Productivity and Utilities

Caption Line Breaker

Paste a transcript and choose a line length to get timed caption blocks.

Your numbers

An example limit. From 10 to 120.

1, 2 or 3.

An example speed used to set how long each caption stays on screen.

More options

Result

Caption blocks

4

14.6 seconds of captions at 17 characters a second.

Longest line
40 characters
Total time
14.6 seconds
Words
47
Words longer than a line
None

Timed captions (Copy result adds the SRT text)

#StartEndCaption
100:00:00,00000:00:04,529Welcome back to the show. Today we are / looking at how captions are timed, why
200:00:04,52900:00:08,000line breaks matter, and what / makes a subtitle easy to read.
300:00:08,00000:00:12,706A good caption keeps a phrase together, / stays on screen long enough to read, and
400:00:12,70600:00:14,647never covers more than two lines.
How this was calculated
  1. Words are added to a caption until 2 lines of 42 characters are full. If the caption is nearly full, it ends at the last sentence end instead.
  2. A two line caption is split where the two lines come out closest in length, and a split after a comma, colon or sentence end is preferred.
  3. Each caption stays on screen for its characters divided by 17, and never less than 1 seconds. Captions follow one another with no gaps.
  4. Copy result gives the number of blocks and then the captions in SRT format: a number, the start and end times, the lines, then a blank line.
  • The times come from reading speed only. They are not matched to the audio, so use them as a starting point and adjust them against the recording. The text stays in this browser.

Calculated using SumPanda.com · https://sumpanda.com/caption-line-breaker/

Results are estimates for general information. Verify important figures before relying on them.

Next step

How to use the caption line breaker

  1. Paste the transcript. Line breaks in it do not matter, because the words are re-flowed.
  2. Set the characters allowed in a line, the lines allowed in a caption, your reading speed in characters a second and the shortest time a caption may stay on screen.
  3. Read the table of timed captions, with the start and end of each. A two-line caption is split so that the lines are close in length, with a pause such as a comma preferred for the break.
  4. Use Copy result to get the number of blocks followed by the captions as SRT text, or download the CSV.

Formula

a caption takes words until the lines are full: at most N lines of M characters

duration = max(shortest time, characters in the caption / reading speed)

start of the next caption = end of the previous one

two-line split chosen to minimize |length of line 1 - length of line 2|, with a bonus when line 1 ends with , ; : . ! or ?

If a caption is nearly full and a sentence ends in its last 40 percent, the caption ends there. A word longer than a line gets a line of its own and is reported.

A 47 word transcript

The sample transcript has 47 words. With 42 characters a line, 2 lines a caption and 17 characters a second, it becomes four captions. The first reads "Welcome back to the show. Today we are" over "looking at how captions are timed, why" and runs from 0 to 4.529 seconds.

The second ends at the sentence end after "easy to read." at 8 seconds, the third ends at 12.706 seconds and the fourth, "never covers more than two lines.", ends at 14.647 seconds. No line is longer than 40 characters, and each caption starts the instant the previous one ends.

Why line length and balance matter

Short lines of about the same length are easier to read than one long line and one short one, because the eyes make the same trip on each. Breaking after a comma or a full stop keeps a phrase together. The page weighs both: it looks at every way of splitting the caption into two lines that fit, and picks the one with the most even lengths, giving credit when the first line ends at a pause.

The limits of 42 characters, 2 lines and 17 characters a second are examples you can change. The right values depend on where the captions will be shown and on the viewers, so follow the rules of your platform if it has any.

Timing and the SRT format

Reading speed sets how long a caption stays: the longer the text, the longer the time, with a floor so short captions do not flash. The times here come from reading speed alone, so they will not line up with speech. Use them as a rough first pass and then adjust them against the audio in a subtitle editor.

An SRT entry has a number, a time range such as 00:00:00,000 --> 00:00:04,529, the lines of text and a blank line. To convert existing subtitle files, use the SRT Subtitle Converter. To count the words first, use the Word Counter, and to check how easy the text is to read, use the Readability Score Calculator.

Assumptions and limits

  • The transcript is plain text. Speaker names and sound cues are treated as ordinary words.
  • Lines are 10 to 120 characters, captions have 1 to 3 lines and the transcript is limited to 200,000 characters.
  • Times are set only by reading speed and the shortest caption time, starting at the time you enter.
  • The first caption starts at 0 unless you change the start time in the advanced options.
  • The text stays in this browser.

Formulas reviewed October 10, 2026. See the calculation methodology for how SumPanda rounds, tests and sources its formulas.

Frequently asked questions

What happens to a word longer than the line?

It is placed on a line of its own and the page lists it in the statistics, because it cannot fit within the limit. Raise the limit or split the word to fix it.

Why do some captions end early?

When a sentence ends in the last part of a nearly full caption, the page ends the caption there instead of starting a sentence it cannot finish. This keeps sentences together.

Are these times accurate for my video?

No. They are based on reading speed only, not on when the words are spoken. Use them as a starting point and adjust them against the audio.

Can I get a file?

Use Copy result to copy the SRT text and save it in a file with the .srt ending, or download the CSV for a table of the same captions.