Sorry. The bus never came.
Six seconds of silence between two speakers, and the output is the input byte for byte. Filling that pause would put one person's line on screen while the other is talking, so a threshold anywhere near this gap is a mistake — the order of the two lines is not the same as the timing of them. Setting the limit to 500 ms, as here, leaves this file alone.
Frequently asked questions
The file splits every sentence across two or three cues. What should I set the gap to?
Start with 1000 ms. A file exported from a transcription service usually puts 50 to 300 ms between the pieces of one sentence, and a second is comfortably wider than that while staying well inside the pauses between sentences. Turn to Separate lines for how the two texts meet, run the whole season through the queue in one go, and then read the timeline preview: it is the fastest way to see whether any pause you wanted to keep has been swallowed.
Will merging cues change when my subtitles appear on screen?
For the cues that merge, yes — and that is the point. The combined cue starts when the first one did and ends when the last one did, so the gap between them is now covered by text instead of being empty. Every other cue in the file keeps its exact times. If a pause inside the merged span was deliberate, raise the limit down until the pair no longer joins, or leave those cues out by fixing them in a separate run.
Do the cue numbers get rewritten after a merge?
No. The surviving cue keeps the number it had and the absorbed cues' numbers disappear, so a file that had cues 1, 2 and 3 can come out with 1 and 3. That is what the format's own numbering is — a label, not a sequence — and leaving it alone means the numbers in your output still match the ones in the version you started from, which matters when you are comparing the two.
Should I merge cues that are already touching?
Only if you meant to. A gap of zero is what a well-timed file looks like: every cue ends exactly where the next begins. Merging across zero joins those pairs and is the same as setting the limit to 0 here — which is why the page that merges overlaps only, next to this one, leaves touching cues alone by design rather than merging them by accident.
Can merging lose any of my text?
No. Every word of every cue ends up in the output; what is lost is the individual timing of each absorbed cue, which is recorded in the loss report. Styling is the one exception to watch for on a styled file such as ASS: a cue can only carry one style, so when two cues in different styles are merged the first one's style is kept and the report counts what could not come along.
Can I merge a whole file into a single cue?
Not by accident. The box stops at 60000 ms, so the widest setting joins cues up to a minute apart — enough for a badly split song or a rushed interview, not enough to collapse an episode into one cue. If a file really is one subtitle's worth of text spread over a hundred cues, delete the ones you do not want first, or check whether what you have is a transcript rather than a subtitle file.