Free tool
Subtitle Timing Calculator
Divide a clip's duration by its subtitle line count to see the average time on screen per line, and find out before you burn the captions in.
AhaNow tools · Updated 2026-11-03 · No sign-up, nothing leaves the page
Summary
This calculator divides a clip's duration by the number of subtitle lines and reports the average time each line can stay on screen. Averages hide the interesting part — real subtitle timing follows the character count of each line — but the average is the fastest way to see that a clip simply cannot hold the number of lines someone planned for it.
Check whether the lines fit the cut
Count cues — every time the on-screen text changes is one. One sentence can be two cues.
The part the subtitles cover, not the whole video.
Round up if you are estimating — under-counting hides the problem.
Average time on screen per line
3.00 s
A comfortable average can still hide one unreadable line. Assign time in proportion to the character count of each line and open the cue list before trusting this number.
Average time per line, not a per-line plan. Use it to find out whether the clip can hold the lines at all; the actual split follows each line's character count.
What the calculator works out
Duration divided by line count. If a ninety-second clip has thirty subtitle lines, each line gets three seconds on average, and the page also reports how many lines the clip could hold at the usual one-second floor. The second number is the useful one when a client sends a script and a finished picture: it says immediately whether the plan and the duration agree.
Both numbers are averages, and averages are honest about that. A clip where every line is three seconds and a clip where half the lines get one second and half get five produce the same average, and only one of them is watchable. The calculation is a first check, not a timing pass.
How to fill in the two fields
- Enter the duration of the segment the subtitles belong to, in seconds. If the subtitles cover a five-minute chapter inside a longer video, use the five minutes, not the full runtime.
- Count cues, not sentences and not words. Each time the on-screen text changes is one cue, even if the change is a single word being corrected.
- Leave out cues that are not dialogue timing — a title card or a lower third is not competing with the spoken words for reading time.
- If the count is an estimate, round up. Under-counting lines makes the average look comfortable when it is not, and the whole point of the check is to catch that case.
Why equal timing does not work
Reading speed is roughly constant in characters per second, so the time a line needs is proportional to its length. Splitting a duration equally across lines gives short lines too much time and long lines too little, which reads as subtitles that lag behind the audio in some places and race ahead in others — even though every cue is technically in sync with its start time.
The practical procedure is to assign time by character count and then clamp: no cue below the floor, no cue so long that viewers read it twice and wait. Where the clamps are needed, the fix is at the other end — split the long line, or merge two short ones — not at the timing stage.
When the average misleads
A comfortable average can hide unreadable individual lines, and a flagged average can come from one very long line while the rest are fine. Always open the actual cue list before rewriting anything.
- Cue length varies. The average says nothing about the longest line, and the longest line is always the one that fails.
- Character density varies by language. The same spoken sentence occupies different amounts of screen in different writing systems, so a per-line average computed for one language does not transfer to its translation.
- Silence is not free. Gaps between cues, music-only sections and pauses all consume duration without consuming lines, which makes the arithmetic per line look better than the viewer's experience.
- Two-line cues are not twice as readable. Most style guides cap a cue at two lines because reading speed drops when the eye has to jump; a two-line cue needs more than double the time of a one-line cue with the same total characters.
- Burned-in captions cannot be adjusted afterwards. A soft track can be re-timed by the viewer's player; a burned-in one cannot, so the arithmetic has to be right before the render.
What to do with the result
If the average is comfortably above the floor and no single line is unusually long, the plan fits and the work is ordinary timing. If the average is below the floor, the line count has to come down: fewer cues means either fewer words or a longer clip, and those are the only two levers. Rewriting the same lines slightly shorter rarely buys enough time to change the answer.
Questions
Should I count sentences or subtitle lines?
Subtitle lines, also called cues. One sentence can be two or three cues when it is long, and a short exchange can be one cue with two speakers. The number that matters here is how many times the on-screen text changes, because that is what the duration has to be divided among.
What is a good duration per subtitle line?
There is no single number. It depends on how many characters are in the line and how fast the audience reads, and both change with the language and the audience. This page tells you when the average falls below the usual floor; deciding how much time a specific line needs is a per-line judgement, not a page-level one.
Why is one second the floor?
It is a long-standing convention rather than a measurement of your audience. Below roughly a second, a full line of text appears and disappears faster than most viewers can read it, so the practical fix is fewer characters per line or fewer lines — not a faster fade. Treat the flag as a prompt to look at the actual lines.
How should I split the total duration across lines of different lengths?
In proportion to their character counts, not equally. If one line has fourteen characters and the next has seven, the first needs twice the time. Equal per-line timing is what makes subtitles feel out of sync even when every cue starts exactly where the audio does.
Does this work for burned-in captions?
The arithmetic is the same, but the stakes are higher. Burned-in text cannot be turned off or restyled, so a line that is too fast is a line the viewer permanently cannot read, and a caption placed over a face cannot be moved. Keep burned-in lines shorter than you would for a soft subtitle track.
Do captions for deaf and hard-of-hearing audiences need more time?
Usually yes. Those tracks also carry sound descriptions such as [door closes] or music cues, which are read on top of the dialogue rather than replacing silence. If you are producing one of those tracks, budget time for the descriptions as well as the words.
